HomeMy WebLinkAboutFeasibility report Vol 1 Engineering and economic aspects 1982I
I
I
I
I
I
I
I •
I
I
I
I
I
I
I
I
I
I
I,
ALASKA POWER AUTHORITY
SUSITNA HYDROELECTRIC PROJECT
FEASIBILITY REPORT
VOLUME 1 -ENGINEERING AND ECONOMIC ASPECTS
FIRST DRAFT
FEBRUARY,~ 1982
~-< -.. --• '·,1 ~ ' '
I
I
I
I
I
;I
I
I
I
I
I
I
I
••
I
I.
I
I
fl r
SUSITNA HYDROELECTRIC PROJECT
FEASIBILITY<REPORT
PRELIMINARY OUTLINE
VOLUME 1 -ENGINEERING AND ECONOMIC ASPECTS
9 -SELECTION OF WATANA GENERAL ARRANGEMENT
9.1 -Site Topography
9.2 -Site Geology -0 c-·c---
9.3 -Geotechnical Design Considerations
9.4·-Seismic Conside~~tions
9.5 -Selection of Reservoir Levels
9.6 -Selection of Installeci Capacity
9.7-Selection of Spillway Capacity
9.8 -Main 2am Alternatives
9.9 -Diversion Scheme Alternatives
9.10 -Spillway facilities Alternatives
9.11-Power Facilities Alternatives
9.12 -Selection of Watana General Arrangement
9.13 -Preliminary Review
9.14-Intermediate Review
9.15-Final Review
10 -SELECTION OF DEVIL CANYON GENERAL ARRANGEMENT
10.1 -Site Topography
10.2 -Site Geology
10.3 -Geotechnical Considerations
10.4 -Seismic Considerations
10.5 -Selection of Reservoir Level
10.6 -Selection of Installed Capzc-ity
10.7 -Se1ection of Spillway Capacity
10.8 -Main Dam Alternatives
10.9 -Diversion Scheme Alternatives
10.10-Spillway Facilities Alternatives
10.11 -Power Facilities Alternatives
10.12-General Arrangement~election
10.13 -Preliminary Review
10.14 -Final Review
11 -SELECTION OF MAIN ACCESS PLANS
11~1 -Background
11.2-Objectives
11.3 -Appt"oach
11.4-Corridor Selection and Evaluation
11.5 -Route Selection an.d Evaluation
11.6-Description of Basic Plans
11.7 -Additional Plans
Page
9-1
9-'1
9-1
9-9
9-12
9-13
9-16
9-19
9-20
9-22
9-27
9-28
9-31
9-33
9-38
9-41
10-1
10-1
1 0.-l
10-7
10-10
10-11
l0-12
10-13
10-13
1:0•16
10-18
10-19
10-21
10-21
10-26
11-l
11-1
11-2
11-2
il-3
11-4
11-7
11-8
r• .,,
••
I
I
-1
I
I
I
I
I
I
I
I
I
I
I
I
I
I
~\··,
I
VOLUME 1 -TECHNICAL AND ECONOMIC ASPECTS (Continued)
J2 -
13 -
11.8-Evaluation Criteria
11.9-Evaluation of Access Plans
11.10-Identification of Conflicts
11.11 -Comparison of Access Plans
11.12-Recommended Access Plan
11.13 ... Mitigation Recommendations
11.14-Tradeoffs Made in the Selection Process
WATANA DEVELOPMENT
12.1-General Arrangement
12.2 -Site Access
12.3 -Site Facilities
12.4 -Diversion
12.5 -Emergency Release Facilities
12.6 -Main Dam
1'2.7 -Relict Channel Treatment
12.8-Primary Outlet Facilities
12.9 -Main Spillway
12.10-Emergency Spillway
12.11 -Intake
12.12 -Penstocks
12.13 -Powerhouse
12 .. 14 -Reservoir
12.15-Tailrace
12.16 -.Turbines and Generators
..
12.17 -Miscellaneous Mechanical Equipment
12 .. 18-Accessory Electrical Equipment
12.19-Switchyard Structures and Equipment
12.20 -Project lands
DEVIL CANYON DEVELOPMENT
13.1 -General Arrangement
13.2 -Site Access
13.3 -Site Facilities
13.4 -Diversion
13.5 -Arch Dam
13.6 -Saddle Dam
13.7 -Primary Outlet Facilities
13.8 -Main Spillway
13.9 -Emergency Spillway
13.10 -Devil Canyon Power Facilities
13.11 -Penstocks
13.12-Powerhouse
13.13 -Reservoir
13.14-Tailrace
13.15 -Turbines and Generators
Page
11-9
11-15
11-21
11-22
11-25
11-26
11-27
12-1
12-1
12-2
12-3
12~8
12-10
12-11
12-31.
12-37
12-41
12-43
12-44
12-49
12-50
12-54
12-55
12-55
12-62
12-70
12-84
12.,.84
13-1
13-1
13-2
13-3
13-6
13-8
13-10
13-14
13-16
13-18
13-19
. 13-20
13-21
13-24
13-25
13-26
,,,
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
··I
I
VOLUME 1 TECHNICP1L AND ECONOMIC ASPECTS (Continued)
13 .. 16 -Miscellaneous Mechanical Equipment
13.17-Accessory Electrical Equipment
13.18-Switchyard Structured and Equipment
13.19 -Project Lands
14 -TRANSMISSION FACILITIES
14.1 -Electric Systems Studies
14.2 -Corridor S~le~tion
. 14.3 -Route Selection
14 .. 4 -Towers, Foundations and Conductor·s
14.5 -Substations
14.6 -Dispatch Center and Communications
15 PROJECT OPERATION
15.1 -Plant and System Operation Requirements
15.2 -General Power Plant and System Railbelt Criteria
15.3 -Economic ·operation of Units
15.4 ... Unit Operation Security Criteria
15.5 w Dispatch Control Centers
15.6 -Susitna Project Operation
. 15.7 -Performance Monitoring
15.8 -Plant Operation and Maintenance
16 -ESTIMATES OF COST
16.1'-Construction Costs
16~2 -Mitigation Costs
. 16'. 3 -Operation, ~ia i ntenance and Rep 1 acement Costs
16.4 -Engineering and Administration Costs
16.5 -Allowance for Funds Used During Construction
16.6 -Escala.tion
16.7 -Cash Flow Requirements
17 -DEVELOPMENT SCHEDULE
17.1 -·watana Development
17.2 -Devil Canyon Development
:7.3-System Development Schedule
18 -ECONOMIC AND FINANCIAL EVALUATION
. 18.1 -Economic Evaluation
18.2 -Risk Analysis
18.3.-Marketing
18',,4 ~ Financial Evaluation
18.5 -Financial Ri~k
·'
>'
0
Page
13-28
13-30
13-35
13-35
14-1
14-1
14-7
14-14
14-19
14-25
14-25
15-1
15-1
15-2
15-4
15-6
15-7
15-8
15-15
15-17
16-l
16-1
16-5
16-6
16-6
16-8
16-8
16-8
To fol1o\'l
18-1
18.-l
18-13
18-25
18-27
18-31
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
••
I
I
l·t
Plate
1
2
3
4
5
6
7
8
9
10
11
12
13 -
14
15.
List of Plates
Title
Railbelt Area
Devil Canyon
Hydro Development
Fill Dam
Watana
Hydro Development
Fill Dam
Watana
Stages Fi 11 Dam
High Devil Canyon
l-lydro Development
S-usi tna I I I
Hydro Development
Vee
. Hydro Development
Denali & Maclaren
Hydro Developments
Preferred Tunnel
Scheme 3
Plan View
Preferred Tunnel
Scheme 3
Sections
Watana
Arch Dam Alternative
Watana
Alternative Dam Axes
Watana
Preliminary Schemes
Watana
Scheme ~JPl
Plan
Watana
Scheme WP3
Sect tons
"
I List.of Plates (cont'd} -.
I P1ate Title -.
16 Wa:tana
I Schemes WP2 & WP3
Plan and Section
17 Watana I Scheme WP2
Sections
•• 18 Watana
Scheme WP4
Plan
I 19 Watana
Scheme WP4
Sections I 20 Watana
Scheme WP3A
I 21 Watana
Scheme WP4A
I 22 Watana
Simulated Reservoir Operation
I 22A Devil Canyon
Simulated Reservoir Operation
I 23 Devil Canyon
Scheme DCl
I 24 Devil Canyon
Scheme DC2
1-
25 Devil Canyon
Scheme DC3
26 Devil Canyon
I Scheme DC4
27 Devil Canyon
I Selected Scheme
28 Alternative Access Corridors
I 29 Alternative Access Routes
'
30 Access .. Plan
I I Recommended Route ~ I :J
·'
••
~
I
·, .... . . . -----------
I
I
I
I
I
I
I
I
I
I
I
I
.I
I
I
I·
I
I
,I
I.
Plate -
31
32
32A
33
34
35
36
37
38
39
40
41
42
·43
List of ..Plates (cant' ~l
Title
Watana
Reservoir
Plan
Watana
Site Layout
Watana
General Arrangment
Layout of Structures
Plan
Watana
· Hydrological Data
Sheet 1
Watana
Hydrological Data
Sheet 2
Watana
General Layout
Site Facilities
Watana
Village and Townsite
Watana
Main Construction Camp Site
Watana and Devil Canyon
Construction Camp Details
Watana
Diversion
General Arrangement
Watana
Diversion Scheme
Sections
Watana
Diversion
Intake Structures
Watana
Downstream Portals
Plan and Section
Watana
Emergency Release
Sections
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Plate
44
45
46
47
48
49
50
51
52
53
54
55
List of Plates !cont'd)
Title
Watana
Main Dam
Plan
vJatana
Main Dam
Sections
Watana
Main Dam
Grouting and Drainage
Watana
Outlet Facilities
Gate Structure
Watana
Outlet Facilities
General Arrangement
Watana
Main Spillway
General Arrangement
Plan and Profile
Watana
~1a in Spillway
Control Structure
Watana
Main Spillway
Chute Sections
Watana
Main Spillway
Flip Bucket Discharge Structure
Watana·
Emergency Spillway
Watana
Power Facilities
General Arrangement
Watana
Power Facilities
Plan and Sections
0
...
I
I
I
I
I
••
I
I
I
I
I
I
I
I
I
I
I
I
I
• ~.~ ..... ~-•'<''---' <' h·'"''-"•" .... _. ·~·"' "' j'> '<"N"
Plate
55 A
56
57
58
59
60
60A
61
62
63
64
64A
65
66
List of Plates (cont 1d}
Title
'\ t wa ana
Power Facilities
Plan, Sections and Elevations
Watana
Power Facilities
Access
Watana
Powerhouse
Plans
· Watana
Powerhouse
Sections
Watana
Transformer Gallery
Plan anG Sections
Electrical Legend
Watana
Powerhouse
Single Line Diagram
\~a tan a Swi tchya rd
Single Line Diagram
Block Schematic
Computer-Aided Coutrol System
Oevil Canyon
Reservoir
f'lan
Devil Canyon
Site Layout
Devil Canyon
General Arrangement
Layollt of Structures
[levi 1 Canyon
Hydrologic Data
Sheet 1
Devii Canyon
Hydrologic Data
Sheet 2
-'. >t ~ ~ ., ·~. . ·-· ....
I
I
••
I
I
I
I
I
I
I
I
I
I
I
I
I
••
I
I
Pl. ate
66A
67
68
69
J
70
71
72
73
74
75
76
76A
77
List of Plates (cont'd)
Title
Devil Canyon
Genera 1 Layout
Site Facil itie~~
Devil Canyon
Temporary V<:llage
Devil Canyon
Construction Camp
Plan
Devil Canyon
Diversion
General Arrangements
Devil Canyon
Di ve\"S ion
Sections
Devil Canyon
Dams
Plan and Profile
Devil Canyon
Main Dam
Geometry
Devil Canyon
Main Dam
Geometry
Crown Section
Devil Canyon
Main Dam
Thrust Blocks
Devil Canyon
Main Dam
Grouting and Drainage
Devil Canyon
Saddle Dam
General Arrangement
Sections
Devil Canyon
Outlet Facilities
Devil Canyon
Main Spillway
General Arrangement
Plan and Profile
I List of Plates (cant • d)
I Plate Title
78 Devil Canyon
•• Main Spillway
Control Structure
79 Devil Canyon
I Main Spillway
Chute
I 80* Devil Canyon
Main SpiYlway
Flip Bucket
I 81 Devil Canyon
Emergency Spillway
I
General Arrangement
82 Devil Canyon
Emergenc~ Spillway
I Sections
83 Devil Canyon
I Power Intake Structures
Plan and Sections
84 Devil Canyon I Power Facilities
85 Devil Canyon
I Powerhouse
Plans
I 86 Devil Canyon
Powerhouse
Plan and Sections
I 87 Devil Canyon
Powerhouse
Sections
I 87A Devil Cdnyon
Trans forme•" Ga 11 ery
I General Arrangement
Plan and Sections
I
88 Devil Canyon Powerhouse
Single Line Diagram
89 Qevil Canyon Switchyard
I Single Line Diagram
I
*Not Included
f
I 0
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I'
I
Plate
90
91
,•
List of Plates (cont'd)
Title
Watana
Construction Schedule
Devil Canyon
Construction Schedule
•
I
I
I
I
I
I
I
I
I
I
I
I
I
I
.I
I
I
I
I
9 -SELECTION OF WATANA GENERA~ ARRANGEMENT
This section describes -the ~volut·ion of the general arrangement of the Watana
project, which, together with the Devil Canyon project, comprises the develop-
ment plan selected as part of Section 8. This section also describes the site
topography, geology, and seismicity of the Watana site relative to the design
and arrangement of the various site facilities. The process by which reservoit
operatiq~} levels and the installed generating capacity of the pm>~er facilities
were estaDlished is also presented, together with the means of handling floods
expected dJring construction and subsequent project operation.
The main components of the Wat ana deve 1 opment are as follows:
-Main dam;
Diversion facilities;
-Spillway facilities;
-Outlet facilities;
-Emergency release facilities; and
-Power facilities.
A number of alternatives are available for each of these components and they can
obviously be combined in a number of ways. The following paragraphs describe
the various components and methodology for the preliminary, intermediate, and
fi na 1 screening and revi e'fl of a 1 ternati ve genera 1 arrangement of the components,
together with a brief description of the se 1 ected scheme. A ~detailed descri p-
tion of the various project components is given in Section 12.
9.1 -Site Topography
The project site is located in a broad U-shaped valley at river mile 183~
approximately 2-1/2 miles upstream of the confluence of Tsusena Creek with the
Susitna River'. The river at the site is relatively wide, although turbulent
(Figure 7.11). On the right bank, the valley rises at an approximate slope .of
2H:lV from river level at elevation 1450 for approximately 600 feet, then
gradually flattens to a maximum elevation of 2350 between the Susitna River and
Tsusena Creek. The left bank rises more steeply from the river for about 450
feet at a slope of 1.4H:1V, then flattens to 3H:lV or less to approximate
elevation 2600.
9.2 -Site Geology
General
This section s·ummarizes the geological and the geotechnical investigations con-
ducted to date and the geologic conditions present at the Watana ~i~e. A
detailed description of the geology and site investigations is presented in the
1_980-1981 Geotechnical Report (1).
9-1
(a) §eologic Setting
The Watana site is located on the western side of a Tertiary age (2 to 70
m.y.b.p) intrusive body. The rock is primarily a gr~y to green medium
grained crystalline igneous rock of diorite-quartz diorite composition.
Associated-with the pluton are andesitic dikes and volcanic flows, which
are generally the fine grained equivalents of the intrusive rocks, and
volcaniclastic·sedimentary rocks.
The underlying rock is hard, fresh, and of good quality. It is slightly
weathered at the surf ace and a 1 ong joint surfaces~ The overburden is
generally thin on the valley walls, thickening to the north of the damsite,
and consists of glacially derived silts~ sands, and gravels~ Shear, frac-
ture, and alteration zones have been delineated highlighting two major
structural features to be considered in the design of the dam. No evidence
of major structural deformation or faulting has been found. Permafrost
conditions exist on the north facing slopes with temperatures near the
freezing point. '}
(b) Geo loai ca 1 and Geotechni ca 1 Investigations
Surface and subsurface investigations for the site have been conducted by
several organizations at different times. Preliminary reconnaissance work
was done by the USSR in the 1950s. During the years of 1975 and 1978 the
COE conducted site specific investigations to determine the suitability of
the site. As part of the. current study program, more detailed
investigations were undertaken in 1980 and 1981 to establish the technical
feasibilitoy of the project. The investigations have included air
. reconnaissance, air photo interpretation, geologic mapping of rock and
surficial materia 1, dri 11 i ng of both rock and overburden, including 1 n-ho le
geophysical tests and seismic refraction surveys. Both insitu and
laboratory tests have been performed to determine the engineering ~
characteristic of soils and rocks. The location of drill holes and other
investigations is shown in Figures 9.1 and 9.2.
Geo 1ogi c mapping was concentrated in the immediate proposed damsite area
between Tsusena and Deadman Creeks. All accessible areas were mapped for
the rock and overburden exposures. The lithology or type of material~ bed-
ding, jointing, weathering, degree of consolidation, exposure size, and
elevation were noted and plotted on maps for use in the interpretations.
Seismic refraction surveys were performed throughout the investigations on
both banks, the river channe 1, the borrow areas, and the upper slopes of
t~e damsi te area to determine the bedrock depth and other si gni fie ant
features. A total of feet of seismic refraction traverse were run at
the Watana site. Information gathered was integrated and used with the
other subsurface information for correlation and development of a thre.e
dimensional representation of site charactel·istics.
0
9-.2
I
•••••
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-----~--~-'---~~~--~-----~~-~---
I
~· 0 .
I
I
I
I
I (c)
I
I
I.
I
I
I
I
I
I
I
I
I
Di amend core and rotary dri 111 ng were performed in the foundations and
abutments of the proposed project structures. A total of . feet was
drilled in 28 holes with feet of core recovered. In-haTe permeability
tests were conducted in each hole upon completion and geophysical logging
and borehole photographs were performed in selected borings.
To monitor the ground water and ground temperature conditions at the site,
piezometers and thermistor strings were installed in selected drill holes
both in 1978 and 1980. A regular monitoring program for those instruments
has been established to collect data.
A series of tests were performed on the rock recovered from coring to
determine the engineering char acteri st i cs of the rock mass. The resu 1 ts of
these investigations were used to develop the geologic picture of the dam-
site and the adjacent areas .. Figures 9.3 and 9.4 present the rock outcrop
map and the stl"uctural geology map of the site. The results of the labora-
tory rock tests are summarized in Table 9.1.
Construction Material Investigations
Extensive investigations have been conducted both prior to and during the
current studies to identify sufficient quantities of suitable materials for
the construction of an embankment dam and for concrete aggregates. The
investigation methods have included geologic mapping, auger and rotary
drilling, excavation of test pits, seismic refraction surveys, and
laboratory tests. A total of rotary holes, auger holes, test
pits, and linear feet of srumic refraction surveys have been-
performed.---:rhe locations of the various potential sources of embankment
material and concrete aggregates together with the locations of boreholes
and test pits are shown in Figure 9.5.
( i) Rock F111 Material
Two sources for the rockfill material, designated Quarry A and
Quarry B, have been identified. The rock in Quarry A is located on
the south bank and primarily contains a relatively thin layer of
andesite (an extensive igneous rock) overlying diorite. The diorite
is generally hard, durable and fresh, and suitable for use as
rockfill in the dam. Quarry B is an alternative source of rockfill,
however, it is estimated that sufficient quantities of rock are
available in Quarry A to meet the project requirements.
( i i) Impervious Core Material
Two sources have been identified for the impervious core material
and are dt~si gnated as Borrow Area D and Borrow Area H. Borrow Area
Dis located within 1.5 miles upstream of the damsite on the north
bank. The upper few feet of materia 1 comprises tundra, topsoi 1 ~ and
boulderss and is underlain by glacial tills composed of dense
gravelly silty sands with some clay. The tills range from 15 to 25
feet in thickness and overlie sandy gravelly clay. A composite
grain size curve for these soils is presented in Figure 9.6. The
I
mat$rl a 1 is well graded with natural water content at about the
plastic limit. Figure 9.7 presents the results of Atterberg limits : •.
on the finer portion of the material.. More than adequate quantities
of materi aJ are present in Borrow Area D for the impervious fi 11
requirements at Watana. .J
An alt.ernative source of core material, designated Area H, is locat-
ed approximately 5 to 7 miles downstream from the damsite on the
south bank of the river in the Fog Lakes a~ea. This area contains a
relatively thick layer of till composed of silt, sand, and gravel
with some cobbles. A composite grain size curve for the till
materia 1 is prese!1ted in Figure 9.8 and the Atterberg 1 i mit test
results in Figure 9. 9. The material is quite similar to that in
Borrow Area D except that the natural water content is higher.
Sufficient quantities of impervious fill material are also available
at this location if required.
(iii) Fi 1 ter Materia 1
Borrow Area E has been identified as a primary source of material
for filter and transition zones of the embankment. This area is
located at the confluence of the Tsusena Creek and the Susitna River
approximately 2.5 miles downstream from the damsite. The area is
covered by about 2 feet of organics and silt and is underlain by a
few feet thick layer of silty sand to clean sand. Below that a
.thick layer of sandy and gravelly material exists. A composite
grain-size di stri but ion curve for Borrow Area E materia 1 is
presented in Figure 9-10. Sufficient quantities are available in
this borrow area to meet the project requirements for wilter
materi a 1 s. The material wi 1] have to be processed to meet the
gradation requirements of specific zones. Additional alternate
sources of material i dent i fi ed inc 1 ude Borrow Area ,C and Borrow Area
F, at greater distances from the damsite. Also, some material from
Borr0\'1 Area D and riverbed alluvium (as described later) is suitable
with processing to be used as filter and/or transition materia 1.
{iv) Gravels and Gobbles for Shells
Seismic refraction survey investigations were performed within
miles upstream and miles downstream from the damsite in the-
Susitna River valley:-These investigations confirmed that suffi-
cient quantities of granual material are available for use in the
supporting she 11 zones of the dam. In the upstream she 11 of the
embankment, to meet design requirements, the material will require
processing to remove fines and \tast age of materia 1 3/8 inch or
sma 11 er in size, and ov~rsi ze materia 1 1 arger than 18 inches.
Available data indicates that the grain size distribution of these
materials will be similar to that from Borrow Area E, with probably
a higher percentage of coarser material.
9~4
I
I
I
I
I
I
.I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
(v) Concrete Aggregate
The material avai 1 able from Borrow Areas E, C, F, and the riverbed
all uvi urn is suitable for use as coarse and fine aggregate for con-
crete. Processing wi 11 be required to produce desired gradations ..
The coarser particles are rounded and petrographic analyses have
indicated the material to be of good quality. Sufficient quantities
are available within the identified sources.
(d) Geologic Conditions
A summary of overburden and bedrock conditions is presented in· the follow-
ing paragraphs:
(i) Overburden
Overbur·den thickness is generally thin on the 1alley walls and
thickens away from the proposed damsite to the north (Figure 9.3).
On the lower slopes, the overburden consists primarily of talus.
Above the break in slope where the topography becomes more gentle~
glacial silts, sands, gravels, and boulders are encountered. Sub-
surface investigations have indicated the contact between the
overburden and bedrock to be relatively unweathered.
The depth of the river alluvium beneath the proposed dam averages
about 80 feet 1 up to a maximum of about 100 feet, and consists of
sand, silt, coarse gravels, and boulders. Very little is known at
this time about the denseness and character of this alluvium. A
deep bedrock depression has been delineated on the north side of the
river and is discussed under Section 9ol3.
(ii) Bedrock Lithology
The Watana site is underlain by a ·serie of sedimentary, volcanic,
and plutonic rocks. The damsite is primarily underlain by an
intrusive dioritic body which varies in composition from granodio-
rite to quartzdiorite to diorite. The volcanic rocks (dior1te and
andesite) are generally finer grained equivalents of these intrusive
rocks. The sedimentary rocks consist of tuffaceous siltstones anti
graywackes. The geologic map of the·site is shown on Figure 9.3.
The quartz diorite is light gray and is found primarily upstream
. from the damsite axis. The texture is massive and the rock is hard,
competent, and fresh except within the shear zones~ which are dis-
cussed later. The diorite is a dark grayish green rock with massive
appearance. It is hard, competent, and generally-fresh. These two
grades of di.orite occur in alternating zones on the order of several
hundred feet wide. Weathering is limited to a very thin zone on the
exposed surface, and along the joints to feet _depth. These
rocks have been intruded by mafic and felsic dikes which are
generally only a few feet thick with tight contacts. These. dikes
generally trend parallel to major joint sets. The rock, downstream
from these diorites, is a series of extrusive rocks ranging from
rhyolite to andesite and basalt. Andesite porphyry is the more
prominent of these rocks. The rock is a medium to dark gray to
9-5
...... ' . . ~,..,'!
-
green and contains quartz diorite inclusions~ The contact of the
andesite with the underlying diorite is generally slightly weathered
and fractured. On the north bank, the andesite-diorite contact at
the downstr•eam is coincident wi-th the 11 Fingerbustern shear zone.
A sequence of sedimentary rocks composed of essent i a 11 y vc71 cani c
debris is exposed downstream from the damsite and is comprised of
generally snund sandstones and siltstones. This sequence is
overlain by andesite.
(iii) Bedrock Structures
-Joints
There are two major joint sets and two minor joint sets at the
site (Figures 9 .4 and 9 .12). These joint sets are described in
Table 9.2. Set I~ which is the most prominent set strikes 320°
(N40W) and dips 80° NE to vertical. This set is found throughout
the damsite and parallels the general structural trend in the
region. Joint Set I has a subset, which strikes 29o= to 300° (N60
-70W) with a dip of 75° NE.. This subset is localized in the
downstream area near where the diversion tunne 1 porta 1 s are
planned. This subset also parallels the shear zones in the down-
stream area of the site.. Set I I trends northeast to east and dips
vertically. This set is best developed in the upstream portion of
the damsi te area, although it is prominent in the downstream areas
with a more easterly strike. No other structural features were
found with orientations in this set. Sets III and IV are minor
sets but can be locally strong. Set IIi forms numerous open
joints on the c 1 iff faces near the 11 Fi ngerbuster", and sever a 1
shear zones parallel this orientation. Set IV appears to result
from stress relief from glacial unloading and/or valley erosion.
The average spacing, of Joint Sets I and I I is 1 to 2 feet and 6 to
12 inches, respectively;'····The spacing of Joint Sets III and IV is
quite variable and can range from a few inches to several feet.
-Shears and Fracture Zones
Several shears, fracture zones, and alteration zones are present
at the site. For the most part, these are small and discontinu-
ous. During mappinglt all zones greater than 10 feet in width were
mapped and delineated on the geologic may (Figure 9.4).
Shears are defined as having breccia~ gouge, and/ or s·l i ckens l ides
indicating relative movement and are found in two forms. The
first type is found only in the diorite and are characterized by
breccia of sheared rock that has been healed by a matrix or very
fine grained andesite/diorite. The contacts, although irregular,
are tight and unfractured. The zones were found to have high RQDs
and to be fresh and hard. The second type is common to all the
rock types and consists of breccia and/or gouge of fine grained
t"ock types and consists of breccia and/or gouge of fine grained
rock material in a silt/clay matrix. These are soft, friable, and
often have secondary mineralization of carbonate and chlorite
showing slickensides. These are genera11y less than 1 foot wide.
I
I
I
I
I
I
I
'&
I
I
••
I
I
I
I
I
I
I
I
'I
I
I
I
I
I
I
I'
I
··-
••
I
I
I
I
I
I
I
I
Fracture zones are a 1 so common to a 11 rock types ranging from a
6-inch to 30-foot wide (generally less than 10 feet) zone of
closely spaced joints that are often iron oxide stained or carbon-
ate coated. Where exposed, the zones tend to form topographic
lows.
In the a 1 terati on zones, the felspars and mafic minerals of the
rock have been chemically altered by hydrothermal solutions to
clay and chloriteQ The degree of alteration encountered is highly
variable across the site. These zones are rarely seen in outcrop
as they are easily eroded tnto gullies but were encountered to
some degree in all the boreholes. The transition between fr.esh
and altered rock is gradational and the zones may range to 20 feet
thick although are usually less than 5 feet. Core recoveries are
generally very good and the rock quality is dependent on the
degree of alteration.
(iv) Significant Structural Features
The Watana site has several significant geologic features consisting
of broad areas of the shears, fractures and alteration zones
described previously.
The two most prominent areas have been named 11 The Fins" and the
11 Fingerbuster". "The Fins 11 is located on the north bank of tht.~
river upstream from the diversion t unne 1 intake. It is
approximately a 400 foot wide area characterized by three major
northwest trending zones of shearing and alteration that have eroded
into steep gullies. These alteration zones are separated by intact
rock bands (ribs} 5 to 50 feet wide. The 20-foot wide upstream zone
of the series coincides with the diorite/andesite porphry contact.
The other two zones, approximately 55 and 30 feet wide, are fi 11 ed
with severely altered talus. This area is characterized by a 300 -
310° orientation (NSO to 60W) and near vertical dips of its
component shears and by low seismic refraction velocity of the rock.
The extension of the zone is extrapolated by topographic and seismic
lows northwest to a sheared/altered outcrop on Tsusena Creek.
The 11 Fi ngerbuster 11 is 1 oc a ted downstream from the d amsi te and is
exposed in a 40 foot wide deep talus filled gully along the andesite
porphyr-y/diorite contact. The rock is severely weathered with
closely spaced joints trending parallel to Set I (330°) and Set III
(0°). Slickensides indicate vertical displacement. The extension
of this zone to the south is ba~ed on a strong north-south topo-
graphic lineament, although no outcrops are exposed. Becapse of the
1 ack of exposure and defi nab 1 e continuity of this feature~ its
. location orientation is extrapolated.
A prominent alteration zone wa~ encountered in BH-12 on the south
bank. The·ho1e encountered aprroximately 200 feet of hydrothermally
altered rock. Although core recovery in thin boring was very good,
the quality of rock was relatively poor and a zone of significant
artesian pressure was encountered. The south eastern extension of
this zone passes under. the core of the dam.
9-7
(e) Ground Water Conditions
The gr1'lund water regime in the bedrock is confined to movement a 1 ong
fractur-es and joints. Measured water 1 eve ls have ranged from to
feet below surface. It is assumed that the ground water in the non-pe_r_m-a--
frost areas on the north side .of the river is a subdued rep 1 i ca of the
topography with the gradient towards the Susitna River and its tributaries.
Artesian condit7ons are present in isolated fr.acture isolation zones.
(f) _Permafrost Conditions
Permafrost conditions exist on th~ north facing slopes (left bank) of the
damsite area. Measurements in the borings indicate that it penetrates to a
depth of feet and show marginal temperatures within loC of freezing.
Only sporad1 c areas of permafrost have been encountered on the right bank~
{g) Reservoir Geology
The topography of the \~atana Reservoir and adjacent slopes is characterized
by a narrow V-shaped stream-cut valley superimposed on broad shapes
U-shaped glacial valley. Overburden masks much of the bedrock especially
in the 1 ower and uppermost r'1aches of the reservoir. Figure 7.11 shows the
general Watana reservoir geology. ·
The lower portions of the Watana reservoir are predominantly covered by a
veneer of glacial till with scattered outwash deposits. On the south side
of the Susitna River, the Fog Lakes area is characterized by a fluted
ground moraine surface. Upstream in the Watana Creek area, a broad flat
plain is mantled with glacial till and semi-consolidated Tertiary sedi-
ments. The·se are predominantly stratified~ poorly graded, fine sands and
silts with some clays.. The river valleys contain significant amounts of
alluvial deposits and reworked outwash. Ice disintegration features such
as Kanes and eskers have been observed in the river valley.
A non-conformable contact between argillite and the diorite pluton in the
ct3ffisite area was mapped approximately three miles upstream of the damsite.
Semi-consolidated, Tertiary age sedimentary rocks, and volcanics of
Trias sic ag'e are present just downstream from the confluence of \~at ana
Creek and the Susitna Rvier. These volcanics consist of metabasalt flows
with thin interbeds of metachert, argillite, marble, and metavolcaniclastic
rocks. Metamorphosed from Watana Creek to Jay Creek. Tife rocks between
·Jay Creek and Oshetna Creek are metamorphic amphibolite and minor amounts
of greenschist and foliated diorite.
The main structural feature of the Watana Reservoir is the Talkeetna Thrust
Fault which trends northeast -southwest~ The Talkeetna Thrust Fault
crosses the Susitna River approximately eight miles upstream from the
damsite,. This fault has been studied in detail as part of the seismic
studies, and has been determined to be inactive in relation to the project
design.
9-8
I
I
I
I
I
I
I
I
I
:I
I
I
I
I
I
I
I
I
I
I
1:
I
I
I
I
I
I
I
·-
.-.
I
I
I
I
I
'I
I
I
9~3 -Ge9technical Design Considerations
This section deals with the geotechnical aspects of design of the dam and other
major stt"uctures at the Watana site.
(a) Main Dam Excavat.i on and Foundation Treatment
As discussed previously, the riverbed alluvium ranges up to approximately
100 feet in depth. The character of the this material is difficult to
define~ however~ its stability during a strong earthquake event is
quect.ionable. Considering the nature of the materi,al, and the height of
the Jam, the riverbed material will be removed entirely within the limits
of the dam. th~ overburden materia 1 on the abutments is relatively thin,
except for gu 11 i es and pockets. Most of this materia 1 is-frozen and wi 11
become unstable when thawed and is therefore unsuitable for the dam
foundation. Accordingly, the overburden on the abutments will also be
required. Details of foundation treatment are preseflted in Section 12.
The presence of numerous shear zones, containing gouge material, indicates
the need to remove ali weathered rock under the impervious core and
upstream and downstream filter zones. Excavation will include shaping of
valley walls along the abutments to provide a proper contat surface in
accordance with good modern design practice. Excavation under the outer
shells will include removal of loose rock blocks, extensively weathered
rock and local reshaping as necessary. The strength of the rock foundation
is otherwise adequate to support the embankment and associated reservoir
1 oads.
The two major geologic structures at the site, "The Fins" on the upstream
side and the HFi ngerbuster" on the downstream side, have a major influence
on the overall project layout., however, they do not directly affect the darn
at its proposed location. ·
Extensive permafrost is present on the Jouth bank (as deep as several
hundred feet) and sporadic pet"'mafrost has been not~d on the north bank,
This permafrost is within loC of freezing and is protected in most part by
the thick ·tundra and heavy vegetation. During the foundation excavation,
the 1 ass of i nsu l at i ng tundra may cause th aw1 ng of overburden and may
result in unstable slopes and/or mud flows. Therefore, proper care and
planning will be required during removal of this material.
(b) Contra 1 of Under seep age and Up 1 i ft ,
A grout curtain wi 11 be provided under the core of the dam and extending
several hundred feet into the abutment beyond the dam. The initial phases
of the grout curtain wi 11 be exploratory in nature, in order to identify
areas in the dam foundation rock that require more extensive treatment and
provide detailed information of the foundation. A series of drain holes
wi 11 be drilled downstream from the grout curtain to form a continuous
drainage curtain. These holes wi 11 drain into underground grouting/drain-
age g a 11 eri es, which wi 11 permit moni taring of seepages and water pressures
and access for necessary remedial work if required at a later date. Si nee
9-9
-"': .. ~~
the rock mass is permafro.st ... affected, ground thawing will bEl required prior
to grouting. Consolidation grouting is all planned under the core and the
filters to provide a good contact surface free from open joints and
fractures.
(c) .Cofferdams and Dewatering
Becauso. of the pervious nature of the thick riverbed desposits and the
relatively high pool behind the upstream cofferdan, major dewatering opera-
tions wi 11 be required during the foundation excavation and unti 1 the dam
construction reaches above the diversion stage pool level. A slurry trench
cutoff is currently proposed beneath the upstream cofferdam to control
water flows during diversion. Further exploration is necessary in the
riverbed to better define the extent and condition of the a 11 uvi a 1 mate-
rials at the cofferdam site prior to construction.
(d) Undergro~nd Structures
The rock conditions at the Watana site are suitable for the construction of
tunnels and underground caverns. From the geological and geotechnical
viewpoint, the location and the orientation of these structures is influ-
enced by the orientation and location of ·rock discontinuities. Permafrost
conditi-ons will not have any major adverse impact except where thawing may
be required for grouting.
The RQD values indicate that 85 percent of the rock is of a good to exct=1-
lent category. The remaining 15 percent represents poor quality rock
associated with rock discontinuities. The major joint sets at the t~atana
site are oriented at N40W (Set I) and N45E (Set II). Other four joint sets
are minor. The major shear and fracture zones a 1 so parallel these general
trends. The most favorable orientations for the tunnels and the large
underground caverns are those with their long axes perpendicular to the
major joint sets. These factors have been a major factor in selection of
the alignments of the tunnels and major caverns to achieve maximum stabi 1-
ity and minimum support requirement.
Although little is known at this time about the insitu stress regime at the
site, the general tectonic stress regime within the region is in a compres-
sion mode.. The unconfined compressive strength of the rock ranges from
to and suggests that overstressing problems such as spalling or
slabing are not likely. Conventional rock bolt support using 1 inch
diameter' bo 1 t s is genera 11 y considered adequate in most areas with spans
less than 40 feet. For larger spans and in areas of poor quality rock, the
support requirements have been determined on a case-by-case basis. In the
case of large span openings, intersection of n~arby vertical and subhori-
zontal joints can create unstable blocks in the crown. Allowances have
· been made for the use of support measures such as shotcrete, welded wire
fabric, and concrete lining in areas of potentially poor rock quality and
water carrying tunnels under high head (such as penstocks).
9-10
I
I
I
I
1
a·
I
'I
~ I
:I
I
I
'I
I
I
I
I
I
-.• ~
I
I
I
I
I
I
I
I
I
I
I'
I
I
I
I
I
I
I
I
"
(e)
Although the rock mass by itself is fairly impervious, intersection of rock
di scontinui ties may cause ground water problems during construction and act
as path of seepage and high pore pressures during operation. Provisions
have, therefore, been made for consolidation/ring grouting and suitably
placed drain holes to reduce the risk of a build up of high pore
pressures.
Tunnel excavation can be performed using conventional drill and blast
techniques or higD production mechanical excavations. Sufficient informa-
tion is not available at this time to make this decision, and for feasibil~
ity assessment purposes, conventional drill and blast methods have been
assumed.. The excavation of powerhouse caverns wi 11 be performed by dri 11
and blast using a primary heading, side slash and bench excavations
approach.
The spacing between long tunnels has been set at 2 .. 5 times the diameter of
the 1 argest tunne 1. The spacing between the major caverns has been set
such that a pillar thickness of 1.5 times the span of the larger cavern is
mai nt ai ned.
Stabi lity_of Soi 1 and Rock Slopes
In most areas the excavation slopeS? wi 11 be in the rock. The slopes in the
overburden, where necessary, have been based on the nature of soi 1, ground
water table, and the height of the slope. In general, slopes in overburden
will not be steeper than 2H:lV below the water table and 1.5:1V above the
water tab 1 e. A bench of adequate width wi 11 be provided at the over~burden
rock contact to accommodate any local slumping or slope failure and to
intercept and d 1 spose of ground/seepage \vater. Fl atter s 1 opes may be
required where frozen ground may be become unstable because of high pore
pressures oduring thawing.
The slopes in rock are controlled by the joint dips and orientations.
Since major joint set dips are almost vertical, lH:lOV slopes are con-
sidered reasonable up to 40 feet in height.. Where the height of slope
exceeds 40 feet, a minimum of 10 root wide bench have been provided every
40 feet depth to facilitate construction and to provide access for future
maintenance .. These berms will also intercept falling loose rock pieces and
surface/ground water drainage. Locally, rock bolting or similar support
techniques and drain holes have been provided in appropriate areas to
maintain stable rock slopes.
Excavation of tunnel portals will be accomplished by liberal use of pattern
rock bolting and some provision for concrete/shotcrete to reduce the risk
of unstab 1 e s 1 opes~ Speci a 1 det ai 1 s have been incorporated .; n areas where
slopes-intersect or cross larger shear zones or otherwise unstable rock.
9-11
(f) Use of Excavated Rock in Dam Construction
Since most of the rock excavation wi 11-be within the diorite and andesite
·rocks, the quality of rock will be acceptable for use in the rockfill
portions of the dam. The exception to this could be excavation in poor
quality or weathered rock which will-result in unacceptable fill. The use
of the rockfill in the dam will be be limited to portions of the downstream
shell, and in zones of rip rap material.· Proper quality control wi11 be
exercised in selecting this material.
(g) Relict Channel
A deep bedrock depression exists on the north bank of the river extending
from about 2,500 feet west of Deadman Creek northwest-toward Tsusena Creek.
The depth to.bedrock is as much as 400 feet below the surface and the
reservoir level. The overburden consists of several sequences of glacial
deposits, lake sediments, and a 11 uvi urn varying in thickness and character
both laterally and with depth. Some of these granular deposits exhibit
high permeability, and ice inclusions were noted at a depth of several
hundred feet suggesting the possibility of permafrost. The ground water
surface has not been well defined and a perched water table has been
encountered in at least one boring indicating artesian pressure, and is
also evidenced by the presence of several surface lakes. With the proposed
range of reservoir levelss these overburden deposits will become saturated.
A bedrock contour map of the Re 1 i ct Channe 1 area is presented in Figure
9.13.
A saddle dam of relatively low height is planned across the topographic low
of this Re 1 i ct Ch anne 1 . Det ai 1 s of the potentia 1 design prob 1 ems to be
dealt with in the Relict Channel and the proposed methods of treatment are
discussed further in Section 12. Additional investigation will be
necessary to properly characterize the subsurface _condition and the final
detai 1 s. of foundation treatment in the area prior to construction.
9.4 -Seismic Considerations
The seismicity of the Susitna Basin and the sources of earthquakes are discussed
in Section 7 of this report. This section presents the implications of the
sei smi city on the design of the Watana project.
(a) Seismic Design Approach
For earthquake engineering and design considerations, the project struc-
tures have been classified as either critical structures or non-crttical
structures. Critical structures inc 1 ude the dam and simi 1 ar major --struc;;ii·
tures whose failure may result in sudden and uncontrolled release of large
volumes of water which may endanger property and lives downstream. The
non-critical structures are those structures whose failure can be assessed
as an econpmic or financial loss to the projec_t in terms of lost revenue,
repair~ and/or replacement cost .. Critical structures will be designed to
safely withstand the effect of the ''Safety Evaluation Earthquake" (SEE) for
the site. No si gni fi cant damage to these structures wi 11 be accept_ed under
these conditions. The design of non-critical structures for earthquake
conditions is undertaken using conventional Uniform Building Code recommen-
dations ..
9-12
I
I
I
I
-1
I
I
I
I
I
1
I
I
I
I
I
I'
I
I
1
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
For design of critical structures the effective acceleration for the SEE
has been determined as 0.8 x actual SEE acceleration. In the case of the
earthfill structures, designs are basad on analyses using a projected time
hi star y for the se lee ted e,arthquake event. For other structures a corres-
pondingly scaled response spectrum is used. The selected SEE for Watana
was based on a consideration of two of the most severe events which might
occur. The first of these is the "terrain'1 or "detection level 11 earthquake
which has been characterized as follows:
-Magnitude: 6-1/4 to 6-1/2 -- --
-Location: Approximately 3 km from structure
-Maximum Acceleration·: ~1ean 0.55 g to 0.60 g
84th percentile 0.70 g
-Peak Spectral Acceleration: Mean 1.37 g to 1.50 g
84th percentile 1.77 g
The duration of this event is relatively short and the time history
approach to design was not developed. The response spectra for this event
are shown on Figure 10.7A. The effective peak acceleration for design of
structures is then:
-Design a max= 0.8 x 0.70 g = 0.56 g
(Sa) max : 0.56 g x 2.5 = 1.40 g
The second earthquake source to be considered for design at Watana is the
Benioff Zone.
(b) Safety Evaluation Earthquake for Watana
Although the "terrain .. earthquake would result in more severe ground
motions, the duration of these motions is relatively short and the likeli-
hood of occurrence of such an event is extremely small. A more likely
source of strong ground shaking at the Watana site is the Benioff Zone.
The estimated mean peak response spectrum for the SEE for this event is
presented in Figure 7.14, along with the 84th percentile response spectrum.
A maximum horizontal acceleration level for the 84th percentile response
spectrum for the Benioff event is approximately .
The design of the Watana Dam has been based in the projected time history
for this event as shown in Figure ___ , and as discussed in Section 12.
9.5 -Selection of Reservoir Levels
This section describes the approach used and the results of the evaluations made
in the determintion of optimum Watana reservoir 1evel. The selected elevation
of the Watana dam crest is based on considerations of the value of the hydro-
e 1 ectri c energy produced from the associ a ted reserv'oi r~, and geotechni ca 1 con-
straints on reservoir levels. Firm energy, average annu~l energy, construction
costs and operation and maintenance costs were determined for the Watana
deve.lopment with dam crest elevations of 2240, 2190 and 2140 feet. The relative
9-13
value of energy produced for ·each of these three dam elevations was then deter-
mined bymeans of the OGP generation planning model as discussed in Section 6,
to-determine the long term present worth cost of meeting the Railbelt system
energy demand. Finally the physical constraints imposed on dam height and
reservoir elevation by geotechnical considerations were reviewed and incor-
porated into the crest elevation selection process.
(a) Methodology
Firm and average annual energy produced by the Susitna development are
base(' on 32 years of hydrolog;:cal records. The energy produced was deter-
mined by using a mult-reservoir simulation of the operation of the Watan·a
and Devil Canyon reservoirs. A variety of reservoir drawdowns were
ex ami ned, and drawdowns producing the maxi mum firm energy consistent with
engineering feasibility and cost of the intake structure were selected (see
Section 9.11). Minimum flow requirements were established at both project
sites based on mechani ca 1 p 1 ant-re 1 ated restrictions and dm-Jnstream
fisheries considerations.
As discussed in Section 9.6, to meet system demand the required maximum
generating capability at Watana in the period 1993 and 2010 ranges from 665
MW to 908 M~~. For the reservoir level determinations, energy estimates
were made on the basis of assumed average annua 1 capacity requi rememts of
650 MW at Watana in 1993, increasing to 1020 MW at ~~atana in 2007, \t:'ith an
additional 600 MW at Devi 1 Canyon coming on line in the year 2002. Yearly
~ystem demand and monthly and daily load patterns within the Railbe1t over
a 29 year period·were based on forecasts developed as described in Section
5 and 6. The long term present worth costs of the generation system
required to meet the Railbelt energy demand were then determined for each
of the three· crest elevations of the Watana dam using the OGP V model.
As discussed in Section 6, these present worth costs are based on economic
parameters, not of inflation. The construction cost estimates used in the
OGP V modeling process for the Watana and Devil Canyon projects were based
on preliminary conceptual layouts and construction schedules. Further
refinement of these 1 ayout s has taken p 1 ace in the process of deve 1 opment
of the estimates presented in Section 16. These ref''\nements have no signi-
ficant impact on the reservoir 1 eve 1 selection.. The oasis of assumed costs
for construction and operation of alternative energy generation facilities
is also discussed in Section 6.
(b) Optimization
Optimization of the Watana reservoir level was based on an evaluation of
three dam crest elevations of 2240, 2190, and 2140 feet. These crest
elevations apply to the central portion of the embankment with appropriate
allowances for freeboard and seismic slumping, and correspond to maximum
operating levels of the reservoir of 2215, 2165 and 2115 feet, respec-
tively. Average annual energy calculated for each case using the reservoir
simulation model are given in Table 9 .. 3, together with corresponding
project construction costs.
9-14
I
I
;I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I'
I
I
I
I
I
I
I
•••
I
I
I
I
·I
I
I
I
I
I
I
In the determination of long term present worth of production costs, the
Susitna capital costs were adjusted to include an allowance for ·interest
duri n9 .construct 1 on and then used as input to the OGP V mode 1. Si mu 1 a ted
annual energy yields were distributed on a monthly basis by the reservoir
operation model to match as closely as possible the project energy demand
of the Railbelt and then input to the OGP V model. The long term present
worth production costs of meeting the Railbelt energy demand using the
Susitna development as the primary source of energy, were thus determined
for each of the three reservoir levels.
The results of these evaluat1ons are shown in Table 9 .. 4, and plots showing
the variation of the long term present worth with dam crest elevation are
shown in Figure 9 .17. This figure indicates that on the basis of the
assumptions used, the minimum long term present worth of production costs
occurs at a Watan.a dam crest elevation ranging from approximately 2160 to
2200 feet (reservoir levels 2140 to 2180). A higher dam crest will still
result in a development which has an overall net economic benefit relative
to displaced energy sources. However, it is also clear· that as the height
of the Watana dam is increased, the unit costs of additional energy
produced at Watana is somewhat greater than fo·r the displaced alternative
energy sources~ Hence, the long-term present worth of the overall system
increases. Conversely, as the height of the dam is 1 owe red, and thus
Watana produces less energy, the unit cost of the energy produced by
alternative generation sources to replace the lost Susitna energy, is more
expensive than Susitna energy. In thise case also, the long-term present
worth again increases.
(c) Conclusions
It is important to clearly establish the overall objective of setting the
Watana reservoir level. An objective which is to minimize the long term
present worth of energy cost will lead to selection of a lower reservoir
level than an objective vJhich is to maximize the amount of energy which can
be obtained from the available resource, while still doing so economically
and within accepted technical and environmental constr~nts.
The three values of long term present worth developed by the OGP V computer
runs defined a relationship between long term present worth cost and Watana
dam height which is relatively insensitive to dam height. There is an
indication that a small difference in system present worth occurs as the
Watana dam crest is raised or lowered ovesr the range considered. However
these di.fferences are of the same orde~ as the inaccuracies which a~e
inherent in capital cost estimates for the development of such major
generation facilities within the Railbelt. Little value would thus be
gained from analyzing intermediate dam heights to further define the curve.
The insensitivity is highlighted by the graph of present worth against dam
height in Figure 9.17. This figur€ shows these slight variations in
context within the total long term present worth cost of the system.
9-15.
Thus~ from an economic standpoint, the optimum crest elevation could be
considered as varying over a range of elevations of as much as 50 to 100
feet. The governing factors is establishing the upper limit of dam height
were consequently physical and geotechni ca 1· considerations, a 11 owing the
objective of maximizing the economic use of the Susitna resource still to
be. sati sfi ect~ ·
The normal maximum operating level of the reservoir was therefore set at
elevation 2185 feet. At this level, for up to the 1:10,000 year flood
occurrence, there will be no danger of over-topping the low lying portion
of the relict channel on the right side of the river. In the· unlikely
event of floods of greater severity, a freeboard dike in the low area of up
to 10 feet in height has been incorporated in the design" ·With this
approach, the Watana project will develop the maximum energy reasonably
available without incurring the need for costly water retaining structures
in the relict channel area.
9.6 -Selection of Insta 11 ed Capacity
The generating capacity to be installed at both Watana and Devil Canyon was
determined on the basis of generation planning studies described in Sections 6
and 8, together ~ith appropriate consideration of the following:
-Available firm and average energy from Watana and Devi 1 Canyon;
-The forecast energy demand and peak load demand of the system;
-Available firm and average energy from other existing and committed plant;
-Capital cost and annual operating costs for Watana and Devi 1 Canyon;
-Capital cost a.nd annual operating costs for alternative sources of energy
and capacity;
-Environmental constraints on reservoir operatioh; and
-Turbine and generator operating characteristics.
{a) Methodology
The following procedure was used to select the installed capacity at
Watana:
-The firm and average energy available at both Watana and Devil Canyon
was determined using a reservoir operation computer si mui ati on program
based on the 32 years of hydrological record developed as described in
Section 7 (see Plate 22).
-An assessment was made of the a 1 tern at i ve therma 1 energy required to meet
the predicted 1 cad forecast, using a computer ~; mul at ion of economic 1 oad
'dispatch from av ai 1 ab 1 e p 1 ant in the firm year. This determined the
optimum schedu 1 i ng and capacity of new therma 1 p 1 ant required to meet the
. minimum Loss-of-Load Probability (LOLP) criterion for system security.
- A determination was then made of the generating capacity required to
utilize the available energy from the Susitna Project in the hydrological
years of record, based on the following assumptions:
9-16
c
"I' s_"' .
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
' •'•
I
I
I
I
I
I
I
tl
I
I
I
I
I
I
I
I ,,
I
I'
I
I
. In a wet year, hydro energy in excess of system demand) displaces
thermal energy (from coal,· gas turbine, combined cycle, or diesel
p 1 ant).
. In an average year, where thermal energy is required to meet system
energy demand, hydro energy is used either to satisfy peak demand with
thermal energy supplying base load (Option 1); or hydro energy is used
to supply base load requirements with thermal energy at peak demand
(Option 2). The actual choice is based on made on economic load
dispatch criterla.
. Devil Canyon energy is used predominantly as base load energy because
of environmental constraints on downstream flow variations.
• The maximum installed capacity required was determined on the basis of
the est ab 1 ished peak generating capacity described above plus any hydro
standby or spinning ~eserve requirement~
(b) Total Installed Capacity ..
The required total capacity at Watana in a wet year (determined as
descri be·d above), exc 1 udi ng st-andby and spinning reserve capacity, is
summarized below. The capacities are based on the Battelle medium load
forecast.
Capacity (MW)
Demand Year Option 1 Option 2
1993 801 801
1995 839 839
2000 862 742
2002 (In~ 1 • Devi 1 Canyon) 660 655
2005 (Incl. Devil Canyon) 750 740
2010 (Incl. Devi 1 Canyon) 908 900
On the basis of this evaluation, the ultimate power generation capability
at Watana was selected as 1000 MW for preliminary design purposes, to allow
a margin for hydro spinning reserve and standby for forced outage. This
installation also provides a low cost margin in the event than an
accelerated growth of demand occurs.
(c) Unit Capacity
Selection of the unit size for a given total capacity is a compromise
betw~en the initial least cost solution, generally inv9lving a scheme with
a smaller number of large capacity units, and the improved plant efficiency
and security of operation provided by a larger number of smaller capacity
9-17
units. _Other factors include the size of each unit .as a proportion of the
total system load and the minimum anticipated load on the station_. Any
requirement Jor a mlnimum downstream flow would also affect the selecticm,
since, for example, Francis turbines will. not operate effectively at less
than about 50 percent full output. Growth of the actual load demand is
also a significant factor~ ·since the unit installation may be phased to
~atch the actual load growth~
The number of units and their i ndi vi du a 1 ratings were determined by the
requirement to deliver the design peak capacity in the critical demand
month of December, at minimum December reservoir level, with turbine wicket
gates fully open. In addition, unit selection was based on consideration
of the following:
-Rate of load growth with time;-
-Load following capability at part station operation;
-Efficiency variation with load and head; p
-Minimum acceptable load on each machine;
-Minimum downstream compensation flow;
-Standby capac1ty and spinning reserve; and
-Sensitivity to change in forecasted load growth.
An examination was made of the economioc impact on power plant production
costs for various combinations of unit numbers and rated capaci ty:s which
would provide the selected capability of 1000 MW and satisfy the considera-
tions outlined above. As discussed above, for any given installed capa-
city, plant efficiency increases as the number of units increases. This is
illustrated in Figure 9.18. The assumed capitalized value of the resulting
additional annual energy used for this evaluation was 1000 mills per kWh;
based on economic parameters developed in previously described system
studies. Variations in unit numbers and capacity will affect the cost of
the power intakes, penstocks, powerhouse, and tailrace; the differences in
these capital costs were estimated and included in the evaluation. The
results of this analysis are presented below.
Number of
Units
4
6
8
Rated Capacity
of Unit
(MW)
250
170
125
Capitalized Value of
Additional Energy
($ Millions)
40
50
Additi anal
Capital Cost
($ Millions)
31
58
It is apparent from th1s analysis that a six-unit scheme is the most
economic alternative. This scheme also offers a high degree of flexibility
and security of operation compared to the four-unit alternative, as well as
advantages if unit i nsta 11 ati on is required to be phased to match actua 1
load growth. The net economic benefits of the six-unit scheme are greater
than those of the eight-unit scheme, whi 1 e at the same time, no significant
operational or scheduling advantages are associated with the eight-unit
scheme. Accordingly, a scheme incorporating six units each with a rated
capacity of 170 MW, for a total of 1020 MW~ has been adopted for all
alternatives. -
:_.,
9-18
I
I
I
I
J
ll
I
I
I
I
I
I
I
'I
I
I
I
I
I
I
I
I ' ·.
••
1:
I
I
I
••
I
I
I
I
I
.I
'I
I
I
I
I
For proj,ect .design and cost estimating purposes in the current studys the
installed capacity of 1020 MW has been assumed. From generation planning
and financial analyses, certain advantages may be gained from staging the
installation of generating equipment over a somewhat longer period. These
aspects wi 11 be addressed further during detai 1 ed design of the project ..
The power f aci l i ties and associated equipment are described in detail in
Section 12.
9.7 -Selection of Spillway Capacity
Normal design practice for projects of this magnitude, together with applicable
design regulations, require that the project be capab 1 e of passing the probab 1 e, "c .•.
maxi mum flood (PMF) routed through the re·servoi r without overtopping the damo
In addition to this requirement, the project should have sufficient spillway
capacity to safely pass a major f1 ood of 1 esser magnitude than the probab 1 e max-
imum flood without endangering the main dam or ancillary structures, in a manner
which will avoid injury or loss of life, or damage to the project itself. The
frequency of occurrence of this flood, known as the spi ·1 lway design flood or
Standard Project Flood (SPF), is generally selected on the basis of an evalua-
tion of the risks of the project if the spillway flood is exceeded, compared to
the costs of the structures required to safely discharge the flood. A list of
spillway design flood frequencies and magnitudes for several major projects is
presented below.
Spi 11 way Design Flood Frequencies and Magnitudes
Spi~lwax Design Flooa Basin Spillway Capacity
Peak PMF After Routing (cfs)
Project
~------~----~--~--~----~~--~~~------------~------
Frequency Inflow {cfs) (cfs) Design*
Mica, Canada
Churchill Falls,
Canada
New Bullards, USA
Oro vi 11 e, USA
Guri, Venezuela
(final stage)
Itaipu, Brazi 1
Sayano, USSR ·
PMF
1:10,000
PMF
1:10,000
PMF
PMF
1:10,000
250,000
600,000
226,000
440,500
1,000,000
2,195,000
480!t000
250,000
1,000,000
226,000
711,400
1,000,000
2s195,000
N/A
150,000
230\JOOO
170~000
440,500
1,000,000
2,195,000
680s000
*All spillways except Sayano have capacity to pass PMF with surcharge
On the basis of the foregoing, a spillway design flood with a return frequency
of 1:10,000 years was selected for Watana.
9-19
. -
></ ;:; ~ ,~,..., ~ '''• -.<"-•4 ''' ,, ,,.,,.. . ..,,".' ' '•"~ ..... ~;;. "''""' • <;,••-."''•·•••!;~ ... ,_ -~~~--·J "'" "•••"'>~>-'""~•:...,._ 0 h ••' ''"'"" "'""' 0 .,,,)>,...,., •• , •. -' •• • • • < '• ., "••' • • •••• •-~ oo-<o.---"'"•'"w --~~ "' •'""'" '"' •• ,,., • 0 • ....
The flood frequency analysis undertaken as described in Section 7.2 indicates
the following values:
Flood
Probable Maximum
Spillway Design
Frequency
--1:10,000
Inflow Peak
326~000 cfs
156~000 cfs (0.47 PMF)
-AdditioP.al capacity required to pass the Ptt1F discharge will be provided by an
emergency spillway consisting of a fuse plug and rock channel cut on the right
bank.
9.8 -Main Dam Alternatives
This section describes the a 1 ternati ve types of dam considered at the Wat ana
site and the basis for the selected. alternative.
(a) Comparison of Embankment and Concrete Type Dams
The selection between an embankment type or a concrete type dam is usually
made based on the configuration of the ·~·alley, the condition of the founda-
tion rock and depth of the overburden, and the relative availability of
construction materials. Previous studies by the COE envisaged an embank-
ment dam at Watana. Initial studies as part of these current evaluation
included comparison of a earthfi 11 dam with a concrete arch dam at the
site. An arrangement for a concrete arch dam alternati ;e at Watana is
presented in Plate 11. The results of this analysis indicated that the
cost of the embankment dam was somewhat lower than the arch dam, based on
the use of concrete costs significantly lower than comparable costs used
for Devil Canyon. This preliminary evaluation did not indicate any signi-
ficant advantages for the concrete arch relative to the arrangement of
other structures, or the construction schedu 1 e relative to the use of a
concrete arch.
Based on the overall cost differences described above, and the likelihood
that the cost of the arch dam wou 1 d increase re 1 at i ve to that of the
embankment dam, the arch dam a·lternative was eliminated from further
consideration.
(b) Selection of Dam Type
The deve 1 opment of the design of the main dam, togethP.r with a description
of the various features of the dam, is given in Section 12. The dam is, of
course, the central and most costly component of the project, and a brief
discussion of the development of the finally selected design, together with
some of the factors which influenced deve 1 opment of the genera 1 arrangement
are presented in this section.
Selection of the configuration of the embankment dq.m cross-section \~as un-
dertaken within the context of the following basic considerations:
-The avai 1 ability of suitable construction materials within economic haul
distance, particularly impervious core material;
'9-20
. : ..
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
'I.
!
I
I
I
·~
I
I'
••
I
I
I
I
-The requirement that the dam be capable of withstanding the effects of a
significant earthquake shock, as well as the static loads imposed by the
reservoir and its own weight;
-The relatively limita· constrtiction season available for placement of
compacted fi 11 materia 1 s.
The exp lor at ion program undertaken during 1980 and 1981 indicated that ade-
quate quantities of materia 1 s sui tab 1 e for dam construction were located
within reasonable distance from the site. The locations of potential
borrow materials for the dam are shown on Figure ·9.6. The well graded
silty sand material from Borrow Area D is considered the most promising
source of impervious filt. Compaction tests indicate a natural moisture
constant slightly on the wet side of optimum moisture content, so that
control of moisture content will ·be critical in achieving a dense imper-
vious core \'lith high shear s"trength.
Potential sources for the upstream and downstream shells included either
river gravel from borrow areas along the Susitna Riyer~ or compacted rock
fill from structural excavation of quarries.
The main dam wi 11 consist of a compacted impervious core protected by fine
and coarse filter and transition zones on both the upstream and downstream
slopes of the core. The upstream and downstream outer supporting f"ill
zones wi 11 comprise t"'e 1 ati ve ly free draining materials such as compacted
gravel or rockfill, providing stability to the overall embankment struc-
ture. The location. and inclination of the impervious core is fundamental
to the design of the embankment. Two basic alter·natives exist in this
regard.:
- A vertical core located centrally within the dam; and
-An inclined core with both faces sloping upstream.
The advantages and disadvantages of these two alternatives are discussed in
Section 12. A central vertical core was chosen for the embankment based on
a review of precedent design and the nature of the avai 1 able potential i m-
pervious material.
In order to evaluate the relative sensitivity of the project arrangement.to
changes in.exterior dam slopes, two alternatives were used in the prelim-
inary review:
-2.4H: lV upstream and 2H:1V· downstream
-2.25H: lV upstream and 2H:lV downstream
As part of the i ntermed1 ate review, the volume of the dam with an upstream
slope of 2o4H:lV was.computed for four alternative dam axes. The location
of these alternative axes are shown on Plate 12. The results of this
comparison are described below:
9-21 !
. .. .J
Alternative
1
2
3
4
Total Volume
(millions c.y.)
69.2
71.7
69.3
71.9
During the intermediate review, the upstream slope of the dam was flattened
to 2.75H:1V. This slope was based on a conservative estimate of the
effective shear strength parameters of the available construction
materia 1 s, as well as a conservative a 11 owance in the design for the
effects of earthquake shock on the dam.
During the fi na 1 review stage, the exterior upstream slope of the dam was
steepened from 2.75H:lV to 2.4H:lV, reflecting the results of the static
and dynamic design analyses being undertaken at the same time as the
general arrangement studies. This section was used for the final review of
alternative schemes. Further refinements to the design were subsequent 1 y
incorporated in the final design presented in Section 12, but these did not
influence the selection of the final scheme.
9.9 -Diversion Scheme Alternatives
The topography of the site essentially dictated that diversion of the river
during construction be accomplished using one or two diversion tunnels with
upstream and downstream cofferdams protecting the main construction area.
The configuration of the river in th~e vicinity of the site favors location of
the diversion tunnel or tunnels on the right bank, since the tunnel length for a
comparable scheme on the left bank could be approximately 2,000 feet greater.
In addition, rock conditions on the right bank are more favorable for tunneling
and excavation of intake and outlet portals. Notwithstanding these considera-
tions, the selection process for establishing the final general arrangement
included examination of tunnel locations on both banks.
(a) Design Flood for Diversion
The recurrence interval of the design flood for diversion is generally
established based on the characteristics of the flow regime of the river,
the length of the construction period for which diversion is required and
the probable consequences of overtopping of the cofferdams. These last two
considerations are usually evaluated as part of an economic risk analysis
in which the cost of the diversion scheme, and the risks involved in
exceeding the capabilities of the scheme.. This type of analysis wi.ll be
undertaken as part of the detailed design phase of the project, but for the
purposes of the feasibility analysis, design criteria and experience from
other projects similar in scope and nature have been used.
9-22 .,
I
I
'I
I
I
I
'I
I
I
I
I
'I
I
I
'I
I
I
I
I
I
I
I
:a
I
I
' I
I
I
I
I.
I
I
I
I)
I
I
'" I .• ''-·
At Watana, damage to the parfia1ly completed project tr~ould be significant,
or more importantly, would probably result in at least a one-year delay in
the completion schedule •. A ·preliminary evaluation of the construction
schedule indicates that the diversion scheme would be required to operate
for 4 or 5 years until the dam was completed sufficiently to permit initial
filling of the reservoi~. A design flood with a return frequency of 1:50
years was selected based on experience and practice with other major
hydroelectric projects. This approximates a 90 percent probability that
the cofferdam 1Ni 11 not be overtopped over the critical 5 year constructi·on
period. The equivalent inflow for the design_ flood together with average
flow char acteri st i c s of the river s i gni fi cant to diversion are presented
below:
Average annual flow
Maximum average monthly flow
Minimum average monthly flow
Design flood inflow (1:50 years)
(b) Goff erd ams
7,860 cfs
23,100 cfs (June)
890 cfs (March)
81,:100 cfs
The character and considerable depth of riverbed alluvium at both cofferdam
sites indicate that embankment type cofferdam structures waul d be the on 1 y
technically and economically feasible alternative at Watana. For the
purposes of establishing the overall general arrangement of the project and
for subsequent diversion optimization studies, the upstream cofferdam
section adopted comprised an initial closure section approximately 20 feet
high constructed in the wet, with a zoned embankment constructed in the
. dry. The downstream cofferdam comprises a closure dam structure approxi-
mately 30 feet high placed in the wet. Control of underseepage through the
relatively pervious underlying alluvium mat·erial will be achieved by means
of a soil/ bentonite slurry wall.
The selected cofferdam sections are described in more detail in Section
12.
(c) Diversion Tunnels
A basic consideration in evaluation of any diversion tunnel'scheme is an
ex ami nation of the advantages and disadvantages of concrete-lined tunne 1 s
compared to unlined tunnels. Preliminary hydraulic studies indicated that
the design flood routed through the diversion scheme would result in a de-
sign discharge of approximately 80,500 cfs. For concrete-lined tunnels,
design velocities of the order of 50-feet per second have been used in sev~
eral projects. For unlined tunnels~ maximum design velocities ranging from
10 fps in good quality rock to 4 fps in less competent material are
typical. Using a maximum permissible velocity of 10 fps!! four unlined
tunne 1 s each with an equivalent diameter of 50 feet waul d be required to
pass the design flow. Alternatively~ a design velocity of 50 fps would
theoretically permit the use of one concrete~lined tunnel with an
equivalent finished d1qmeter of 44 feet. The unlined tunnels would require
4.5 times as much excavation as the lined alternative, together with at
least four times as much tunnel support cost. This would only be partially
9-23 __
(d)
offset by the cost of the concrete lining~ Apart from co'St, the most
important single factor relates to the security and reliability of the
diversion scheme. The tunne 1 s wi 11 undoubted 1 y traverse numerous
unfavorable geological conditions and structures, as yet undefined, during
construction. The reliability of an unlined tunnel is more dependent on
rock conditions than is a lined tunnel, particularly given the extended
period during which the diversion scheme is required to operate. These
considerations, together with cost and the somewhat questionable feasibi 1-
i ty of a tunne 1 with a diameter approaching 50 feet in this type of rock,.
are considered sufficient to eliminate consideration of unlined tunnels for
the diversion scheme. ·
The following alternative lined tunnel schemes were examin~d as part of
this analysis:
-Pressure tunnel with a free outlet;
-Pressure tunnel with a submerged outlet; and
-Free flow tunnel.
Pressure tunnels are designed to flow full and accordingly must withstand
internal pressure. The most widely used type of pressure tunnel for diver-
sion has the crown of the outlet portal submerged during all flow
conditions ..
Emergency Release Facilities
While not an integral part of the diversion scheme itself, the emergency
release facilities greately influenced the number, type, and arrangement of
the diversion tunnels selected for the final scheme.
At an early stage of the study, it was established that, in accordance with
current design practice, some form of low level release facility was
roequired to permit lowering o~ the reservoir in the event of an extreme
emergency. Since the primary discharge facilities wil1 be located near the
crest of the dam, they would be ineffective if the reservoir level had to
be reduced below approximate elevation 1950. The most economical
alternative available would involve converting an existing diversion tunnel
to permanent use as a low level outlet facility. Since it obviously would
be necessary to maintain the diversion scheme in effective service during
construction of the low level outlet works, two or more tunnels would be
required if this a 1 ternati ve was adopted. The use of two diversion
tunnels, while not contributing to the overall economy of the project,
prov1des an additional measure of security to the diversion scheme in case
of the loss of service of one tunnel during an emergency. The use of two·
tunnels also provides greater flexibility in construction scheduling:>
particularly since concrete-lined tunnels are required. Additionally,
potential problems with stability of two smaller openings are likely to be
1 ess severe than for the 1 arger spans ,g.ssoci ated with a single tunnel ..
9-24
I
I
I
I
I
I
I
I
I
I
I
I
I
I.
I
I
I
I
"I
J.
I
I
I.
I
If operation of the emergency low level release facilities is required, it
\t~i ll extend over a considerable period of time. Discharge of the faci li-
ties at the heads required could r<esult in serious erosion downstream ..
This requirement necessitated some form of energy dissipation prior to
returning the reservoir water to the river. Given the space restrictions
imposed by the size of the diversion tunne 1, it was decided to uti 1 i ze a
double expansion system with concrete plugs within the tunnel. The
operation of the expansion chamber is described in Section 12. The use of
this arrangement requires that the chamber be located above tai lwater to
prevent ~cavitation in the area of the emerging jets from the downstream
plug. The implications of this restraint require that if a diversion
tunnel is to be used as part of the emergency low level release facilities,
it must act as a free flow tunnel.
II (e) Optimization of Diversion Scheme
I
' .,
I
I ,,
I \.
~--·
I
' I
I
I
Given the considerations described above relative to design flows, coffer-
dam configuration and alternative types of tunnels, an economic study was
undertaken to determine the optimum combination of upstream cofferdam
height and tunne 1 type and diameter ..
Capital costs were developed for three heights for an upstream cofferdam
embankment with a 30 foot wide crest and exterior slopes of 2H:1V. A
freeboard allowance of 5 feet for settlement and wave runup and 10 feet for
the effects of downstream ice jamming on tailwater was adopted.
Capital costs for the a::·,;)ociated tunnel alternatives included allowances
for excavation, conc\·ete liner, rock bolts, and stee'l supports. Casts were
a 1 so deve 1 oped for the upstream and downstream por·t a 1 s. inc 1 udi ng
excavation and support. The cost of intake and out 1 et gate structures and
associated gates was determined not to vary significantly with tunnel
diameter and was excluded from the analysis.
A right bank configuration was selected· and the corresp?nding tunnel length
in all cases was assumed to be 4,700 feet.
Curves of headwater elevation versus tunnel diameter for the various tunnel
alternatives with submerged and free outlets are presented in Figure 9.13.
The relationship between capital cost and crest elevation for the upstream
cofferdam is shown in Figure 9.14. The capital cost for various tunnel
diameters with free and submerged outlets is given in Figure 9.15.
The results of the optimization study are presented in Figure 9.16" and in-
dicate the following optimum solutions for each altern-ative.
Cofferdam
TypP of Tunnel Diameter (ft) Elevation (ft} Total Cost ($)
2-Pressure tunnels 30 1595 66,000,000
2-Free flow tunnels 32.5 1570 68,000,000
2-Free flow tunnels 35 1545 69,000,000
9-25
I
The foregoing indicates that a relatively small cost differential (4 to 5 J.~."·. · ..
percent) separates the various alternatives in range of tunnel diameter :
from 30 to 35 feet.
(f) Selected Diversion Scheme J
Although a scheme incorporating two 30 foot diameter pressure tunnels with
submerged outlets is marginally the most economical soiution as discussed
in (d), at le.ast one tunnel must have a free outlet if it is to be con-
verted into a low level outlet tunnel.
An important consideration at this point is cofferdam closure. For the
pressure tunnel scheme, the invert of the tunnel entrance is below riverbed
elevation, and once the tunnel is complete diversion can be accomplished
with a closure dam section approximately 10 feet high. The free flew
tunnel scheme however requires a tunnel invert approximately 30 feet above
riverbed level, and diversion will involve an end-dumped closure section 50
feet high.. Two basic problems are associated with closure embankments of
this height -velocities during final closure would be quite high, requir-
ing 1 arge size stone to remai~n in p 1 ace, and subsequent sea 1 i ng of the
closure embankment-in the wet must be done at significant depth, with
relatively less control than for lower embankments.
In consideration of these problems and restraints, a combination of one
pressure tunnel and one free flow tunnel (or pressure tunnel with free out-
1 et) was adopted. This wi 11 permit i ni ti a l diversion to be made using the
lower pressure tunnel, thereby simplifying this critical operation and
avoiding potentially serious delays in the schedule. Two alternatives were
re-evaluated as follows:
Tunne 1 Diameter
(feet)
30
35
Upstream Cofferdam
Crest Elevation Approximate Height
(feet) (feet)
1595
1545
150 <:
100
More detailed layout studies indicated that the higher cofferdam associated
with the 30 foot diameter tunnel alternative \vould require locating the in-
let portal further upstream into "The Fins" shear feature. Since good rock
conditions for portal construction are essential, and the 35 foot diameter
tunnel alternative would permit a portal location downstream of 11 The Fins 11,
this latter alternative was adopted. As noted in {e), the·overall cost
difference was not significant in the range of tunnel diameters considered,
and the scheme incorporating two 35 foot diameter tunnels with an upstream
cofferdam crest elevation 1545 was incorporated as part of the selected
general arrangement.
The various components of the selected diversion scheme are described in
Section 12.
9-26
I
I
I
I
I
I
I
I
I
I
'I
I
' I
I
I
I
.I
I
I:
I
I,
I
I
I
I
I
I.
I
I
·~-
1 '·' . !I
I
,---~--,-, -.·.~----,.,-, .. ' l
9.10-Spillway Facilities Alternatives
As discussed in Section 9.7, the project has been designed to safely pass the
floods with the following return frequencies:
Flood
Design Flood
Probable Maximum Flood
Frequency
1:10,000 years
Spillway
Discharge ( cfs)
120,000
235,000
Discharge of the spillway design flood wi 11 require a gated service spillway on
either the 1 eft or right bank. Three basic alternative spi 11way types were
ex ami ned:
/
-Chute spillway with flip bucket;
-Chute spillway with stilling basin; and
-Cascade spillway.
Consideration was also given to combinations of these alternatives with or
without supplemental facilities such as valved tunnels and an emergency spillway
fuse plug for handling flood con~itions.
Clearly the selected spillway alternatives will greatly influence and be
influenced by the project general arrangement. A discussion of the development
of the general arrangement is presented in Section 9.12.
(a) Energy_ Di ssi pati on
The two chute spillway alternatives considered effect energy dissipation
either by means of a flip bucket which directs the spillway discharge in a
free-fall jet into a plunge pool in the river well downstream from the
structure, or a stilling basin at the end of the chute which dissipates
energy in a hydraulic jump. The cascade type spillway limits the fr,ee fall
height of the discharge by uti 1 i zing a series of excavated steps do\'Jn to
river level, with energy dissipation at each step and reduction of the
velocity heads.
All spillway alternatives were assumed to incorporate a concrete agee type
control section controlled by fi.xed roller vertical lift gates. Chute
spillway sections were assumed to be concrete lined, with ample pr 'lision
for air entrainment in the chute to prevent cavitation and pres sur::.:. ~eli ef
drai~s and rock anchors in the foundation. A detailed description 6i the
selected spillway alternative is given in Section 12.
(b) Environmental Mitigation
During development of the general arrangements for both Watana and [}evi 1
Canyon~ a restriction was imposed on the allowance of excess dissolved
nitrogen in the spillway discharges. Supersaturation occurs when aerated
flows are subjected to pressure increases~ forcing excess nitrogen into
9-27
1
"f
I
I
:· ,,;.;,.S· ... , .. ~,. 'S •• ~.,.1 .. ~ ,,,:"·v~· "·"·· ·, . ,m.~· .,.,,_,. ~,. ~'*'"'''!'" ~.·'··~-~.~·''-"I"''""'~·~;.,_~~.!~~!..!C._-'~~tb_.:....•_!_>::::_,·~-·':_._~~~:.c0~~,,.u, •. _,<!•'~.·~·"',.,"'""'·-·'"''"''i'';.,..._~cc~"'~\·<"'"'·'""' .. '' ,,..~-. ... ,H-.• ...... ...-fll.•:.·-..._ .. ,.1--,..·~"·"'"""'-•~•-H•""-•~~-.""'' '~·~'-'"~""'····~·-'>-"-·,·· w ·-.~·#-~--.-~.t .. .-.c•
solution~ This occurs when water is subjected to pressures approaching two
atmospheres and would occur in deep plunge pools or at large hydraulic
jumps.. The excess nitrogen would not be dissipated within the downstream
Devil Canyon reservoir and a buildup of nitrogen concentration could occur
throughout the body of water. It would eventually-be discharged downstream
from Devil Canyon with extreme··Jy harmful effects on the fish population.
On the basis of an evaluation of the related impacts, and discussions with
interested federal and state environmental agencies, spillway facilities
were designed to limit discharges of nitrogen supersaturated water from
Watana to a recurrence period of not less than 1:50 years.
9.11 -Power Facilities Alternative
Selection of the optimum power plant development involved consider~ion of the
fol!owing:
-Location, type and size of the power plant;
-Geotechnical considerations;
-Number, type, size and setting of generating units;
-Arrangement of 1 ntake and water passages; and
-Environmental constraints.
The se 1 ecti on of the i nsta 11 ed capacity of 1020 MW at Watana is described in .
detail in Sect1on 9.6. The detailed comparison of power facilities alternatives
is described in Appendix D. A summary of the general conclusions is described
below.,
(a) Comparison of Surface and Underground Powerhouse
Preliminary studies were carried out to compare the construction casts of a
surface powerhouse and an underground powerhouse at Watana. These studies
were undertaken on the basis of preliminary conceptual layouts assuming
units and an installed capacity of 840 MW. The comparative cost --.---estimates for powerhouse civil works and electrical and mechanical equip-
ment (excluding common items) indicated an advantage in favor of the under-
ground powerhouse of $16,300,000. The additional cost for the surface
powerhouse arrangement is predominantly associated with the longer pen-
stocks and the steel linings required. Although construction cost esti-
mates for a 1020 MvJ plant would be somewhat higher, the overall conclusion
favoring the underground location would not change.
The underground powerhouse arrangement is also better suited to the severe
winder conditions in Alaska, is less affecteci ~Y river. flood floes in
summer, and is aesthetically less obtrusive~ l)is arrangement has
therefore.been adopted for further development.
(b} Comparison of Alternative Locations
Preliminary studies were undertaken during the development of conceptual
project layouts at Watana to investigate both right and. left bank loc·ations
for power facilities. The configuration of the site is such that left bank
locations generally required longer penstock and/or tailrace tunnels and
were therefore strictly more expensive.
9-28
I
I
I
I
I
,I
I
I
I
' I
I
I
I
:I ..
ill ~
I
I
I
I
I
J
I "'· ....... '-
' I -,.;
I
I· .......
I
1
I.
I
I
I ·.
' t
I
~--
'I
I~.
The location of the left bank was also not favored because of indications
that the underground facilities would be located in relatively poor quality
rock. The underground powerhouse was therefore located on the right bank
such that the major openings lay between the two major shear features (*'The
Fi ns 11 and the "Fi ngerbuster••) ..
(c) Underground Openings
Cost estimates have been based on assumptions of full concrete lining of
the penstocks and tai 1 race tunnels. The 1 atter is a conservative assump-
tion for preliminary design; in practice, a large proportion of the tail-
race tunnels could be unlined, depending on the· actual rock quality
encountered.
The mini mum center-to-center spacing of rock tunne 1 s and caverns has been
assumed for layout studies to be 2.5 times the size of the larger
excavation.
(d) Selection of Turbines
The sel~ction of unit type is governed by the available head and flow, an~
economic considerations. For the design head and specific speed, Francis
type turbines have been selected; these have a reasonably flat
load-efficiency curve over a range from about 50 percent to 115 percent
rated output, and a peak efficiency of about 92 percent.
The number and rating of 0 individual units is discussed in detail in Section
9.6. The final arrangement selected is six units of 170 MW rated at
minimum reservoir level (from reservoir simulation studies) in the peak
deman9 month (December) at fu 11 gate. Th.e unit best efficiency output at
rated head {680 feet) is 181 MW.
(e) Transformers
The selection of transformer type, size, location and step-up rating is
descr1bed in Section 12.18 and summarized below:
-Single phase tran·sformers are required because of transport limitations
on Alaskan roads and railways;
-Direct transformation from 15 kV to 345 kV is preferred for overall
system transient stability;
-An underground transformer gallery has been selected for minimum total
cost of transformers, cables, but, and transformer losses; and
- A grouped arrangement of three single ~hase transformers for each two
units has been selected to rer~\;ce the physical size of the transformer
gallery and to provide a tran.iformer spacing comparable with the unit
spacing.
9-29
(f) Power Intake and Water Passages
The power intake and approach channel are significant items in the cost of
the overall power facilities arrangement. The size of the intake is
controlled by the number and minimum spacing between the penstocks!~ which
in turn is dictated by geotechnical considerations (Sections 9.2 and 9.3) ..
The preferred penstock arrangement comprises six inrlividual penstocks, one
for each turbine~ With this arrangement, no inlet valve is required in the
powerhouse since penstock dewatering can be performed by using the control
gate at the intake. An alternative arrangement with three penstocks was
considered in detai 1 to assess any possible advantages. This scheme would
require a bifurcation and two inlet valves on each penstock and extra space
in the powerhouse to accommodate the inlet valves. Estimates of relative
cost differences are summarized below:
Item
Intake (increment)
Penstocks (increments)
Bi fur·c at ions
Valves
Power·house
Capitalized Value of Ext~a Head
Loss
Total
Cost Difference
6 Penstocks 3
0
($000)
Penstocks
-20.0
-3.0
+ 3.0
+ 4.0
+ 8.0
+ 6.0
-2.0
Despite a marginal saving of $2 million (or less than 2%) in favor of three
penstocks, in a total estimated cost of $120 million, the arrangement of
six individual penstocks has been retained. This arrangement provides
improved flexibility and security of operation.
The preliminary design of the power facilities involves two tailrace
tunne 1 s 1 eadi ng from a common surge chamber. An alternative arrangement
with a single tailrace tunnel was also considered, but no significant cost
saving was apparent.
Optimization studies on all water passages were carried out to determine
the minimum total cost of initial construction plus the capitalized value
of anticipated energy losses caused by conduit friction~ bends and changes
of section. For the penstock optimization, the construction costs of the
intake and approach channe 1 were inc 1 uded, as a function of the penstock
diameter and spacing. Similarly, in the optimization studies for the
tailrace tunne 1 s, the costs of the surge chamber were included~ as a
function of tailrace tunne 1 diameter.
(g) Environmental Constraints
Apart from_the potential nitrogen supersaturation problem discussed in
Section 9.10, the major environmental constraints on the design of the
power facilities are:
9-30
I
·'1;. ~ -
I
I .j
I
I """'
• "';;'_ .. '
I .... -
I
I "'-•
I
I
I
I ... -~
' a
I '
I
I
I
I
I ·-
1
I
'
' t.
I
t
I
I
I
' t
I
' I
I
.. Contra l of downstream river temperature's; and
.. Control of downstream' flows.
The intake design has -been modified to enable power plant flows to be drawn
from ~he reservoir at four different levels through the anticipated range
of dra\'idown in order to control the downstream river temperatures within
acceptable limits.
Guaranteed minimum flows at Gold Creek during the critical summer months
have been studied to mitigate the project impacts on salmon spawning
downstream of Devil Canyon. These minimum flows represent a constraint on
the reservoir operation, and influence the computation of average and firm
energy from the Susi tna development. These studies are discussed in detai 1
in Section 15.
In average to wet years, the Watana development will be capable of
operating as a daily peaking plant for load following. The actual extent
of daily peaking will be dictated by unit availability, system demand, unit
generating costs, system stability, etc., (as described in Section 15).
Predicted downstream water level fluctuation caused 'By daily peaking at
Watan& is within acceptable limits ..
9.12 -S"election of Watana General Arrangement
Preliminary alternative arrangements of the Watana Project were developed and
subjected to a series of review and screening processes. The layouts selected
from each screening process were developed in greater detail prior to the next
review, and where necessary, additional layouts were prepared combining the
features of two or more of the alternatives. Assumptions and criteria were
evaluated at each stage and additional data incorporated as ne~essary. The
selection process followed the general selection methodology established for the
Susitna project, and is outlined below.
(a) Selection Methodology
The determination of the project genera 1 arrangement at ~4atana was
undertaken in three distinct review stages: preliminary, intermediate., and
final. ·
(i) Preliminary Review
This comprised four steps:
-Step 1: Assemble avai 1 able data;
Determine design criteria; and
Establish evaluation criteria.
-Step 2: Develop preliminary layouts based on the above data and
design criteria including all plausible alternatives for
the constituent facilities and structures.
-Step-3: Review all layouts on the basis bf technical feasibility,
readily apparent cost differences, safety, and environ-
ment a 1 imp act.
,, '
,-;.';
(i i)
(iii)
~ Step 4: Select those layouts that can be identified as most
favoriible~~Pased on the evaluation criteria determined
under Step lc, taking into account the preliminary nature
af the work at this stage. ·
Intermediate Review
This involved a series of 5 steps:
-Step 1: Review all data, incorporating additional data from other
work tasks.
Review and expand design criteria to a greater 1 eve1 of
detail.
Revie~ evaluation criteria and modify, if necessary~
-Step 2: Revise selected layouts on basis of the revised criteria
and additional data. Prepare plans and principal
sections of layouts.
-Step 3: Prepare quantity estimates for major structures based on
drawings prepared under Step 2.
-Step 4:
-Step 5:
Develop a preliminary construction schedule to evaluate
whether or not the selected layout will allow completion
of the project within the required time frame.
Prepare a preliminary contractor's type estimate to
determine the overall cost of each scheme. ·
Review all layouts on the basis of technical feasibility,
cost impact of possible unknown condi ti ens and uncertain-
ty of assumptions, safety, and environmental impact.
Se 1 ect the two most f avorab 1 e 1 ayouts based on the
evaluation criteria determined under Step 1~
Fi na 1 Review
-Step 1:
-Step 2:
Assemble and review any additional data from other work
tasks.
Revise design criteria in accordance with additional
available data.
Finalize overall evaluation criteria.
Revise or further develop the two layouts on the basis of
conclusions from Step 1. Determine over a 11 dimensions of
structures' water passages' gates' and other key ; terns ·•
' 9-32
I
I
I J
I
I
I
'
' I
t
I
I
I
J;
-~
f
'
' I "-*
II
I -
I
~..--..
·t
~ ..
I
I
I
'
' I ~
t
I
I
I
I J•
' t
I
I.
I
-Step 3: Prepare quantity take-offs for all major structures.
Review cost components within a preliminary contractor's
type estimate using the most recent data-and criteria,
and develop a construction.schedule.
Determine overall direct cost of schemes.
-Step 4: Review all layouts on th~ basis of practicability,
technical feasibility, cost, impa:ct of possible unknown
conditions, safety, and environm~ntal impact.
-Step 5: Select the final layout on the basis of the evaluation
criteria developed under Step 1.
(b) Design Data and Criteria
As discussed above, the review process included assembling of relevant de-
sign data, establishing preliminary design Criteria, and expanding andre-
fining these data during the intermediate and final revie\vs of the project
arrangement. The design data and design cri ter i a which eva 1 ved through the
final review is presented in Table 9.5. Data and criteria developed during
the preliminary and intermediate review stages are given in Appendix D for
reference.
(c) Evaluation Criteria
The various layouts were evaluated at each stage of the review process on
the basis-of the criteria summarized in Table 9.6. The criteria listed in
Table 9.6 illustrate the progressively more detailed evaluation process
1 eadi ng to the final se 1 ected arrangement.
9.13 -Preliminary Review
The development selection studies described in Section 8 involved comp~risons of
hydroelectric schemes at a number of sites on the Susitna River. For these
comparisons a preliminary conceptual design was·developed for the Watana project
known as the "DSR Sci1emes 11 • ··
Eight further 1 ayouts were subsequently prepared and ex ami ned for the Watana
project during this preliminary review process, in addition to the DSR scheme.
These eight layouts are shown in schematic form on Plate 13. Alternative 1 of
these·layouts was that recommended for further study in the Development
Selection Report.
This section describes the preliminary review Ur1dertaken of alternative Watana
1 aym·?~s.
(a) Basis of Comparison of Alternatives
Although it was recognized that provision would have to be made for
downstream r·e leases of water during filling of the reservoir and for
emergency reservoir drawdown, these features were not incorporated in these
9-33
preliminary layouts. These facilities would either be inter-connected with
the diversion tunnels or be provided for separately. Since the system
selected would be similar for all layouts with minimal cost differences and
little impact on other structures, it was decided to exclude these
faci 1 i ties from over a 11 assessment at this ear 1 y stage.
Ongoing geotechnical explorations had identified the two major shear zones
c~ossing the Susitna River and running roughly parallel in the northwest
direction. These zones enc'lose a stretch of watercourse approximately
4500 feet in length (see Section 9.2). Preliminary evaluation of the
existing ge.ological data ·indicated that the fracture materials and infill
within the actual shear zones would be unable to support standard tunneling
methJds and would be inadequate for founding cf massive concrete
structures. The originally proposed dam axis was located between these
shear zones, and as no apparent major advantage cppeared to bn gained from
large changes in the dam location, layouts generally were kept within the
confines of these bounding zones.
An earth and rockfi 11 dam as described in Section 9.8 was used as the basis
for all layouts. The downstream slope of the dam was assumed as 2H~1V in
all alternatives, upstream slopes varying between 2.5H:1V and 2.25H:lV were
examined ·;n order to determine the ifluence of variance in the dam slope
on the congestion of the layout. In all these preliminary arrangements,
except that prepared for the DSR, cofferdams were incorporated within the
body of the main dam.
Floods greater than the routed 1;10,000 year spillway flood and up to the
probable maximum flood were assumed to be passed by surcharging the
spillways except in cases where an unlined cascade or stilling basin type
spillway served as the sole discharge facility. In such instances, under
1 arge surcharges, these spi 11 ways wou 1 d not act as efficient energy
di.ssipators but would be drowned out, acting as steep open channels \'lith
the possibility of their total destruction. In order to avoid such an
occurrence the design flood was considered as the routed probable maximum
flood.
On the basis of information existing at the time of the preliminary review,
it appeared that an undergrpund powerhouse cou 1 d be located on either side
of the rive~. A surface powerhouse on the right bank appeared feasible but
was precluded from the left bank by the close proximity Of the downstream
toe of the dam and the adjacent broad shear zone. Locating the powerhouse-
further downstream would require tunneling across the shear zone, whtch
would be expensive, and excavating a talus slope. Furthermore~ it was
found that a left bank surface powerhouse would either interfere with a
left bank spillway or would be directly impacted by discharges from a right
bank spillway.
(b) Description of Alternative
(i) Preliminary DSR Scheme
The preliminary I}DSR scheme as shown on Plate 3 has a dam axis
location similar to that originally proposed by the COE, and a right.
bank double stilling basin spillway. The spillway follows the ·
9-34.
t
I
-.. -'
I _ ..
I
I
'
'
' I ....
t
I
I
I
' I
I
' I
I
I
I ~--
, 'I
{1_
' I
'
' I
' f
I
I
I
t
' I
I
I
. ,
I
shortest line to the river avoiding interference with the dam and
-discharges downstream, almost parallel to the flow~ into the center
of the river. A substantial amount of excavation is required for
the chute and stilling 'basins, although most of this material could
probably be used in the dam~ A large volume of-concrete is also
required for this type of spillway, however, and the system would be
very costly. The maximum head dissipated within each stilling basin
is approximately 450 feet, within world experience, and cavitation
and erosion of the chute and basins should not be a problem if the
structures are properly designed. Extensive erosion downstream
would not be expected. The diver·sion fo11o~rs the shortest route,
cutting the bend of the river on the right bank, and has-inlet
portals as far upstream as possible-without having to tunnel through
11 The Fi nsu. It is possi b 1 e that the underground powerhouse is in
the area of "The Fi ngerbuster", but it caul d be located upstr.eam
almost as far as the system of drain holes and galleries just
downstream of the main dam grout curtain.
(ii) Alternative 1
This alternative is that recommended for further study in the
Development Selection Report and is similar to the preliminary DSR
1 ayout, except that the right side of the dam has been rotated
clockwise, the axis relocated upstream and the spillway changed to a
chute and flip bucket. The revised dam al~gnment resulted in a
slight reduction in total dam volume compared to the DSR alterna-
tive. A localized downstream curve was introduced in the dam close
to the right abutment in order to reduce the length of the spillway.
The alignment of the spillway is almost parallel to the downstream
sec~ion of the river and it discharges into a pre-excavated plunge
pool. in the river approximately 800 feet downstream from the flip
bucket~ This type of spillway should be considerably less costly
than one incorporating a stilling basin, provided that excessive
excavation of bedrock within the plunge pool area is not required.
Careful design of the bucket will be required however, to prevent
excessive erosion downstream causing undermining of the valley sides
and/or build up of materia 1 downstream which could cause elevation
of the t ai lwater 1 eve 1 s ~
(iii) Alternatives 2 through 20
Alternative 2 consists of a left bank cascade spillway with the main
dam axis curving downstream at the abutments. The cascade spillway
would require an extremely large volume of excavation but it is
probable that most of this material, with carefule scheduling~ could
be used in the dam. The excavation would cross "The Fingerbuster"
and extensive dental concrete would be required. In the upstream
portion of the spillway~ velocities would be relatively high because
of the narrow configuration of the channel and erosion could take
place in this area in proximity to the dam. This discharge from the
spillway enters the river perpendicular to the general flow but
velocities would be rei atively low and should not cause substanti a1
erosion prol1lt:p1s. The powerhouse is in the most suitable location
for a surtace ~alternative where the bedrock is close to the surface
and the overall slope is approximately 2H:lV •
"
Alternative 2A is similar to Alternative 2 except that the upper end
of the channel is divided and separate control structures are.
provitied ... This division would allow the use of one structur·e or
upstream channe 1 whi 1 e maintenance or remedi a 1 ~1ork is being
performed on the other.
Alternative. 2B is similar to Alternative 2 except that the cascade
spillway is replaced by a ~ouble stilling basin type structure.
This spillway is somewhat longer than the sinli 1 ar type of structure
on the right bank in A 1 tern at ive 1. However, the s 1 ope of the
ground is less than the· rather steep right bank and it may be easier
to construct, a factor which may partly mitigate the cost of the
longer· structure. The discharge is at a sharp angle to the river
and being more concentrated than the cascade could cause erosion of
the opposite bank. ·
• Alternative 2C is a derivative of 28 with a similar arrangement,
except that the double stilling basin spillway is reduced in size
and augmented by an addi ti ona1 emergency spillway in the form of an
inclined~ unlined rock channel~ Under this arrangement the concrete
spillway acts as the main spillway, passing the 1:10,000 year design
flood with greater flows passed down the un 1 i ned channe 1 which is
closed.at its upstream end by an erodable fuse plug. The problems
of erosion of the opposite bank still remain, although these could
be overcome by excavation and/or slope protection. Erosion of the
chute would be extreme for significant flm'ls,. although it is highly
unlikely that this emergency spillway would ever be used.
Alternative 20 replaces the cascade of Alternative. 2 with a lined
chute and flip bucket. The comments relative to the flip bucket are
the same as for Alternative 1 except that the left bank location in
this instance requires a lange~ chute, partly offset by lower
construction costs bee ause of the flatter slope, and the flip bucket
discharges into the river at an angle which may.cause erosion of the
opposite bank. The underg~·ound powerhouse is located on the right
bank, an arrangement which provides an ov~~rall reduction ·of the
length of the water passages.
(iv) Alternative 3
This arrangement h~s a dam axis location slightly upstream from
Alternative 2, but retains the downstream curve at the abutments.
The main spillway is an unlined rock cascade on the left bank which
passes the design flood. Discharges beyond the 1:10,000 year flood
would be discharged through the auxiliary concrete-1 i ned chute and
flip bucket spillway on the right bank. A gated contra 1 structure
is provided for this auxiliary spillway which gives it the flexibi 1-
ity to be used as a backup if maintenance should b·e required on the
main spillway.. Erosion of the cascade may be a prob 1 em, as
mentioned previously, but erosion downstream should be a less
important consideration because of the low unit discharge and the
infrequent · operat 1 on of the spi 11way. The diversion tunne 1 s are
situated in the right abutment~ as with p·revi ous arrangements, and
are of simi 1 ar cost for all these alternatives.
9-36
I
a ~
t)
'
' I
t
' i "'"
·a
< ~
I
I
I
.I
t
I
' .,.--:;,.
'
I
I ;.
W> ,.. .,
'
'
' t
I
' I ..
I
I ......
I
I
I
' t . t
••
(v) Alternative 4
-This alternative involves rotating the axis of the main dam so that
the left abutment is relocated approximate1y 1000 feet downstream
from its Alternative 2 location. The relocation results in a
-reduction in the overall dam quantities but would require siting the
impervious core of the dam directly over the "Fi ngerbuster" shear
zone at maxi mum dam height. The 1 eft bank spillway, consisting of
chute and flip bucket, is reduced in length compared to other left
bank locations, as are the power facility water passages. The
diversion tunnels are situated on the left bank; there is no advan-
tage to a right bank location~ since the tunnels are of similar
length owing to the overall downstream relocation of the dam.
Spillways __ and power facilities would also be lengthened by a right
bank location with this dam configuration.
(c) Selection of Schemes for Further Study
A basic consideration ·during design developm-ent was that the main dam core
should not cross the major shear zones because of the obvious problems ~Jith
treatment of the foundation~ Accordingly, there is very little scope for
realigning the main dam apart from a slight rotation to place it more at
right angles to the river.
Location of the spillway on the right bank results in a shorter distance to
the river anc allows discharges almost parallel to the general direction of
river flow. The double stilling basin arrangement of the preliminary DSR
scheme would be extremely expensive, particularly if it must be designed to
pass the probable maximum flood. An alternative such as 2C wou.ld reduce
the magnitude of design flood to be passed by the spi 11way but would only
be acceptable if an emergency spillway with a high degree of operational
predictability could be constructed. A flip bucket spillway on the right
bank, discharging directly down the river, would appear to be an economic
arrangement~· although some scour might occur in the plunge pool area.. A
cascade spillway on the 1 eft bank caul d be an acceptable so 1 ut:i~11 providing
most of the excavated materia 1 could be used in the dam, and adequate rock
conditions exist.
The length of diversion tunne 1 s can be decreased if they are 1 ocated on the
right bank. In addition, the tunnels would be accessible by a preliminary
access road from the north, which is the most likely route. This location
would also avoid the area of 11 The Fingerbuster" and the steep cli.ffs which
would be encountered on the left side close to the downstream dam toe ..
The underground configuration assumed for the powerhouse in the.se prelimi-
nary studies allows for location on either side of the river with a minimum
of interference with the surface structures.
Four of the preceding layouts., or variations of them, were selected for
further· study:
0
9-37
(i) A variation of the preliminary DSR scheme, but with a single
stilling basin main spillway on the right bank, a rock channe 1 and
fuse plug emergency spillway, a left bank underground powarhouse and
a right bank diversion scheme;
(ii) Alternative 2 with a right bank flip bucket spillway, an underground
powerhouse on the left bank, and right bank diversion; ·
(iii) A variation of Alternative 2 with a reduced capacity main spillway
and a right bank rock channel with fuse plug serving as an emergency
spillway; and
(iv) Alternative 4 with a :left bank rock cascade spillway, a right bank
underground powerhouse, and a right bank diversion.
9.14 -Intermediate Review
For the intermediate review process, the four schemes selected as a result of
the preliminary review were examined in more detail and modified. A description
of each of the schemes is given below and shown on Plates 14 through 19. The
general locatio~s of the upstream and downstream shear zones shown on these
plates are approximate, and have been refined onthe basts· of subsequent field
investigations for the design_studies described in Section 12.
(a) Description of Alternative Schemes
The four schemes at"e snown on Plates 14 through 19:
(i) Scheme WPl (Plates 14 and 15)
This scheme is a refinement of Alternative 1. The upstream slope of
the dam is flattened from 2.5:1 to 2.75:1. This conservative
approach was adopted to provide an assessment of the possible
impacts on project layout of conceivable meansures which prove
necessary in dealing with severe earthquake design conditions.
Uncerta1nty with regard to the nature of river alluvium also led to
the location of the cofferdams outside the limits of the main dam
embankment. As a result of these conditio~s, the intake portals of
the diversion tunnels on the right bank are also moved upstream from
"The Fins". A chute spillway with a flip bucket is located on the
right bank together with the underground powerhouse.
(ii) Scheme WP2 (Plates 16 and 17)
This scheme is derived from the DSR layout. The main dam and diver-
sion facilities are similar to Scheme WPl except that the downstream
cofferdam is relocated further downstream from the spillway outlet,
and the diversion tunnels are correspondingly extended. The main
spillway is located on the right bank, but the two stilling basins
of the preliminary DSR scheme are combined into a single stilling
basin at the river 1 evel. .An emergency spillway is a 1 so located on
the right bank, and consists of a. channe 1 excavated in rock, dis-
charging downstream from the area of the relict channel. The
9-38 . )
I
' ,,
I
' I
t
'
' 'I ....
I
:t
a·
I
'
' ,,
' ,,
I
'
'
' I
t
I
I
' t
tl\ -·
'
'
' I
I
t
' I
(b)
channel is closed at its upstream end by a compacted earthfi fl fuse
plug and is capable of discharging the flow differential between the
probable maximum flood and the 1:10,000-year design flood of the
main spillway 8 The underground powerhouse is 1 oc ated on the left
bank.
(iii) Scheme WP3
This scheme is similar to Scheme WPl in all respects, except that an
emergency spillway is added, consisting of right bank rock "channe 1
and fuse plug (see Plate 16).
{iv) Scheme WP4 (Plates 18 and 19)
The dam location and geometry for Schemo WP4 are similar to that for
the other schemes. The diversion is on the right bank and
discharges downstream from the powerhouse tailrace outlet. A rock
cascade spi 11way is 1 oc ated on the 1 eft bank and is served by two
separate contra 1 structures with downstream sti 11 i ng basins. The
underground powerhouse is located on the right bank.
Comparison of Schemes
The main dam is in the same location and has the same configuration for
each of the four layouts considered. The cofferdams have been locat!=d
outside the limits of the main dam in order to allow more extensive
excavation of the alluvial material and to ensure a sound rock foundation
beneath the complete area of the dam. The overall design of the dam is
conservative, and it was recognized during the evaluation that savings in
both fill and excavation costs can probably be made after more detailed
study. ·
The diversion t unnE! 1 s are l oc at ed on the right bank. The upstream flatten-
i ng of the dam slope necessitates the location of the diversion inlets up-
stream from "The Fins" shear zone which wi 11 require extensi v·e excava);i on
and support where the tunnels pass through this extrem1.~ ly poor rock zone
and could cause delays in the construction schedule.
A low-lying area exists on the right bank above the area of the relict
channel, and t~is is closed by an approximately 50-foot high saddle dam. A
slurry trench cutoff will be combined with grouting to seal the .200-f,oot
depth of pervious material infilling this channel.
A summary of capita 1 cost estimates for the four alternative schemes is
given in Table 9.7.
The results of this intermediate--analysis indicate that the chute spillway
with flip bucket of Scheme WPl is the least costly spillway alternative.
9-39
(c)
The scheme has the additional advantage of relative1y simple operating
characteristics. The control structure has provision for surcharging to
pass the design flood. The probable maximum flood can be passed by
additional surcharging up to the crest level of the dam. In Scheme WP3 a
similar spillway is provided, except tha the control structure is reduced
in size and discharges above the routed design flood are passed through the
rock channel emergency spillway. The arrangement in Scheme WPl does not
provide a backup faci] i ty to the main spillway, so that if repairs caused
by excessive plunge pool erosion or damage to the structm·e itself require
remova 1 of the sp,i llway from service for any 1 ength of time, no alternative
discharge facility would be available. The additional spillway of Seheme
WP3 would permit emergency discharge if it were absolutely required under
extreme circumstances. ·
The stilling basin spillway (Scheme WP2) would reduce the potential for ex-
tensive erosion downstream, but high velocities in the lower part of the
chute could cause cavitation even with the provision for aeration of the
discharge. This type of spillway wou'l d be very costly, as can be seen from
Table 9.7.
The feasibility of the,rock cascade spillway is entirely dependent on the
qua 1 i ty of the rock, which dictates the amount of treatment required for
the rock surface and cdso the proportion of the excavated material which
can be used in the d,:tm. For determining the capital cost of Scheme \4P4,
conservative assumr:cions were made regarding surface treatment and the
portion of materi a1 that woL·ld have to be wasted.
The diversion tunnels are located on the right bank for all alternatives
ex ami ned in the intermediate review. For Scheme WP2, the downstream
portals must be located downstream from the stilling oasin, resulting in an
increase of approximately 800 feet in the length of the tunnels. The left
bank location of the powerhouse requires its placement close to a suspected
shear zone, with the tailrace tunnels-passing through this shear zone to
reach the river. A 1 anger access tunne 1 is a 1 so requ·i red, together with an
additional 1,000 feet in the length of the tailrace. The left-side
location is remote from the main access road, which will probably be on the.
north side of the river, as will the transmission corridor.
Selection of Schemes for Further Study
Examination of the technical and economic aspects of Scheme WPl through WP4
indicates there is 1 itt l e scope for adjustment of the dam axis owing to the
confinement imposed by the upstream and downstream shear zones. In
addition, passage of the diversion tunnels through the upstream shear zone
could result in significant delays in construction and additional cost.
From a comparison of costs in Table 9.7, it can ba seen that the flip
bucket type spillway is the most economi ca 1, but because of the potential
for erosion under extensive operation it is undesirable to use it as the
on·ly discharge facility. A mid-level release will be required for emer---
gency drawdown of the reservoir, and use of this release as the first-stage
service spillway with the flip bucket as a backup facility would combine
flexibility and safety of oper~tion with reasonable cost. The emergency
rock channel spillway would be retained for di schiirge of flows above the
route 1:10,000 year flood.
9-40
I
'
'
'
'
'
! '
' I
'
I •
I .,
i
I
I
t
I
t
I
I
I
' I •
, .•
I
' t
'
'
' I
t
I
' I
' I
t
The stilling basin spi1lway is very costly and the operating head of 800
feet is beyond precedent experi.ence. Erosion downstrceam should not be a
pt"Oblem but cavitation of the chute could occur. Scheme \vP2 was therefote
elimi·nated from further ·consideration ..
The cascade spillway was also not favored for technical and economic
reasons. However, this arrangement does have an advantage in that it pro-
vides a ·means. of preventing nitrogen supersaturation in the downstream
discharges from the project which could be harmful to the fish population,
as discussed in Section 9.10. A cascade configuration would reduce the
di sso 1 ved nitrogen content ... and hence~ this a 1 ternati ve was ~et ai ned for
further evaluation. The c~tJacity of the cascade was r;educed and the
emergency rock channe 1 spi'llway was included to take the extreme floods.
The results of the intermediate review indicated that the following compon-
ents should be incorpo~"ated into any scheme carried forward for f~nal re-
view:
... TvJo diversion tunnels located on the r·ight bank of the r~iver;
·· An underground powerhouse also 1 oc ated on the right bank;
An emergency spillway~ comprising a rock channel excavated on the right
bank and discharging well downstream from the right abutment. The
channel is sea1ed by an erodible fuse plug of impervious material
designed to fail i.f overtopped by_the reservoir-; and
- A compacted earthfi 11 and rockfi 11 dam situated b-etween the two major
shear zones which tiaverse the project site.
As discussed above, two specific alternative methods exist with respect to
routing of the spillway design flood and mi nimi zing the adverse effects of
nitrogen supersaturation on the downstream fish populatior~. These alterna-
tives are:
- A chute spillway with flip bucket on the right bank to pass the spi 1lway
design flood, with a mid-level release system designed to operate for
floods with a frequency of up to about 1:50 years; or
- A cascade spillway on the 1 eft bank.
Accordingly, two schemes were developed for fruther evaluation as part of
the fi na i review process. These schemes are described separate 1 y in the
paragraphs be 1 ow.
~e15 -Final Review
The two schemes considered in the final review process were essentially
deviations of Schemes WP3 and WP4.
(a) Scheme WP 3 A (Plate 20)
This scheme is a modified version of Scheme WP 1 described in Section 9.14
with an emergency spillway as included in Scheme WP 3. Because of
9~41
schedu 1 i ng and cost considerations, 1 t is extreme 1 y import ant to rnaint ai n
the diversion tunnels downstream from "The Fins .. n It ,is also important to
keep the dam axis as far upstream as possible to avoid congestion of the
downstr,eam structures. Far these reasons, _the in 1 et portals to the
diversion tunn~ls were located in the sound bedrock forming the downstream
boundar,y of the '~Fingerbuster. u The upstream cofferdam and main dam are
maintained in the upstr-eam locations as shown on Plate 14. As mentioned
previously, additional criteria. have necessitated modifications in the
spillway configuratton, and low-level and emergency drawdown outlets have
been introduced.
The main modifications to the scheme are ~s follows:
( i) Main Dam
Further investigation of preliminary design studies and review of
world practice sugg.ests that an upstream slope of 2.4H :IV-would be
acceptable for the rock shell. Adoption of this slope results not
only in a reduction in dam fill volume but also in a reduction in
the base width of the dam which permits the main project components
to be located between the major shear zones.
The downstream slope of the dam is retained .as 2H:1V. The coffer-
dams remain outside the limits of the dam in order to allow complete
excavation of the riverbed alluvium.
(i i) Diversion
In the intermediate revlew arrangements, diversion tunnels passed
through the broad structure of 11 The Fins,11 an intensely sheared area
of breccia, gouge, and i nfill s. Tunne 1 i ng of this materia 1 wou 1 d be
difficult, and might even require excavation in open cut from the
surface. High cost would be involved, but more important would be
the time taken for construction in this area and the possibility of
unexpected delays. For this reason, the inlet portals have been
relocated downstream from this zone with the tunnels located closer
to the river and crossing the. main system of joi nt.i ng at approxi ...
mately 45°. This arrangement allows for shorter tunnels with a more
favorable orientation of the in let and out 1 et porta 1 s with respect
to the directions of river flow and di'version inflow and outflow at
the portals.
A separate low-1 eve 1 in 1 et and concrete-1 i ned tunnel is provided
leading from the reservoir at approximate elevation 1550 feet to
downstream from the diversion plug where it merges with the
.diversion tunnel closest to the river. This low-level tunnel is
designed to pi::\S flows in excess of 2000 cfs acting as a low-level
release durin~ reservoir filling. It will also pass up to 10~000
cfs under 500-foot head to allow emergency draining of the reservoir
as discussed in Section 9 .. 9.
9-42
I
I
' I
I
I
I
I
I
I
t
I
t
t
-
t
t
~
t
I
' .,
I
t
I
I
'
' I
I ,.
' I .,
I
t
I
I
(iii)
{iv)
Initial closure is made by lowering the gates to the tunnel located
c 1 osest to the river and constructing a concr:ete c 1 a sure p 1 ug in the
tunnel at the location of the grout curtain Ulnderlying the core of
the main dam. On completion of the plug~ thEi 1ow-leve1 release is
opened and controlled d1 scharges are passed downstream. The gates
within the second port a 1 are lowered and a mass concrete c 1 osure
p 1 ug constructed on a 1 i ne with the grout cur·tai n. Afi>-=r c 1 osure of
the gates, filling of the reservoir can commence.
Emergency and Out let Release Faci 11~;; es
As a pro vision for drawing down the reservoir in case of emergency,
a mid-level release is provtded~ The intake to these facilities is
located at deptch adjacent to the power facilities intake
structures. Flows wi 11 then be passed downstr·eam thr"ough a hi red
shaft and tunne 1, existing beneath the downstream end of the main
spi 11 way flip bucket, as ·described be 1 ow. In order to overcome
potential nitrogen supersaturation problems, Scheme WP 3 A also
incorporates a system of fixed core valves into the downstream end
of the emergency release facility. The valves were sized to
discharge in conjunction with the powerhouse operating at 7S percent
capacity, flows up to the equivalent routed 50-year flood. Six
valves are required, located on branches off a steel manifold and
protected by i ndi vi dua 1 upstream c 1 osure gates. The v a 1 ves are
partly incorporated into the mass concrete block forming the flip
bucket of the main spillway. The rock downstream is protected from
erosion by a concrete facing slab anchored back to the sound
bedrock.
Spillways
As discussed in Section 9.10 above, the designed operation of the
main spillway facilities was arranged to limit discharges of
potentially nitrogen supersaturated water from Watana to flows
having an equivalent return period of not less than 1:50 years ..
The main chute spillway and flip bucket discharge into an excavated
plunge pool in the downstream river bed. Releases are controlled by
a three-gated agee structure 1 oc ated adjacent to the out 1 et release
and power intake structures just upstream from the dam centerline.
The design discharge is approximately 80,000 cfs corresponding to
the routed 1:10,000-year flood (120,000 cfs) reduced by the 40,000
cfs flet11.s attributable to outlet release and power faciliti.es
discharges. The plunge pool is formed by excavating the alluvial
river deposits to bedrock; and as this approaches the limits of the
calculated maximum scour hole, it is not anticipated that~ given the
infrequent discharges, significant downstream erosion wi'll occur.
The emergency spillway is provided by means of a channel excavated
in rock on th~ right bank~ discharging \'/ell downstream from the
right abutment in the direction of Tsusena Creek. The channe 1 is
sealed by an erodible fuse plug of impervious material designed to
9-43
fail if overtopped by the reservoir; although some preliminary
excavation may be. necessary. The crest level of the plug wi 11 be
set at elevation 2230 fe~t, well below that of the main dam. Th~
channel will be capable of passing the exce·ss discharg~ of floods
greater than the l:IO,OOQ .... year flood up to the probable maximum
.flood·of 235,000 cfs. ·
(v) Power Facflities
The power intake is set slightly upstream from the dam ·centerline.
deep within·sound bedrock. at the downstream end of the approach
channeL. The intake consists of six units with provi sian in each
unit for drawing flows from a. variety of depths covering the
complete drawdown range Of the reservoiro This facility also .
provides for drawing water from the different temperature strata
within ·the upper part of t.he reservoir~ and thus regulating the
temperature of the downstream di scharg~s close to the natural
temperatures of the river. For this p:~el imi nary conceptual
arrangement, flow withdrawals from difterent levels is affected by a
series of upstream verti.ca1 shutters moving in a single set of
guides and operated to form openings at the r~qui red 1 eve 1.
Downstream from these·shutters each unit has a pa.ir of wheel-mounted
closure gates which will i-solate the individual penstocks.
The six penstocks are 18-foot-di ameter, concrete-lined tunnels
inclined at 55° immediately downstream from the intake to a nearby
.horizontal portion leading to the powerhouse. This horizontal por-
tion is steel-lined for 150 feet upstream from the turbine units to
extend the seepage path to the powerhouse and contain the flow with-
in the fractured rock area caused by b 1 asti ng in the adjacent power-
house cavern.
The six 170 MW turbine/generator units are housed within the major
powerhouse cavern and are setviced by an overhead crane which runs
the 1 ength of the· powerhouse and into the service area adjacent to
the units. Switchgear, area maintenance room and offices are 1ocat~
ed within the main cavern, with the transformers situated downstream
in a separate gallery excavated above the tailrace tunnels. Six
inclined tunnels carry the connecting bus ducts from the main power
hall to the transformer gallery upstream. A vertical elevator and
vent shaft run from the power cavern to the main office. bui 1d1ng and
control room located at the surface.. Vertical cable shafts., one for
each pair of transformers, connect the transformer gallery to the
switchyard directly overhead. Downstream from the transformer
· ga.llery$ the unde~lying draft tube tunnels merge fnto two sur.ge
chambers, one chamber for three draft tubes, which also house th'e
draft tube gates for isolating the units from the tailr.ace .. The
gates are operated by an overhead tt"aveling gantry located in the---
upper part of each of the surge chambers. Emerging· from the ends of
the chambers, two concrete-lined, low-pressure tailra<:e tunne.ls
carry the discharges to the river. Because of space restrictions at
the river, one of these tunnels has been merged with the downstream
end of th.e diversion tunnel. The other tunnel emerges in a separate
portal with prQvi si on for the i nsta llati on of bulkhead ga.tes.
I~ ..
' : i
'
' I
I
I
I
I
I
I
I
.I t.
t
I
I
I
i
t
·~
'I
I
t
I
I
I
I
I
I.
t
' I
' I
I
I
I
I
t
< .-:-
(b)
The qrientation of water passages and underground caverns ts such as
to avoid as far as possible the ma.i n, alignment of the excavations
running paralle-l to _the major Joint sets as described in Section
9.3~ .
(vi) Access
Access is assumed to be from the north (right) side of the river~
Permanent access to_ structures close to the river is by a -road along
the right downstream river ba.nk and then· vi'a'··-a tunnel passing
through the concrete formin~ the flip bucket. A· tunnel fr6m this
point to the power cavern provides for vehicular access.. A
secondary access road across the crest of the dam passes dawn the
1 eft b an_k of the v a11 ey and across the 1 ower part of the dam.
Scheme WP 4A (Plate 21)
This scheme is similar in most respects to Scheme WP 3A previously dis~
cussed~ except for the spillway arrangements.
(i) Main Dam
The main dam axis is similar to that of Scheme -wP 3A, except for a
s 1 i ght downstream rotation at the 1 eft abutment at the ·spillway con-
trol structures.
(ii) Diversion
The diversion and l0\'1 head releases are exactly similar for the two
schemes.
(iii) Emergency and Outlet ~eleas~. Facilities
The emergency drawdown re 1 ease faci 1 i ty is separated from the main
sp'i llway for this scheme. The emerging re 1 ase consists of a
low-level gated outlet structure located upstream, discharging up to
30,000 cfs well into the river through a concrete-lined~ free-flow
tunnel with a ski jump flip bucket. This facility may also be·
operated as an auxiliary outlet to augment the main left bank
spi 1lway.
( i v) S pi 11 ways
The main left bank spillway is capable of passing a design flow
equivalent to the 1:10,000-year flood through a series of 50-foot
drops into sha 11 ow pre-excavated p 1 unge poo 1 s.. The emergency
spillway is designed to operate during floods of greater magnitude
up to and including the PMF.
Main spillway discharges are coQtrolled by a broad multi-gated
control structure discharging into a shallow stilling basin. The
feasibility of this arrangement is governed by the qua·lity of the
r·ock i.n the area.~ requiring both durabi l).ty to withstand erosion
caused by spillway flows, and a high percentage of sound rockfill
material that can be used f~om the ex{:avation directly in the rna1n
dam.
9-4b
'.)
I
I
I' I
I
l
{c)
!.;
On the: basis -Of the .sit·e information deve.foped concurrent Ty-with the
gen~ral arr·angement studies, it became app.arent that the major shear
zone::_ known to exist in the left bank area extended ·further -
dO\·Jnsl're.am than initial st·udies have indicated. The cascade
spillway channe 1 was therer:ore 1 engthened to avoid the shear area at
the lower end of the cascade. The arrangement shown on Plate 21 for
Scheme WP 4A does not reflect this relocati'on, which would increase
the overall cost of the scheme.
The_ emergency spillway consisting of rock channe 1 and fuse plug is
simi 1 ar to that o.f the right bank spillway scheme.
(v} Power Facilities
The power facilities are similar to those in Scheme WP3A.
--
Evaluation of Final Alternative Schemes
An evaluation of the dissimilar features for each arrangement (the main
spillways and the discharge. arrangements at the downstream end of the
outlets) indicates a saving in capital cost of $197,000,000, excluding
conti ngenci_es and indirect cost, in favor of Scheme WP 3A. · If this
difference is adjusted for the savings associ atad with using an appropriate
proportion of sxcavated material as rockfi 11 in the main dam, this repre-
sents a net overall cost difference of approximately $110,000,000 including
contingencies, engineering, and administration costs.
As discussed above, although limited information exists regarding tne qual-
ity of the rock in the downstream area of the left bank, it is known that a
major shear .zone runs through and is adjacent to the area presently
allocated to the spillway. This waul d require re locating the· left bank
cascade spillway several hundred feet farther downstream into an area where
the rock quality is unknown and the topography less suited to the gentle
overall slope of the cascade. The cost of the excavation would substan-
tially increase compared to previ ou.s ass\lmpt 1 ens, irrespective of the rock
quality. -In addition, the resistance of the rock to erosion and the suit-
ability for use as excavated material in the main dam would become less
certain. The economic feasibility of this scheme is largely predicated on
this 1 ast factor, si nee the abi 1 i ty to use the materi a1 as a source .of
rockfi 11 for the main dam represents a major cost saving.
The prabl em of the occurrence of nitrogen supersaturation can be overcome
by t,he use of a regularly operated di-spersion type valve outlet facility
in conjunction with the main chute spillway. As this scheme presents a
more econcmic solution with fewer potential problems concerning the·
geotechnical aspects of its design, the right bank chllt-~ ~;'~·\angement has
been adopted as the final selected scheme.
9-46
. ·~
-~
~
1
t
t
I
I
I
t
I
t
I
I
I
I
I
I
·-,,_.--
; ·•
., . . . /t\~ .... ~,J.J;.u~::., .... ~""-"""'""',u;MI;~
·.I
... ,
I
1
I
I
I
I
I
I
I
I
I
' I ,,
I
I
I
TABLE 9.3: COMBINED WATANA AND OCVIL CANYON OPERATION
Watana Dam Watana* Devil. Canyon* iota! Average
Co.st Crest Elevation Cost Cost Annual f;nergy
(ft MSL}
2240 (2215
reservoir elevation}
2190 (2165
reservoir elevation)
2140 (2115
reservoir elevation)
($ X 106 ) ($ X 106) ($ X 10 6) -
4,076 .1' 711 5;787
3.,785 1,711 5,496
3,516 . 1' 711 5,227
Watana Project alone (prior to year 20D2)
Crest Elevation
(ft MSl)
2,240
2,190
2,140
Average Annual
Energy (GWh)
3,542
3 1 322
3,071
* Estimated costs in January 1°82 dollars, based on preliminary conceptual
designs, including relict channel drainag·e blanket a11d .20 per.cent
contingencies ..
(GWh)
6,809
6,586
6,264
~I
I
I
...
. "
I:
I
, I I'
TABLE 9.4~ LONG-TERM PRESENT WORTH I
W~tana Dam Long Term I Crest Elevat.ion System-Presen~* '
(ft MSL) Worth ($ x 10 ) -
~ 2240 (reservoir
elevation 2215) 7' 123 I 2190 (reservoir
elevation 2165) 7,052
2140 (:r-eservoir I elevation 2115) 7,084
* January 1982 dollars. I
I
I
I
I
I
I
I
I
I
I
"' :~~; I c:!:
:·::·\', '>
'"-
·-..
0 .".J ' ;
I
'" I
I
I
I
.I
I
I
I
I
I .:.
I
; -.
I
I
I .,
' ':,."; ' t;:-.
,.,. ,
.. -··--. -... ·~~~~.,...,..--,---._.,.,.n·"'"· ......... ......., ·-
TABLE 9.5: OES!'GN DATA AND DESIGN CRITERIA .FOR
fiNAL REVIEW OF LAYOUTS
Rivel' flaws ·-
Average flow (over 30 years of recotd);
Prabable maximum flood. (routed):
Maximum inflow with return period of 1.:10,000 years:
Maxinwn 1:lo·,ooO-year routed discharge:
Maximum flood with return period ·Of 1:500 years:
Ma~imum flood with return period of 1:50 years:
Reservoir normal maximum operating level:
Reservoir minimum operating level;
Dam -
Type:
Crest elevation at point of maximum super elevation:
Height:
Cutoff and foundation treatment:
~stream slope:
Downstream slope:
Crest width:
Diversion
. Cofferdam type:
Cutoff and foundation:.
Upstream cofferdam crest elevation:
Da'ftnstream·cofferdam crest elevation:
Maximum pool level during construction:
Tunnels ·
final closure:
Releases during impounding:
Spil~~
Design floods:
Main spillway -Capacity:
-Control str,Jcture:
Emergency spillway ~ Capacity:
... Type:
Power Intake . .
Type·:
Nt.Jllber of intakes:
Draw-off requirements:
Drawdown:
-~· ~ -: .. '. .. ..
:::::::~
7,860 r.:fs
235,000 cfs
155,000 cfs
120,000 cfs ·
116,000 cfs
87 1 000 .cfs
2,215 ft
2,030 ft
Rock fill
2,240 ft
890 ft above foundation
Core founded em rock; grout curtain and
downstream drains
1V:2.4H:1V
1V:2H:1V
50 ft
Rock fill
Slurry trench to bedrock
1,585 ft
1,475 ft
1,580 ft
Concrete lined,
Mass concrete plugs
6,000 cfs mirimum via bypass to ou.tl.et
structure
Pass~s PMF, preserving.integrity of dam
with no loss of life
Passes routed 1:10,000.;.year f'lood with no
damage t;.o struntures
Routed 1 :10,000-year flood
with 5 ft surcharge
Gated ogee crests
PMF minus 1:10,000 year flood
Fuse..,plug
Reinforced concrete
6
Multi-level corr.esponding to temperat:ure
strata
1135 feet
. -. ~
'c
··--···A<·•·"·:;L_.li -:t.. _...:~
''-
... -F-~ .. ~~:.. ~ .L·~-!~·,2_,'.' "'-·~~~~~~
\
\\
TABLE 9.5t {Cont'd)
Type:
Nunber of \)enstocks:
Powerhouse
Type:
Transformer area:
Control.room and administration:
Access -Vehicle:
.. Personna 1:
Powerplant
Type of turbines:
Nllllber of rating:
Rated net head: ·
Design flow:
Normal maximum gross head:
Type of generator:
·Rated output:
Po.wer factor:
Frequency:
Transformers:
Water passages:
Surge:
. Average tailwatex· elevation (full generation):.
Concrete-lined tunnels with downstre~am
steel liners
6
Un~ergr.ound
Separate gallery
Surface
Rock tunnel
Elevator from surface
Francis ""'
6 X 170 MW
690ft
31500 C 1 ~S per unit
745 ft
Vertical s~nchronous
148 MVPi
0.9
60 HZ
148 MVA -13.8-345 kV, 3-phase
2 concret.e-lined tunnels
Separate surge chambers
1,458 ft
' ••
I
I
I
I
I
I
I
I
:
I
I
I
I
I
I
' l
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
PRELIMINARY REVIEW
Technical feasibility
Co~atibility of layout
with known .geological
and topographical ~·ite
features
Ease of constructi~~r·;
Physical dimensio:--.;J
of component. structures
in certain loc~tions
Obvious cost dl:t' ferences
of comparable structures
Environmental accept-
ability
Operating characteristics
TABLE 9,6: tVALUATJON CRITIEFtt:
INTERMEDIATE REVIEW
Technical Feasibility
Co!JfJatibilit 't of layout
with known geological and
topographical site featur~s
Ease of construction
Overall cost
Environmental accept-
ability
Operating characteristics
Impact on constructicn
schedule
~1
fiNAL REVIEW
Technical feasibility
Compatibility of layout
with known geological and
topographical site features
Ease of construction
Overall co&t
Environmental impact
Mode of operation of spill-
'Nays
Impact on construction
schedule
,Jesign and operating limita-
tions for key structures
.-~-,-.--,-.'"-0'_____ -o--~-----;--.--.----------=----·~~-e-cc--e
TABLE 9. 7: SUMMARY OF COMPARATIVE COST ESTir;~;:;:e
lNTERtt;EDIATE REVIEW OF ALT~RNATIVE ARRANGEMENTS
(January 1982 $000,000)
WP1 WPZ WPJ
Diversion 101.4 112.6 "101.4
Service Spillway· 128.2 208.3 122.4
Emergency Spillway -46.9 46.9
Tailrace Tunnel 13.1 13.1 13.1
Credit for Use of Rock in Dam (11 .. 7) (31. 2) ( 18 •. 8}
Total Non-Common Items 231.tl 349.7 265.0
Common Items 1643.0 1643.0 1643 .. 0
Subtotal 1874.0 1992.7 1908.0
Lamp & Support Costs· (16%) 299oB 318 .. 8 305.3
Subtotal 2173.8 2311.5 . 2213.3
Contingency (20%) 434.8 462.3 442.7
Subtotal 2608.6 1773.8 2656.0
Engineering and
(12.5%) Administration 326.1 346.7 332.0
'
lOTAL 2934.7 3120.5 29138.0
WP4
1(..3.1
~67.2
a~o
(72.4)
305.9
1643.0
194.8.9
311 .. 8
2260.7
452.1
2712.8
339 .. 1
3051.9
. '\1
I
I
I
I
I
I
I
I
I
1 ..
I
I
I
I
I
I
I
I
t·'
I
I
I
I
I
I
I
I
I
I
I
I
·I
I
I;
I.
·I
l
I!!MMQ
•••rm
_.,.m?
I
l
I
\·
··.~
0 • -..£;$
SCALE e~5iiiiiiiiiiiiiiiiiiii!
LEGEND-~
BOREHOLES AND TEST Pr!'::
. f971J,COE ROTARY DRILL ~ORI~G .....---.......
GEOPHYSICAL SURVEYS~ .· ·. ·~
SEISMIC AEFR~CTION SURVEy £WO Ofl. Tl.lftlnNG
57~. DAMES & MOORE
.~978, SHANNON 8 Wl!.!iON
1980-111, WOODWARD•CLYDE C~SULTANTS
0
SCALE
,.
I
I'
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I ·~
It~-
I
!
..---_....;....J/
+~ ---------
:
WATANA
EXPLORATION MAP
LEGEND
BOREHOLES AND TE$T PITS•
Gl OM ... ·l t·t·n·· .. ,t· of..·.· J·. OIA~ONO CQRE aoRtN&,"OR!Zot{f.L . ~ · ~£CTICHtAt~N
..... ~AAJ
.. TPIHt .-,,AA! IA~H<lE TEST PIT
GEOPtiVSICAL $UR'IEYS: '
£ ... lltiiMIC MniACTtot: SUIIIVE'1 EMil
Oft TUMfiMI POINT
. OM·C ttta,oAYts a MOoRE
s-.1 lt78,1MiAHMON a WIL~
I
SL 10•1 tHO~II,W®DW~·CLYDE c::oMim.TANTS
o zoo 400 FEET
SCM.£
REFERENCE• BASE . .lt4AP FROM COE.l918-1••200'
WAUNA TOPOGRAPH'I',SHE£T8al3
OF 26,COORDitfATES IN FEET,Al.ASI<A
STATE PLANE (ZoNE4}
.CONTINUES OFF
MGE
COE 1 lV78( }
s aW.tt:?a ( )
wcc,l978 < .. )
~· . ~ ·-
FIGURE 9 .. '2·0
~·
.
iil
.. ,,
I
I
I
I
I
I
I
I
·I·
I
·I
I
I
I
I
I
I
_ ____:l __ _
··~-
I +
l
!
Ill
~···~···
.-··'
\
-----------
.
·~
----::;J _.:t•---#-·'<"'
WATANA
ROCK OUTCROP MAP
LEGEND
LITHOU>3Y:
.. "~"•,..f ANDESITE .PORPkYRY,INct..UOESMINoft.
• • ,. DACITE AND LATtTE
CONTACTS:
----... LIMIT OF OUTCROP
CONltlUR LINES:
------.... TOPOGRAPHIC CONTOVR ~TERVAL
50 FEET
..... --~--.........----.. ~ ..
+
+
0 200 400 FEET
SCALE
+
' . .
i . FIGURE 9 .. 3
-~-_,-.. --.. :0----~---,---
."Y"
I
I
··~-
I
I
li
I
I,
I
I
I
I
I
I
I,
I
I
I
I .. a
lt I,· i
i
.. · .. --
·WATANA
GEOLOGIC MAP
SCALE
LEGEND
LttHOU>GYt D :=~~..a~ltlE:$
•to.••;.. . ~ ~.iNCt.Uo£:$ .~ •• • liiiNOit DAC11£ JND Ul11'1E
j~,.v ~~ OIOftlttllfOft~
CONTACTS: .
J'!!. LI'I'HOLbGIC, OUtED WHERE WEfii'E).,
0Ht WH£RE. tQtOWN
STRUCTUfi£:
Em• .SHEAR. WIOTH. ~ l'tWC• to~ ~ ~ lMl£SS w .SHOWfil.
TQ..__ ~.·~ LESS nwe.··l9 n:tt •. · ..)"""A" INCUNEO, VERTICAL.£Xmll ~.
·.JOtDWN
e;;;:.:;;::t ~ z• WIDTH ~"r14111il ~ 10 f'Ef'f; veRTICAL UNI..DS .,. .. SfiOWM:
To-·::..--. ~-!~WilTH.LES$ TRoUt .. • T.r 10 FEET ~u ~ .£XTENT •• ,. WHERE ·~~WM. '
'i'~h ~ ·~~~~~·.rJr!li~~· I'JO EXCEPT FCR OPEN .. JQtNTS.)
[:J A!.TERKOON ZONE, 'WI>nt AS s .... :
OTMER!
w··a t ,t GEOLOGIC SEC'nON LOCAll(lM
A WJ•I JOINT STATtotli
0 200 .. oo fEET
FIGURE. 9 •. 4
--.·· '·
:t
1-
I
I
1:
I.
I
I
I
I
I
I
I
I
I
'I
I
I
!I
1500
1000
TAILRACE
TUNNEL
~ACCESS ~
0
TUNNELS
':r.·I'OO
toO ao o
I J J £l..
-1710.1·.· ·. . . -Tit.· ·. ~0.~ ... 0 ~. :~_/ #'
. ~,-.
-15M • . .,.'I>
0--_0
JDCVERSION
TUNNELS
~·
AREAS OF'FitACTURE
1DNES a MINOR SHEARS
(HIGH ANCILE DIPS)
TRENO~IO·
·WATANA
GEOLOGIC SECTION w-1
SHEET I OF 2
. OH-s 4
~-·
I
9USITNA
..
I
I
t
I
L!GEND
UTHOLOG'f;
gr.~J.i: ~.1.......-FP!OfTIATto
r---1·. •OIOitr!E. TO. GUMTZ. t:dlftt. INa 'IDES 1......-..J . ...oR~ .
n,-.~ ::-.-: .. '""· · ANDDitt POR;BIMrt,INCUJIIf:IIIINO!t
. .• • •. • ••. QriQ1'E: • Ul'J"JE
lq"\.. ..,,I*MtTE ~
CONTACTS~
----~· "rtP -~ RO(;I(
--.~ •. I:IQD 'WMJ;RE· IHFEJ.:D
-:.c,~.·Sitowrc w.at:· tiUTf1t
~~ FP.ACTIJR£. . ~ONE.l..Wimt .SI40WN ·~ ~ GIIIEATIR TtWf. ·.u FEET . . c:: ·] ALT£JW'ION lilt£, \IIIDTH M .ltfOWN·
GEOPtftSICAL SURVEYS:
tsw·• ~CliQM WITH SE1SMlC ~
OM•C Jt75, ~ a _,OOM
SW-I 1978, SHANNON a Wl.SON
Sl. 80-2 1910, WOOOWUIO·CUO£ ~SUIJ»CT$
SL 1!·21 1981 • WOOOMRD·CUDE CONSl.UJWTS
••••••oo• SEISMC VELOC!TY ·~
1 ~ SEISIIIC. VELOCITY .. FEET POl SECQN)
BOREHOLES :
.Ui'HOI.OGY
8H·I ·
!&SHEAR
All'£RIITION ZONE
~:f . COE NOTARY a I>IAMCM!lO COR£ 80Mt6S
D fJH AAI ~COM . .aft~Ne
OTHER•
W-5. .I INTERSEC'n~ WITH GEOl..OGIC; '+' SECilON 'W • l5
~;-:-.. GEOL061.C f't.Al"lJM OESCfttatO-' ~~ .
SCALE
FIGURE 9JI
..
I
=~I
I
I
I
I
I
I
I
I
I.
·I
I
I
2800
2000
11500
1000
0
8aTTOM
PRQ.I£CTEO
IIS'W
BOTTOM
PRO.IECTa)
II!'E
WATANA
GEOLOGIC SECTION W-1
SHEET 20F2
1000
o. 100 200 FEET
SCALE
FIGURE 9Jta
,.
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
,I
..1
E
II
COMPOSITE JOINT· PLOT
SOUTHEAST QUADRANT
N•721
II
COMPOSITE JOINT PLOT
NORTHEAST QUADRANT·
N•!525
w
w
E
E
II
COMPOSITE JOINT PLOT
SOUTHVIEST QUAt)RANT
N•329
N
COMPOSITE "OlNT PLOT
NORTHWEST QUADRANT
WATANA
COMPOSITE JOINT PLOTS
SCALE
•
0 '--·
NOTES
l. coto!TOURS ARE PEJK:EHT OF JOINTS PER 1"-OF AREA.
CONTOUR INT£R\iU;.-I ,3, a 0 "'·
Z. N EOOAU NUMBEII OF ~TA POINTS.
~. COMPOSITE .PLOTS .INCORPO'MTE ALL JOINT ~~ iFflOM
THEIR ft£SP£CTIVE WAOftlNTS.
.f • .IOINT .t'LCtra FOtt .1011T STAfiO~l (Wol-1,2,3,._5.6.~1 all)
ott
5, FOft .1010'. PLat:TM liiET'HOD .$EE FIGURE
1
FIGURE 9;12 [iil
;1. · ..
' .
I
••
'.1 . . .
-.,
I
I
I
·-
1
I
I
I
I
I
I
I
I
I
··--
··--
'''tm
JI'WQQQ
··--
"' . .:. ..
. ·•
..
I SC.AL£ o 4 8 MILES ~!!!!!iiiiiiiiiiiiiiiiiil' i:: LOCATION MAP
0 YJOO 2000 FEET
SCALE
..
}
t WATANA
RELICT CHANNEL TOP OF BEDROCK .
FIGURE 9.12q
l
I
I
I
I
I
I
I
I
I
I
I
1:
I
1600 --------~~--~----~--~----~----~---r~--~---;
\
\
\
\
\
(') \
LESS THAN 3 '
ENTRANCE.
SUBMERGED
-1550 ~--------+----------~--~~;-~---------r~------~ .....: u.. -
0
TYPICAL
TUNNEL
SECTION
1450 ~--------L---------~--------~--~~------------~ 25 30 35 40 45
TUNNEL DIAMETER (FT.)
WATANA DIVERSION
HEADWArER · ELEVAT1 ON I TUNNEL DIAMETER
FIGURE $.13
I
'I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
l .. < • < <
-,_;
u.. -
z
0 -1-
<(
>
lJJ
...J
lJJ
~
<(
0
' i I
f
I 1600~-------h~------~~----~~------~
{
f
1
I • t
1550 t----f----+-----t------4-------1
1500~------~~------~------~------~
lOX 10 6 2.0X 10 6 30XJ0 6 40XI06 "
CAPITAL COST S
WATANA DIVERSION
UPSTREAM COFFERDAM COSTS
FIGURE 9.14
I
.~ C'
I
I
I
I
,
I -CD
0 -
X
I ... -
....
Cl)
I 0
(..)
...J < 1-
I -Q.
< (..)
I
I
I
I
I
I Q
I,
I
I
80----~--~--------~~~----~--------~--~---r~--~--~
50
40
30
20
15
0
0
TYPICAL
TUNNEL
SECTION
20
*DAM c STS INCLUDE AM
HEIGHT 15 1 ABOVE HelAOWAT~R
I
ELEVAT ON FOR FREEBOARD.
25 30 35 40
TUNNEL DIAMETER (FT.}
WATANA DIVERSION
TUNNEc·a COFFERDAM COST I TUNNEL DIAMETER
FIGURE 9.15
45
I
I
I
I
I
I
I
I
1:
·~
I
I
I
'.1
I
I
I.
I
I
-U)
0 -
X --
(I)
1-
(I)
0
0
..J
~
a.. <t
0
..J
<t t-
0 t-
80
70
60
0
TYPIC At:
TUNNEL.
SECTION
50
15 20 25 30 35
TUNNEL DIAMETER (FT.)
WATANA DIVERSION
TOTAL COST I TUNNEL DIAMETER
l
40
FIGURE 9.16
I
I
I
I. •
I
I -ID
0
I -)( ... -:r;
1-
I a::
0
==
1-z
IJJ
I (/)
IJJ a:
0..
:E
IJJ
I }-
(/)
>-(/)
:E
I a:
UJ
1-
(!)
z
0
I ..J
I
I
I
I
I
I
I
7300
72.00
7100
~
7000
6900
6800
6700
6600
6500
2140
0
.-
·-·-·
/~
~ ~
. I
.·
l
f
~
!
t
' I .
t
-..
2160 2180 2200 222.0 ~240 22.60
DAM CREST ELEVATION (FEET)
SELECTION OF RESERVOIR LEVEL
FiGURE 9.17
I
I
I
I
I
I
I
I
I
I;
I
I
I
I
I
I
I
I
.
0 1.
10 ... SELECTION OF DEVIL CANYON GENERAL ARRANGEMENT
This section describes the development of the -general arrangement of the Devil
Canyon project. The site topography, geology, and seismicity of the Devil Can-
yon site are described re 1 at ive to the design and arrangement of the various
site facilities, in a manner similar to that presented in Section 9 for the
Watana site. The method of handling floods during construction and subsequent
project 6peration is also 9utlined in this section.
The reservoir level fluctuations and inflow for Devil Canyon will essentially be
contra lled by operation of the upstream Watana project. This aspect is also
briefly discussed in this section. A detailed descriptioo of the various proj-
ect components is given in Section 13.
10.1 -Site Topography
The Devil Canyon site is located at river mile 152 of the Susitna River, approx-
imately 31 mi 1 es downstream from the Watana site, in a 11 V" shaped sect ion near
the entrance to the canyon which is about 2 miles long. The valley wall on the
left side of the river rises w~ry steeply from Elevations 900 to 1300 on the
left bank at a slope of approximately 0.4H:lV to a relatively gently sloping
plateau area which reaches Elevation 1600 within the general project area. On
the right side, the valley is less pronounced, rising at about 1.1H:lV to Eleva-
tion 1500, then much more gradually to approximate Elevation 1900. The steep
left bank features overhanging cliffs and detached blocks of rock.
10.2 -Site Geology
This section summarizes the gecilo~ical and geotechnical investigations and in-
terpretations conducted to date and the conditions present at the proposed Devil
Canyon site. The detailed description of the site investigations and the geo-
logic and geotechnical conclusions are presented in the 1980-81 Geotechnical
Report ( 1).
(a) Geologic Setting
Devil Canyon has been eroded through hard metamorphosed sedimentary rocks,
argillite and graywacke of excellent quality (Figure 7 .11). The bedding
strikes roughly· parallel to the river and dips to the south. Overburden is
generally thin to nonexistent. Stress relief cracks and open joints paral-
lel the gorge and extend more than 100 feet from the canyon walls ..
On the left (south) bank, a series of small 1 akes paral1el the valley.
Deep overburden up to 80 feet thick ilhas been encountered in this area which
probably represents an old stream buried under glacial material. A highly
sheared and fractured zone is present under this buried stream. Work per-
formed during this study, however, showed this feature to be of no seismic
concern (3).
A large alluvial fan exists at the confluence of Cheechako Creek with the
Susitna Kiver, about 1,000 feet upstream from the damsite. This area is
the rna in source of materia 1 for the concrete aggregates and the fi 1 ter
materials for the saddle dam.
10-1
(b) Geological and Geotechnical Investigatf~ns
Sur·face and subsurface investigations have been conducted by several organ-
; zat ions at different times. During the per tod from June ·1957 to August
1958, the USBR conducted geologic mapping and subsurface investigations at
this site. The subsurface investigations included drilling of rock and
overburden, in-hole testing, test pits, and laboratory tests. Subse-
quently, in 1978, the COE conducted seismic refraction surveys to expand on
this work to assess the suitability of the site. · During the years 1980 and
1981, more detailed investigations of geologic features were performed as
part of the current work program to establish the technical feasibility of
the project. These investigations have included air reconnaissance, air
photo.-in~erpretation, geologic mapping of rock and overburden ir.cluding in-
hole geophysical tests, and seismic refraction surveys .. Both in situ and
1 aboratory tests have been performed to determine the engineering charac-
teristics of soils and rocks. The location of drill holes, test pits, and
other investigations is shown in Figure 10.1. ·
Geologic mapping was concentrated in the immediate damsite area to define
the geology of the site in as much detail as possible. Under extremely
difficult conditions of accessibility, ground mapping traverses ·were accom-
plished by making maximum use of technical climbers in the gorge to augment
mapping activities on foot along the upper slopes. At each station, the
applicable 1 itho1ogy or· type of overburden, bedding, jointing, weathering,
degree of consolidation, exposure size, and elevations was noted and plot-
ted on maps for use in the interpretation. All accessible areas with rock
outcrops were mapped.
Seismic refraction surveys totalling 3~300 feet were performed on the south
side of the canyon across the lake area and the alluvial fan to delineate
their extent and characteristics.
Diamond core dri 11 ing was performed on the upper slopes and at river level
of the canyon.. The holes were cased through the overburden leaving access
for future te.sting and instrumentation. A total of 4,800 1 inear feet in 29
holes have been drilled at the Devil Canyon site. Comprehensive logs have
been developed for each hole. Water pressure testing using inflatable
packers was conducted in the holes for permeability calculations. Geophys-
ical logging and borehole photography were attempted during the 1980 season
in selected holes.
I
I
••
I
I
I
I
I
I
I
I
I
I
To monitor the ground water and ground temperature conditions at the site, I
piezometers and thermistors were installed in selected drill holes. Peri-
odic readings of these instruments, after stabilization, have been con-
ducted to give more detailed informat·ion of the conditions that may be en-·•·
countered during construction.
A series of tests was performed on the rock recovered from coring to deter-.•...
mine the engineering characteristics of the rock mass. The results of
these investigations were compiled, correlated, and interpreted to develop
the geologic picture of the damsite and the adjacent areas. The results of
the 1 aboratory rock tests are summarized in Table lQ.L J
10-2 ••
I
I
I
I
I.
I
I
I
I
I
I
I
I
I
I
I
I
I
I
(c) Construction Material Investigations
Most of the investigations for the construction materials at the Devil
Canyon site were performed during the years 1957 and 1958. Additional ex-
plo~ation during the years 1980 ar.d 1981 was undertaken to supplement the
previous work. The investigations have included geologic mapping, auger
drilling, excavation of test pits and test trenches, seismic refraction
surveys, and laboratory tests. A total of 8 auger holes, 2 test pits, and
4,100 linear feet of seismic refraction survey have been completed. The
location of the various sources for the concrete aggregate and the saddle
embankment dam are shown in Figure 10.1.
A major source of construction materials for the Devil Canyon Project is an
alluvial fan of deposits, which lies near the Cheechako Creek confluence
approximately 1,000 feet upstream from the arch damsite. The area contains
large quantities of sands and gravels with inclusions of boulders and
cobbles above the river level. Under a thin mantle of organic material, a
3-to-4-foot layer of silty sand overlies a layer of sandy gravel with
traces of silt and some cobble5 and boulders. This layer of sandy gravel
is about 80 feet thic~. With proper processing, this source will provide
coarse and fine aggregate for the concrete, fi 1 ters for the embankment dam
a'nd cobbles for the upstream shell of· the embankment dam. A composite
grain-size curve for this material is presented in Figure 10.2.
( i) Rockfi 11 Materia 1 for the Saddle Dam
The required quantities of rockfill can be obtained from the area
designated Quarry Site K. The rock in this site is primarily dior-
ite. It is hard, durable and fresh, and suitable for the embankment
construction. Also, the suitable portions of the rock excavated for
the found at ion of the arch dam and other project structures, such as ·
underground power facilities and main and emergency spillways, are
adequate for use in the embankment, subject to appropriate schedul-
ing of excavation and fill operations. Sufficient fill-quantities
are available at the site to meet all requirements.
(ii) Impervious Core Material for the Saddle Dam
No suitable source for the core material for the saddle dam has been
identified at this time near the site. For current feasibility
assessment purposes, it is planned that the core material will be
transported from Borrow Area D near the vJatana site, where suffi-
cient quantities of suitable material have been identified. A dis-
cussion of the engineering characteristics of those materials is
presented in Section 9. Additional investigations will be performed
in the future in an at tempt to 1 ocate a potentia 1 source nearer the
Devil Canyon site.
(iii) Filter Matet"'ial for the Saddle Dam
filter and transition zone materials will be obtained from the allu-
vial fan Borrow Are a, as discussed above.
10-3
(iv) Gravels and Cobbles for the Saddle Dam
If needed, sufficient quantities of clean gravel and cobbles can
also be obtained from the alluvial fan with proper processing."
(v) Concrete Aggregate
The coarse and fine aggregate for the concrete structures wi 11 also
be obtained from the alluvial fan upstream. The results of the lab-
oratory testing, presented in Table 10.1, indicate that the material
from this source is of adequate quality. The gravel particles are
generally rounded with accompanying subangular sands. Petrographic
analyses indicate that the material includes quartz diorites~ gran-
ites, andesites, diorites, dacites, metavolcanics, rocks, ap1ites,
breccias, schists, phillites, arqi1lites, and amphtbolites. Gener-
ally, the material has less than 2 percent deleterious constituents
such as chert, muscovite, and argillite.
(d) Geologic Conditions
The overburden and bedrock conditions at the Dev i1 Canyon site are discus-
sed in this section.
(i) Overburden
The valley walls at the Devil Canyon site are very steep and al"'e
generally covered by a thin veneer of overburden consisting primari-
ly of talus at the base (Figure 10.3). The flatter upland areas are
covered by 5 to 35 feet of overburden of g1 aci a1 origin. The topo-
graphic depression along the elongated lakes on the south bank has
an overburden covering in excess of 85 feet of g1 aci al materials.
The overburden on the alluvial terrace or point bar deposit at the
Cheechako Creek confluence thickens from 100 feet to more than 300
feet over a distance of less than 40U feet.
The river channel alluvium appears to be composed of cobbles~ bould-
ers, and detached blocks of rock and is inferred to be up to 40 feet
thick. A representative cross section across the valley is pre-
sented in Figure 10.4.
(ii) Bedrock Lithology
The bedrock at the Devil Canyon site is a low-grade metamorphosed
sedimentary rock consisting predominantly of argillite with inter-
beds of graywacke (Figure 10.5). The argillite is a fresh, medium-
to-dark gray, very thinly bedded, very fine grained argillaceous
rock with moderately well ... developed fo1 iation parallel to the bed-
ding. The graywacke is a fresh, light gray, mainly fine grained
sandstone with an argillaceous matrix. It is locally a conglomerate
with lithic. fragments up to two inches in size. The graywacke is
well indurated and·exhibits poorly developed to non-existent folia-
tion. The graywacke is interbedded with the argillite in beds gen•
erally less than 6 inches thick. Contacts between beds are tight
10-4
I
I
I
I ,,
I
I
I
I
I
I
I
I
I
·I
I
I
.I
I
I
I
I
I
I
I
.~
I
I
I
I
I
·I
I.
I
I
I
I
-.•
and both rock types are fresh and hard.. Minor quartz veins and
stringers have intruded the argillite. These are generally less
than 1 foot wide and unfractured with tight contacts. Sulphide min-
eralization is common with pyrite occurring in as much as 5 percent
of the rock.
The area has also been intruded by numerous felsic and mafic dikes
ranging from 1 inch to 60 feet wide (averaging 20 feet). The dikes
have northwest to north orientation (Figure 10.5) with steep dips.
When c 1 ose ly fractured they are easily eroded and tend to form steep
talus-filled gullies, some of which exhibit shearing with the host
rock.. The felsic dike~; are 1 ight gray silicic varieties including
aplite and rhyolite. The mafic dikes are fine grained and appear to
be of diorite to diabase composition.
(ii1) Bedrock Structures
Bedding
The arg i 1 1 ite/ graywacke. has been comp 1 ete 1 y deformed as evidenced
by refolded folds and the development of multiple foliations ..
The primary foliation parallels the bedding at 35° to 90~ {N35E
to E), subparallel tb the river, and dips 45° to 80°SE. \\f11en two
or mor·e foliations are parallel, the rock has a very slateyl
phyll it1c appearance, and when oblique, the rock appears massive.
The canyon at the dams i te appears to be contra 11 ed by the: bedding
planes.
-Joints
Four joint sets have been delineated at Devil Canyon, as shown on
the stereo plots (Figure 10.6). Set I (strike 320° to 355~ and
dip 60NE to 70NW) and Set ti (strike 040° to 065~ and 40° to 60~S
dips) are the most significant. Set I joints are the most promi-
nent with spacing of 15 feet to 2 feet, and on the upper canyon
~>;a 11 s of the south bank are open as much as 6 inches (Figure
10.5). Set III is subparallel to the bedding/foliation and 3 when
it intersects with Set I, can cause the formation of loose
blocks. Set III joints (strikes 005° to 030° and dips 85NW to
85SE) are also often open ·an the south bank and may dip towards
the river, creating potential slip planes. This set, however, has
variable spacing and sporadic distribution. The fourth set is a
minor set with low dip angles and variable strike orientatlon.
Joint spacings measured from the borehole cores range ftom less
than 1 foot to more than 10 feet. The spacing and tightness of
the joints increase with depth, and the iron oxide staining and
weathering extends up to ·80 feet.
~ Shears and Ftacture Zones
Shears and fracture zones were encountered in localized areas ·of
the site in noth outcrops and boreholes. Corre 1 at ion of the data
is shown on the interpretation map (Figure 10.5). Shears are
10-5
•w ,•."'
defined as areas containing. breccia, gouge, and/or slickens1 hies
indicating relative movement~ These zones are soft and friable
and are characteri .zed by high permeab i 1 i ty and core 1 oss during
, drilling. Fracture zones, often encountered in conjunction with
tht; shears, are zones of very closely spaced joints. With depth,
these .zones become smaller, tighter, and more widely spaced ..
Where exposed, they are eroded into deep gullies.
The most common trend of these features is northwest, parallel to
Joint Set I. These have vertical to steep northeast dips and are
generally less than 1 foot \vide. Northw·est trending shears are
also associated with .the contacts between the argillite and mafic
dikes and are up to 1 foot wide, wi·ch closely spaced joints prom-
inent in the dike itself~ ·
A second series of shears trend northeasterly, subparallel~ng the
. bedding/foliation and Joint Set II, and have· high angle southeas-
terly dips. These average less than 6 inches in width.
(e) Structural Features
Several. structural features at the Devil Canyon site were investigated dur-
ing· the 1980-81 program.
In summary, these included the east-west trending sheared and fractured
zone beneath the proposed saddle dam area; a bedrock drop-off beneath
Borrow Site G; and bedrock conditions beneath the Susitna.
Seismic refraction and drilling data confirm the existence of a highly
sheared ·and fractured zone on the 1 eft bank beneath the proposed saddle dam
that generally trends parallel to. the.river. The dip on this feature is
inferred_ to be paralle1'or subpara1tel-i~o-_,tl1e bedding/foliation at approxi-
mate 1y 65 o to the south. The 1 in ear extent of the feature has been infer-
red to be approximately 2~500 feet. No evidence was found during the
1980-81 program to suggest movement along this feature. This finding was
also concluded by work done by Woodward-Clyde Consultants (3). Further
investigation of this feature will be required to define its extent and
type of foundation treatment that will be required beneath the saddle dam
in subsequent phases of investigation.
Upstream from the damsite, a several-hundred-foot drop-off in bedrock sur-
face was detected by seismic refraction surveys under the alluvial fan ..
Land access restrictions imposed during the study prohibited any further
investigation of this area. Possible explanation for this apparent ·anamal-
ous drop-off could be attributed to misinterpretation of the seismic data
·or else the lower velocity material could be either a highly fractured rock
in lieu of soil or an offset of the rock surface caused by faulting. The
1 atter interpret at ion is unlikely in that work performed by Woodward-Clyde
Consultants (3) in this area concluded that there was no compelling evi-
dence for a fault.. Future work remains to be done in this area tu more
clearly define this featur.e~
Uetailed examination of rock core and mapping in the river valley bottom
showed no evidence for faulting in the riverbed.
10-6
I
I
'I· ::, ,'
I
I
I
I
I
I
:1
I
••
I
I
I
I
I
I
I
• I
I
I
I
I
I
I
I !.
I
1:
I
I
I
I
I
I
I
(f) Ground Water Co~ditions
Ground water migration within the rock is restricted to joints and frac-
tures.. It is inferred that the ground water level is a subdued replica of
the surface topography with the flow towards the river and lakes. Measured
water levels in the bOl·eholes range average approximately 120 feet below
surface. ·
(g) Permafrost
Preliminary temperature measurements made in the boreholes did not enccunt-
er permafrost conditions on either side of the river.
(h) Devil Canyon Reservoir. Geolo91_
'
The Oevi 1 Canyon reservoir will be confined to a narrow canyon where the
topography is controlled by bedrock. The overburden is thin to nonexis-
tent, except in the upper reaches of the reservoir where alluvial deposits
cover the valley floor. Near the Watana site, 1 ight gray to pink, medium
grain diorite rock is present. This rock is hard, massive, and competent
except on the upland north of the Susitna River where the biotite grana-.
dionite has been badly weathered. The principal rock types in the most
part of the reservoir are the arg i 11 ite and graywacke which are exposed at
the damsite. The,rock has been isoclinically folded into steeply dipping
structures striking generally northeast-southwest~ The argillite has been
intruded by massive granodiorite, and as a result, large isolated roof
pend ants of the argillite and graywacke are found 1 oc ally throughout the
entire reservoir and surrounding areas. The joint measurements at selected
areas indicate structural trends similar to those at the damsites.
10.3 ~-Geotechnical Consideratio~s
The geotechnical investigations to date have been primarily directed toward the
important geological features which may have significant impact on the feasibil~
ity o·f the project. More detailed investigatinns, including exploratory adits,
will be required prior to the detailed design.
(a) Arch Dam Foundation and Abutments
The geologic and topographic conditions are favorable for an arch dam ·at
the Devil Canyon site. The rock is principally hard, competent, and fresh
with weathering 1 imited to joints and shear zones. ~1trt1sive mafic and
felsic dikes, where present, are hard, and the contact with the parent rock
is tight. The orientation of these dikes is generally NW toN and has no
important adverse effect on the stability of the abutments. The unconfined
compressive strength of the intact rock ranges between l6,00U psi and
32,000 psi. The stresses imposed by the arch dam are about l,OOO psi or
less under normal conditions. Even under extreme loading conditions~ the
stresses will be well withln the acceptable 1 imits for bearing considera-
tions. Un the right abutment, the arch dam thrust block wi 11· be seated in
good sound rock. The topcof the hil1 is at approximately Elevation 1500
and no major rock discontinuity is present. However, on the left side,
massive thrust blocks will be required to transfer the loads to competent
rock and to form an abutment to the saddle dam.
10-7
Essentially no continuous, poorly oriented rock d·iscontinuities, which·
might adversely affect abutment stability, have been found. The major
joint set -:at the site, influencing the stability of the south bank aout-
ment,. ·strikes approximately northwest with a near vert ica1 average dip.
The st~b il ity o.f the right abutment (north bank) is controlled by the.
bedding planes and foliations that strike roughly parallel to subparallel
to the canyon walls and dip steeply into the canyon. The bedding planes
generally appear to be tight with undulating surfaces because of the ex-
tensive folding of the rock. Preliminary analyses indicate no stability
problems. Additional rock investigations and in situ testing will be re-
quired during final design to C6hfirm rock properties and the J·esults of
stability analyses.
The dam· and the thrust blocks will be. founded on sound rock. This will
require complete removal of all the overburden and weathered rock.. Along
some of the northwest trending shear zones, the weathering could be as deep
as 200 feet~ Extensive dental excavation may be required in these areas to
form an acceptable foundation. The entire dam foundation area should be
consolidation grouted to fill all the openings and cavities in rock at
shallow depth.
The permeability of rock varies from 1 x lo-4 em/sec at shai low depths
to 1 x lo-6 em/sec at depths below 175 feet. The permeability is con-
trolled by the discontinuities in the rock and may var·y widely from area to
area. A grout curtain will be provided under the entire dam including the
abutments and an appropriate distance beyond the dam into the abutments. A
system of dra.in holes and drainage galleries will be included to control
uplift pressures and to safely release seepage water. A double row curtain
is proposed,. There is little evidence of permafrost at this site; however,
provision should be made for thawing during grouting should permafrost be
encountered.
(b)_ Underground Structures
The rock conditions at the site ·are suitable for the construction of tun-
nels and underground caverns. From the geological and geotechnical view-
point, the 1 ocat ion and the orientation of these structures are influenced
by the orientation and location of rock discontinuities.
The RQD values indicate that about 50 percent of the rock is in the good-
to-excellent category, roughly 40 percent in the fair-to-good category, and
lhe remaining percentage in the poor category. The poor quality rock is
generally associated with fractures and shear zones.. The major joint sets
are oriented northwest (Set I) and northeast (Set II). Both sets are
steeply dipping.. The bedding plane strikes roughly in a NE-E direction and
dips at 45 to 80° southeast. The orientation of the tunnels and the large
underground caverns have been carefully selected to minimize the potentia 1
· adverse effects of these rock d i scant inui ties~
10-8
I
I
I
I
I
I
I
I
I ,,
I
I
I
I
I
I
I
I
I
I
I
~'' '· '·
''.•..c'
I
I
I
I.
I
I
I
'I
I.
I
I
I
I
I
I
I
(c)
. Oeterminat ion of the magnitude and the orienfat1on .of the in situ str~sses
· .. will_ rrot be possible until in situ-testing is undertaken. Nevertheless,
the tectonic setting suggests that the entire site region is in a compres-
sional stress regime. Near valley walls,. the stresses are expected to have
been relieved .and-low horizontal stresses may exist. Considering the un-
confined strength of the intact rock, overstressing problems such as rock
spa11 ing ~and sl abing are not anticipated~-The rock support requirements
will depend on the size and orientation of the openings and the presence
and character of the rock discontinuities intersected. For. the most part,
conventional rock bolt support u~ing 3/4-inch to l-inch-diameter bolts has
been assumed to be adequate for openings less than 40 feet in span. For
larger spans, in areas of poor quality rock and where rock discontinuities
are known to be adversely oriented, support requirements have been deter-
mined on a case-by-case basis~ In the case of 1 arge span openings (such as
powerhouse cavern), special attention has been given to the potential
presence of subhorizontal.Joints where they may intersect almost vertical
joints and may create unstable blocks in the crown of the excavation. The
use of shotcrete, welded wire fabric, and concrete lining will be required
in poor rock quality areas. For power tunne1s; provisions have also been
made for concrete lining and contact/consolidation grouting.
-
Although rock permeabil ities are generally lo"'' to moderate, intersection of
rock discontinuities may lead to ground water inflow problems during con-
struction and cause high-pore water pressures <.lfter the reservoir is flood-
ed. Therefore, provjsions have been made for grouting around tunnels and
caverns, and suitably placed drain holes and drainage galleries have been
provided upstream from the powerhouse and surge chamber.
The excavat1on of tunnels may be performed using conventional drill and
blast techniques or high-production mechanical excavators. Sufficient in-
formation is not available at this time to select an optimum system. For
cost estimating purposes, conventional methods have been conservatively
assumed at this time. The excavation of the powerhouse cavern will be by
drill and blast using a heading, side slash, and benching sequence.
The spacing between long tunnels should be 2.5 times the diameter of the
1 argest tunnel. The spacing between the caverns should be kept so that a
minimum pillar thickness of 1.5 times the span of the larger cavern is
maintained.
Stabi 1 ity of Soil and Rock Slopes
In most areas, the permanent excavation slopes will be confined within the
rock~ except on the left bank, where a deep buried stream exists. The
slopes within overburden will depend on the nature of soil, ground wat·er
table, and the height of the slope. In general, slop.es within the aver-
burden"nave been assumed as 2H:1V or less below the water table and 1 .. 5H:1V
or .less :above the water table. A bench of appropriate width will be pro-
vided at the overburden-rock contact to accommodate any local slumping/
slope failure and to intercept and dispose of ground/seepage water •. Flat-
ter s 1 opes are required in some areas where frozen ground may become un-
stable because of high pore pressures during thawing or where slope height/
soil conditions so dictate.
10-9
D
The slopes of excavations ir rock have bee_n se1ected in accordance with ~he
joint dips. and orientations and the shear. strength, of rock along discontn1-. _
uities. Slopes in intact rock or where discontinuities dip into the exc·a-.
vated face will usually stand steeply without any· structural support.
Slopes par;alleling the discontinuity have, whenever possiole, been laid
back to the same angle as the dip of the rock discontinuity or adequa.te
rock support provided. Wherever possible, permanent cuts have been set .at
stable slopes without the need for rock bolts. In areas where 1 arge pore
pressures·could potentially develop behind the rock cuts, allowances have
been made for drain holes to relieve the pore. pressures~ In general, a
4V:1H overa'll slope hi considered stable~ For slopes exceeding 40 feet in
height, benches have been included every40 feet~ Excavation of tunnel
portals includes pattern rock bolting and appropriate provision for
concrete/shotcrate to reduce the risk of unstable slopes. Special details
are required ·in areas where s 1 opes wi 11 intersect or cross 1 arger shear
zones or otherwise unstable rock.
(d) Saddle Dam Found at ion
The saddle dam on the south bank will be constructed across the buried
stream. The thickness of overburden in this area reaches up to 80 feet.
The bedrock below (argillite and graywacke) area is competent. The imper-
vious core, filters, and outer shells for the saddle dam wnl be founded on
sound rock. The prominent shear zone or fault which was found in the
saddle dam foundation, together with various she&r and fracture zones, has
been treated by means of provisions for con$olid~ion and curtain grouting
under the core as a continuation of foundation treatment for the arch dam.
10.4 -Seismic Considerations
The seismicity of the Sus i tn a Has in and the sources of earthquakes are discussed
in Section 7 of this report. This section presents the implications ofthe
seismicity on the design of the Devil Canyon project.
(a) Seismic Design Approach
For the earthquake engineering and,design considerations, the project
structures have been classified as either critical structures or noncriti-
cal structures .. Critical structures include the dam and similar major
structures whose failure may Y'esult in sudden and uncontrolled release of
1 arge volumes cf water which may endanger property and ·1 ives downstream ..
The noncritical structures are those structures whose failure can be
assessed as an economic or financial loss to the project· in terms of lost
revenue, repair, and/or replacement cost. Critical structures \'Jill be de-
signed to safely withstand the effect of the 11 Safety Evaluation'Earthquake 11
(SEE) for the site. No significant damage to these structures will be
. accepted under these conditions.
10-10
I
I
I
·1·'·. !
'
I
I
I
I
I
I
I
I
I
I
I
I
I ·.
I
.I·
I
I
'I
I
I
I '·
I
I
I
I
I
I
I
I
' I
for ·ctestgn nf critical .structures, the effective acceleration fcrr. the SEE
has been determ-ined as o. 8 :x actual SEE acceleration, together with a cor-
responding scaled respOnse spectrum. The selected SEE corresponds to the
11 terrain 11 or 11 detection 1eve1 11 earthquake ·which has been characterized as
. follows:
-Magnitude:
-Location!
-Maximum Acce 1 er at ion!
-Peak Spectral Acceleration:
6-1/4 to 6-:1/2
Approximate 1 y 3 km from structure
Me·,. 0.55g to 0.60g
84th p~rcentile 0~70 g ·
Mean 1.37g to 1.50g ·
84th percentile l.77g
The response spectra for this event are shown in Figure 10.7A. The effec-
tive peak ~cceleration for design of structures is then: ~
-Design a max = 0.8 x 0.70g =· 0.56g
(Sa) max = 0.56g x 2.5 = 1.4Ug
The design of non-critical .structures for earthquake conditions will be
undertaken using conventional Uniform Building Code recommendations.
(b) Safety ·Evaluation Earthquake
The design of the: Devil Canyon arch dam and other critical structures has
been undertaken using response spectrum analysis. Although the "terrai'n 11
-earthquake would result in more severe ground motions, the duration of
these motions is relatively short. However, the method of analysis used
for the Devil Canyon Dam does not ta~-e duration into account. The most
likely source of strong ground shak·ing at the Uevil Canyon site is, in
fact, the Benioff Zone. The estimated mean peak response spectrum for the
SEE is presented in Figure 10.7, along with the 84th percentfle response
spectrum .. A maximum horizontal acceleration level for the 84th percentile
response. spectrum for the Benioff event is approximately 0.47g.
10.~ -Selection of Reservoir Level
The selected normal maximum operating level at Devil Canyon Dam is Eleva-
tion 1455. Studies by the USSR and COE on the Devil Canyon Project were
essentially based on a simi 1 ar reservoir 1 eve 1, which corresponds to the
tailwater level selected at the Watana site. Although the narrow configur-
ation of the Devil Canyon site and the relatively low costs involved in in-
creasing the dam height suggest that.it might be economic to do so, it is
clear that the upper economic limit of reservoir level at Devil Canyon is
the Watana tailrace level. ·
The detailed studies of reservoir level at Watana (Section 9) indicated
1 ittle change in benefit-cost ratio over a 100-foot range of reservoir
level at the upper limit. Maximization of hydroelectric energy pr_Qd_uction
at the site was found to be an important objective which weighed heavily in
the selection of reservoir level at Watana. Although a detailed determina-
tion has not been undertaken, the same .is likely to be true at Devil Can-
yon. t:,
l0-11
Although significantly lower reservoir levels at Devil Canyon would lead to
lower dam casts; it is clearly evident .that the location of adequate spill-
\1/ay facilities in the nartow gorge would become extremely difficult and
lead to offsetting increases in cost. In the extreme case, a spillway dis-,
charging over the dam would raise. concerns regarding safety from scouring
at the toe of the dam, which have already led to rejection of. such
schemes. · ·
10.6 -Selection of Installed CapacitX
The methodology used for the preliminary selection of installed capacity at-
Watana and Devil Canyon is described in Section 9.6~
Thedecision to operate Devil Canyon essentially as a run-of-river plant with
_ maximum utilization of available flows from Watana was governed by the following
main considerations: · ·
-Daily peaking is more effectively performed at Watana than at Devil Canyon;
and
Excessive fluctuations in discharge from Devil Caryon will have a significant
impact on downstream fisheries.
Given this mode of operation, the required installed capacity at Devil Canyon
has been determined as the maximum capacity needed to uti1ize the available
energy from the hydrologit:al flows of record, as modified by the res lts of
reservoir operation computer simulation analysis (Section 9 .. 6), with the station
operating at 100 percent load factor. In years where the energy from Watana and
Devil Canyon exceeds the system demand, the usable energy has been reduced at
both stations in proportion to the average net head available, assuming that
power flow releases at Watana will also be used at Devil Canyon.
The total capacity required at Devil Canyon 1n a. wet year, excluding standby and
spinning reserve capacity, is summarized below. As discuss€d in Section 9 .. 6,
the capacity shown is based on the Battelle medium load growth forecast.
Demand Year
2002
2005
2010
Capacity MW
370
410
507
The selected total installed capacity at Devil Canyon has been established as
600 MW for feasibility design purposes. This will provide some margin of stand-
by for forced outage and possible accelerated growth in demand.
The major factors governing the selection of the unit size at Oev.il Canyon ·are
the rate of growth of system demand, the minimum stat ion outputs and the re-
quirement of standby capacity under forced outage conditions. The above tab-
ulation indicates that station maximun load in December will increase by about
50 percent from 2002 to 2010 (from 370 MW to507 MW). Station minimum output in
July during ·the same period will vary fr:?~ _about 150 MW to 202 MW ..
10-12
I
I
I
I
I
I
I
:1
;•
I
I
I
I
I
I ----
1
I
' I
I
.I
I
I
I
I
I
I
I
I
I
••
1.·
'.J
I
I .......
I
I
I
I
For feasibility design purposes, the power facilities at Devil Canyon have been
developed for 4 units at 150 IVIW, This arra.ngement ;vill provide for efficient
station operation at part load.. Consideration of phasing of the capacity in-·
stallation of the machines may be desirable as the system demand increases.
However·, the uncertainty of load forecasts this fan into the future~ and the
additional contractual costs of mobi·1ization for.equipment ins_tal1atton are such
that for study purposes at this t irne it has been assumed that a11 units wi 11 be
commissioned by 2002,.
.
The Devil Canyon reservoir wi 11 normally be full in December; hence~ Q.ny forced
outage will result in spilling and a loss of available energy~ The units have .
been rated to deliver 150 MW at minimum December drawdown level; this means that
in o.n average year, with higher reservoir levels, the full station outp:Ut can be
maintained even with one unit on forced nutage.
10.7 -Selection of Spillway Capacity
A flood frequency of 1:10,000 years was se 1 ected for the spi 11 way design on the
same basis as described for ·watana (Section 9), An emergency spillway with an
erodible fuse plug will also be provided to handle larger discharges up to the
probable maximum. flood. As discussed in Section 8 and· elsewhere, the develop-
ment plan envisages completion of the Watana project prior to construction at
Devil c·anyon. According1y, the inflow flood peaks wi11 be significantly ·less at
UeVil Canyon because of routing through the Watana reservoir. Spillway floods
as calculated in Section 7.2 are:
Flood
Probable Maximum
1:10,000 year
Inflow Peak (cfs)
366,000
165,000 .
,
The restrictions with respect to nitrogen supersaturation of.downstream flows
discussed in Section 9 _for Watana also will apply to Devil .Canyon, and dis-
charges of nitrogen-supersaturated water from Devi 1 Canyon will be 1 imited to a
recurrence period of not less than 1:50 years.
10.8 -Main Dam Alternatives
The location of the Devil Canyon damsite \'laS examined during previous studies by
the. USBR and COE. These studies focused on the narrow entrance to the canyon
and led to the recommendation of a concrete arch dam.. ~otwithstandifig this
initial appraisal, a comparative analysis was undertaken as part of thes.e fe.asi ...
bility studies to evaluate the relative merits of the following types of struc-
tures at the same location:
-Concrete thi·n arch;
-Concrete gravity arch; and
-Fill embankment.
(a) Comparison of Embankment and Concrete Type Dams
This analysis was based on the concrete. arch and concrete gravity arch
schemes developed by the COE fh'-1975 and 1978, together with a rockfill dam
alternative developed as part of the current study program. The results _of
lO ... lSo·
the aQalysis indicated a trend in favor of the concrete arch dam alterna·
tive when compared to the gravity or rockfi11 dam alternatives. The
assessment showed that a gravity dam in the narr.ow gorge would tend to
behave similarly to an arch dam but would not have the flexibtlity of such
a structure. The technical feasibility of a gravity dam was therefore
questionable particularly under severe seismic shaking conditions. This
type of dam also ,tended to be more expensive and was; therefore, not
considered further.
ConsidE:ration of a central core rockfill dam at Devil Canyon indicated rel-
atively small 'Cost differences from a conservative arch dam., significantly
thicker in cross section than the finally selected design. Furthermore, no
information was available to indicate that impervious core material could
be found for such a dam in the necessary quantities and within a reasonable
distance. The rockfill dam was accordingly dropped frGtil. further consider_a-
tion. Details of this evaluation are presented in Appendix 02.
Neither of the concrete arch dam 1 ayouts are intended as the final site
arrangement., but were sufficiently representative of the most suitable
arrangement associated with each dam type to provide an adequate basis for
comparison. Each type of dam was located just downstream of where the
river enters Devil Canyon close to the canyon's n~rrowest point which is
the optimum location_for all types of dams. A br.ef description of each
dam type and configuration is given below.
(i) Thick Arch Dam
The main concrete dam will be a single center arch structure, acting
partly as a gravity dam; w1th a vertical cylindrical upstream face
_and a sloping downstream fa,ce inclined at 1V~0.4H. The maximum
height of the dam will be 6.35 feet with a uniform crest width of 30
. feet, a crest 1 ength of approximately 1, 400 feet, and a maximum
foundation width of 225 feet. The crest E:levation will be 1460.
The center portion of the dam wi 11 be founded on a massive mass con--
crete· pad constructed in the excavated river bed. This central sec-
tion will incorporate a service spillway with sidewalls anchored
into solid bedrock and gated orifice spillways discharging down the
steeply inclined downstream face of the dam into a single _large
stilling basin set below river level and spanning the valley.
The main dam will terminate in thrust blocks high on the abutments-.
The 1 eft abutment thrust b 1 ock will incorporate an emergency gated
control spillway structure which will discharge into a rock channel
running well downstream and terminating at a high level in the river
valley.
Beyond the. control structure and thrust block, a low-lying saddle on
the. left abutment wi 11 be closed by means of a rockfili dike founded
on bedrock~ The powerhouse wi 11 house 4 x 150 MW units and will be
located underground within the right abutment. The multi-level
intake wiJ l be constructed integrally with the d.am and connected to
the powerhouse by vertical steel-lined penstocks~
The service spillway will be· designed to pass the 1:10~-000-year
routed flood with larger floods discharged downstream via the emer-
gency spillway ..
10-14
·I
,I
I
.I
I
I
I
I
I
I
I
'I
I
I -...-·
I
I
I
I
I
••••
~-.
I
I .,.
I .... ··
I
I
' I -.
a
I ........
I ..
I """
If
I
I
I
I -.;,·
I
I
{ i i) Thin Arch Dam
The main dam will be a two-center., doub1e-curv_ed arch structure of
similar height to the thick arch dam, but with a 20-f.oot uniform
crest and a maximum base width o..f 90 feet. The crest elevation will
be 1460. The center sect ion wi 11 be founded on a concrete pad, and
the extreme upper.portion of the dam wi11 terminate in concrete
thrust blocks located on the abutments •
~The main service spillway will be located on the right abutment and
will consist of a conventional gated control structure discharging
down a concrete-1 ined chute terminating in a flip bucket. The
bucket will discharg~ into an unlined plunge pool excavated in the
riverbed alluvium and located sufficiently downstream to prevent
undermining of the dam and associated structures.
The main spillway will be supplemented by orifice type spillways
located high i.n the center portion of the dam which will discharge
into a concrete-lined plunge pool immediately downstream from the
dam. An emergency spillway, cons·isting of a fuse plug discharging
into an unlined rock channel, terminating well downstream, will be
located beyond the saddle dan on the left abutment.
The concrete dam will terminate in a massive thrust block on each
abutment which, on the left abutment, will adjoin a rockfill saddle
dam.
The service and auxiliary spillways will be designed to discharge
the 1 :10,000-year flood. Excess floods for storms tip to the prob-
able maximum flood will be discharged through the emergency left
abutment spillway .
(iii) Comparison of Arch Dam Types
Sand and gravel for concrete aggregates are believed to .be avai1able
in sufficient quantities within economic distance from the dam as
discussed in Sections 10.2 and 10.3. The gravel and sands are
formed from the granitic and metamorphic rocks of the area; at this
time it is anticipated that they wil_l be suitable for the production
of aggregates after a moderate amount of screening and washing,.
The bedrock geology of the site is discussed in .Sections 10.2 and
10.3. At this stage it appears that there are no geological or geo-
technical concerns that would preclude either of the dam types from
consideration.
T~te thick arch dam will allow for the incorporation of a main spill-
way chute on the downstream face of the dam discharging into a
spillway located deep within the present riverbed. This spillway
wi 11 be able to pass routed floods with a return frequency of less
than 1:10,000 ye-ars. For the thin arch and rockfill alternatives,
the equivalent discharge capacity will be provided separately
through the abutments.
. 10-15
Under hydrostatic and temperature loadings, stresses within the
thick arch dam will be generally lower than for the thin arch alter-
native. However, finite.' element analysis has shown that the addi-
tional mass of the dam under seismic loadings will produce stresses
of a greater magnitude in the thick arch dam than in the thin arch
dam. If the surface stresses approach the maximum allowable at a
particular section, the remaining understressed area of concrete
will be greater for the thick arch, and the factor of safety. for the
dam wiii be correspondingly higher. The thin arch is, however, a
more efficient design and better utilizes the inherent properties of
the concrete. It is designed around acc:eptabl e predetermined fac-
tors of safety and requires a much smaller volume of concrete for
the actual dam structure.
The thick arch arrangement did not appear to have any outstanding
merits compared to a thin arch dam and would be more expensive be-
cause of the larger volume of concrete needed. Studies, therefore,
continued, on refining the feasibility of the thin arch alternative.
10.9 -Diversion Scheme Alternatives
In this se&tion the selection of general arrangement and the basis for sizing of
the diversion scheme are presented.
(a) General Arrangements
The steep walled valley at the site essentially dictated that diversion of
the river during construction be accomplished using one or two diversion
tunne 1 s ~ with upstream and dO\oJnstream cofferdams protecting the main con-
struct ion area.
The selection process for establishing the final general arrangement in-
cluded examination of tunnel locations on both banks of the river. Rock
conditions for tunneling did not favor one bank over the other. Access and
ease of construction strongly favored the left bank or abutment, the obvi-
ous approach being via the alluvial fan. The total length of tunnel re-
quired for the 1 eft bank is approximately 300 feet greater; however~ access
to the right bank could not be achieved without great difficulty.
(b) Design Flood for Uiversion
The recurrence interval of the design flood for diversion was established
in the same manner as for Watana (see Sect ion 9). Accordingly~ at Devil
Canyon a risk of exceedence of 10 percent per annum has been adopted, equi-
valent to a design flood with a 1:10-year return period for each year of
critical construction exposure. The critical construction time is esti-
mated q.t 2.5 years. The main dam could be subjected to overtopping during
construction without causing serious damage, and the existence of the
Watana facility upstream will offer considerable assistance in flow regula-
tion in case of an emergency. These considerations led to the selection of
the design flood with a return frequency of 1:25 years.
The equivalent inflow) together with average flow characteristics of the
river significant to diversion, is presented below:
0
'10-16
I
I
I
I
I
I .:-~
' J
I
I <-~
I ...
'I
I
I
I ._
I ,,
.....
I
I
I ·-
1
I
I •..r'
I
"
•• "-
1
' I
I
' I ...,.
I
' I
I
' .I
I
(c) Cofferdams
Average annual flow: _
Maximum average monthly flow:
Minimum averagfr~onthly flow:
Uesign flood inflow (1:25 years
routed through Watana
reservoir):
7,860 cfs
23,100 cfs (June)
890 cfs (March)
37,800 cfs
As at Watana, the considerable depth of riverbed alluvium at both cofferdam
sites indicates that embankment type cofferdam structures would be the only
technically and economically feasible alternative at Devil Canyon~ For the
purposes of establishing the overall general arrangement of the project and
for subsequent diversion optimization studies, the upstream coffe·rda.'ll sec-
tion adopted will comprise an initial closure section approximately 20 feet
high constructed in the wet, with a zoned embankment constructed in the
dry.. The downstream cofferdam will comprise a closure dam structure ap-
proximately 30 feet high placed in the wet. Control of underseepage
through the relatively pervious alluvium material will be achieved by means
of a grouted zone.· The nature df the alluvium led to the selection of a
grouted zone rather than a s 1 urry wa 11.
The selected cofferdam sections are described in more detail in Section 13,
(d) Diversion Tunnels
Although studies for the W;:::.an a project indica ted that· concrete-1 ined
tunnels were the most ecof~O~llically and technically feasible so1ution~ this
aspect was reexamined at Devil Canyon. Preliminary hydraul i·c studies indi-
cated that the design flood routed through the diversion scheme v1ould re-
sult in a design discharge of approximately 37,800 cfs. For concrete~ined
tunnels, design velocities of approximately 50 ft/s have been used in sev-
eral projects. For unlined tunnels, maximum design velocities ranging from
10 ft!s in good quality rock to 4 ft/s in less _competent material are typi-
cal. Using a maximum permissible velocity of 10 ft/s, four unlined
tunnels, each with an equivalent circular diameter of 35 feet, would be
required to pass the design flow. Alternatively, a design velocity of 50
ft/s would theoretically permit the use of one concrete-lined tunnel with
an equivalent diameter of 30 feet. As was the case for the Watana diver-
sion scheme, considerations of reliabnity and cost were considered suffi-
cient to eliminate consideration of unlined tunnels for the diversion
scheme.
For the purposes of optimization studies, only a pressure tunnel was con-
sidered, since previous studies (Section 9) indicated that cofferdam clo-
sure problems associated with free-flow tunnels would more than offset
their other advantages.
Pressure tunnels are designed to flow full and, accordingly, must withstand
internal pr·essure. The most widely used type of pressure tunnel for diver-
sion has the crown of the outlet portal submerged during all flow condi-
tions. The tunnel cross-section used for this evaluation was a modified
horseshoe or 11 0 11 sh-aped configuration. The area of this type of section is
10-17
...• ~~··or·~
.. ·,1
(e)
13.7-percent greater than for a circular tunnel with the same diameter of
span. The 11 0 11 -shaped tunnel offers advantages in terms of ease of con-
.struction and scheduling for the spans envisaged at Devil Canyon.
Optimization of Diversion Scheme
Given the considerations described above relative to design flows~ coffer-
darn configuration; and alternative types of tunnels, an economic study was
undertaken to determine the optimum combination of upstr·eam cofferdam ele-
vation (height) and tunnel diameter. _,
Capital costs were developed for these single pressure tunnel diameters and
corresponding upstream cofferdam embankment crest elevations with a 30-foot
wide crest and exterior slopes of 2H:1V. A freeboard allowance of 5 feet
was inc 1 uded for sett 1 ement and vJave run up.
Capital costs for the tunnel alternatives jncluded allowances for excava-
tion, concrete liner~ rock bolts, and steel supports. Costs were also
developed for the upstream and downstream portals, including excavation and
support. The cost of intake and outlet gate structures and associated
gates was determined not to vary significantly with tunnel diameter and was
excluded from the analysis ..
The centerline tunnel length in all cases was assumed to be 2,000 feet,
reflecting a left bank location.
Rating curves for the single-pressure tunnel alternatives are presented in
Figure 10.8. The relationship between capital cost and crest elevation for
the upstream cofferdam is shown in Figure 10.9. The capital cost for
various tunnel diameters is given in Figure 10.10:
The results of the opt im izati on study are presented in Figure 10.11 and
indicate the following optimum solutions for each alternative.
Tunnel Diameter
25 feet
30 feet
35 feet
Cofferdam Elevation
945 feet
945 feet
945 feet
Total Cost
$8,000,000
$6,600,000
$7,100,000
The selection of the diversion scheme was based on economics, a single, 30-
fo-ot-diameter pressure tunnel being selected. An upstream cofferdam 60
feet high, with a crest elevation of 945, was carried .forward as part of
the selected general arrangement.
The various components of the selected diversion scheme are described in
Section 13.
10.10 -Spillway Alternatives
As discussed in Section 10.7, the project has been designed to safely pass
floods with the following return frequencies:
10-18
I
I
t
I -...
I
;I' :.~
I
I
I
I -,,
I
I ...
' I
I
I ,..
I
I
I
' I
I
I
·I '-
1
I
I
'I
·I
I
I
I
t
I
t
' I
10.10 ~ Spillway Alternatives
As discussed in Section 10.7, the project has been designed to safely pass
floods with the following return frequencies:
Flood
Spillway Design Flood
Probable Maximum Flood
Ftequency
1:10,000 years ,,
U i scharge ( cfs)
135,000
270,000
A number of ·alternatives were considered singly and in combination for Devil
Canyon spillway facilities. These included gatd overflows or orifices in the
main dam discharging into a plunge pool downstream, right and left bank chute or
tunnel spillways in the flip buckets or stilling basins for energy dissipation,
and emergency open channel spillways. As described ·for Watana in Section 9, the
seler:tion of spillway facilities greatly influenced ar{d was influenced by the
gene~"Jl arrangement of the major structures. In general, the main spilh'/ay
facilities will discharge the design flood through a gated spillway control
structure with energy dissipation either by a flip bucket which directs the
spillway discharge in a free fall jet into a plunge pool in the river or by a
stilling basin. which dissipates the energy in a hydrau.ic jump. In addition,
similar restrictions apply with respect to limiting nitrogen supersaturation in
the spillway discharges. The various spillway arrangements developed in
accordance with these considerations are discussed in Sections 10.13 and 10.14.
10.11 -Power Facilities Alternatives
The selection of the optimum arrangements for the power facilities involved
consideration of the same factot~s as described in Section 9.11 for Wat.ana.
The selection of the installed capacity of 600 MW at Devil Canyon is des-
cribed in Section 10.6.
(a) Comparison of Surface and Underground Powerhouses
A surface powerhouse at Devi 1 Canyon would be located either at the down-
stream toe of the dam or along the side of the canyon wall. As determined
for Watana, costs favored an underground arrangement. In add it ion to cost,
the underground powerhouse layout has been selected based on the following:
-Insufficient space is available in the steep-sided canyon for a surface
powerhouse at the base of the dam;
-The provision of an extensive intake at the crest of the arch dam would
be detrimental to stress conditions in the arch dan particularly under
earthquake loading, and would require significant changes 1n the arch dam
geometry; . and _,
-The outlet facilities located in the arch dam are designed to discharge
directly ·into the river valley; these would cause significant \'linter
icing and spray problems to any surface structure below the dam.
10-19
-.
(b)
(c)
(d)
Comparison of Alternat ive,Locat-i-ons
..
The underground powerhouse and related facilities have been located on the
·right bank for.the following reasons:
-Generally superior rock quality at depth;
The left bank area behind the main dam thrust block is unsuitable for the
construction of the power intake; and
-The river turns north downstrean from the dam, and hence, the right bank
-power development· is more suitable for extending the tailr·ace tunnel to
develop extra head.
Selection of Units
The turbine type selected for the Uevil Canyon development is governed by
the design head and specific speed and by economic considerations. Francis
turbines have been adopted for reasons similar to those discussed for
Watana in Section 9.11.
The selection of the number and rating of individual units is discussed in
detail in Section 10.6 .. 1he four·units will each be rated to deliver 150.
MW at full gate opening at minimum reservcir level in December (the peak
month). The best efficiency unit output at rated head {575 feet) is 164
MW.
Transformers
Transformer selection is similar to the procedure for Hatana as discus.sed
in Section 9.11.
The arrangement of the transformr~rs at Devil Canyon ;·s described in detail
in Section 13.
Power Intake and Water Passages
For flexibility and security of operation, individual pensto~ks are pro-
vided to each of the four units. As discussed in Section 9.11 for Watana,
it was found that there is no significant cost advantage in using t\«l ·
1 arger diameter penstocks with bifurcation at the powerhouse._
A single tailrace tunnel has been assumed 6,80U feet in length, to develop ·
a further 30 feet of head downstream from the dam. Detai 1 ed des.i gn may in-
dicate this should be changed to two smaller tailrace tunnels for improved
station security; the extra cost involved would be insignificant. T'ne
surge chamber design for two tailrace tunnels would be relatively un-
changed.
The overall length of the intake structure is governed by the selected pen-
stock diameter and the minimum penstock spacing. Deta'iled studies were
10-20
. . . : .
.. . ' . ..
I
.. : •. ·
.....
I
I
I
I -;·
I
.1:
~·
I
I
I
I
I .... ,,
I
I
I
I
I
I
I
f
I
i
I
' I '""
I
I
I
I
' I
-
I
I
I
I
carried out to determine the optimum diameter of the penstocks and the
tailrace tunnel, in a similar manner to that described for Watana in
Section 9.11.
(f) Environmental Constraints
In addition to potential nitrogen-saturation problems caused by spillway
operation~ as discussed in Section 10.10~ the major impacts of the Devil
Canyon power facilities development are:
-Changes in the normal temperature regime of the river; and
-Fluctuations in downstream river flows and levels.
Temperature modeling has indicated that varying the intake design at Devil
Canyon would not significantly affect downstream water temperatures, since
these are effectively controlled by the water released from Watana. Conse-
quently, the intake design at Devil Canyon incorporates a single level
draw-off about 75 feet below normal reservoir operating level (El 1455}.
The Devil Canyon station will be operated as a base-loaded plant throughout
the year, to satisfy the requirement for no siginificant daily variation in
power flow.
10.12 -General Arrangement Selection
The approach to select ion of a general arrangement for Devil Canyon was a sim-
ilar but simplified version of that used for Watana, as described in Section 9.
(a) Selection Methodology
Pre 1 iminary alternative arrangements of the Devil Canyon project were
developed and selected using two rather than three review stages. Topo-
graphic conditions at this site limited the development of reasonab1y feas-
ible layouts, and initially, four schemes were developed and evaluat::r:L
During the final review, the selected layout was refined based on techni-
cal, operational and environmental considerations identified during the
preliminary review.
(b) Design Data and Criteria
The design data and design criteria on which the alternative layouts were
based is presented in Table 10.2. Subsequent to selection of the preferred
Devil Canyon scheme, the information was refined and updated as part uf the
on-going study program. The aescription of the Devil Canyon project pre-
sented in Section 13 reflects the most recent design d-ata for the prcject.
10.13-Preliminary Review
Cons·icleration of the options available for types and locations of various struc-
tures 1 ed to the development of four primary layouts for ex.aminat ion at Devil
Canyon in the preliminary review phase. As discussed above, pr'evious studies
10-21
had led to the selection of a thin concrete arch structure for the main dam, and
indicated that the most .acceptable technical and economic locQ.tion was at the
upstream entrance to the canyon. The dam axis has been fixed in this location
for all alternatives.
(a) Description of Alternative Schemes
0
The schemes evaluated during the pre.l iminary review are described below.
In each of the alternatives· evaluated, the dam is founded on a mass con-
crete plug, constructed on the sound bedrock underlying the riverbed.. The
structure is 635 feet high, has a crest width uf 20 feet, and a maximum
base width of 90 feet. Mass concrete thrust blocks ·are founded high on the
abutments, the left block extendihg approximately 100 feet above the exist-
ing bedrock surface and ·suppor-ting the upper arches of the dam. The thrust
block on the right abutment makes the cross-river profile of the dam more
symetrical and contributes to a more uniform stress distribution.
( i) Scheme DC 1 (See Plate 23)
In this scheme, diversion facilities comprise upstream and down-
stream earthfill and rockfill cofferdams and two 24-foot-diameter
tunnels beneath the left abutment (Section 10.9).
A rockfill saddle dam occupies the lower lying area beyond the left
abutment running from the thrust block to the higher ground beyond ..
The impervious f i 11 ::ut-off for the s add 1 e dam is founded an bedrock
approximately 80 feet beneath the existing ground surface. The
maximum height of this dam above the foundation is approxima~ely 200
feet. ·
The routed 1:10,000-year design flood of 135,000 cfs is passed by
two spillways. The main spillway is located on the right abutment.
It has a design discharge of 90,000 cfs, and flows are controlled by
a three-gated ogee control structure. This discharges down a ·
concrete-1 ined chute and over a ski-jump flip bucket which ejects
the water in a diverging jet into a pre-excavated plunge pool in the
riverbed. The flip bucket is set at Elevation 925, approximate1y 35
feet above. the river level. An auxiliary spillway, discharging a
total of 33,000 cfs, is located in the center of the dam, 100 feet
be 1 ow the dam crest and is contra 11 ed by three whee 1-mounted gates ..
The orifices are designed to direct the flow in to a concrete-lined
plunge pool just downstream from the dam.
An emergency spillway is located in the sound rock south of tne
saddle dam. · This is designed to pass discharges in excess of the
1:10,000-year flood up to a probable maximum flood· of 270,000 cfs,
if such an event should ever occur. The spill way is an un 1 ined rock
channel which discharges into a valley downstream from the dam lead-
ing into t.he Susitna River.
10-22
I
I ,...
t
I
I
I
t
I
I
I
I
I
I .... ,,
I
I
I
I
I
I
' I
I
I
I
I
I
I
I
I
I
I.
I
I
I
I
I
••
' -----;------;:-~.---------;-;:-;------~--.~ .. -,--:--~-.,
The upstre.am end of the channe 1 ·;s closed by an earthfi 11 fuse plug.·.
The plug is designed to be eroded if overtopped by the reservoir~
Thus, as the crest is lower. than either the mainor saddle dams, the
plug would be washed out prior to overtopping of either of these
structures. ·
The underground power facilities are located on the right bank of
the -river, within the bedrock forming the dam abutment. The rock
\'lithin this abutment is of better quality with fewer shear zones and
t; a lesser degree of jointing than the rock on the left side of the
canyon (see Section 10.3), and hence more suitable for underground
excavation.
-
The power intake is located just upstream from the bend in the
valley before it turns sharply to the right into Devil Canyon. The
intake structure is set deep into the rock at the downstream end of
the approach channel. Separate penstocks for each unit lead to the
powerhouse.
The powerhouse contains four 150 ~lW turbine/generator units. The
turbines are Francis type units coupled to overhead umbrella type
generators. The units are servi~ed by an overhead crane running the
length of the powerhouse and into the end service bay. Offices, the
control room, switchgear room, maintenance room, etc., are located
beyond the service bay. The transformers are housed in a separate,
upstream gallery located above the lower horizontal section of the
penstocks. Two vertical cable shafts connect the gallery to the
surface. The draft tube gates are housed above the draft tubes in
separate annexes off the main powerhall. The draft tubes converge
in two bifurcations at the tailrace.tunnels which discharge, under
free-flow conditions, to the river. Access to the powerhouse is by
means of an unlined tunnel leading from an access portal on the
right side of the canyon.
The switchyard is located on the left bank of the river just down-
stream from the saddle dam, and the power cables from the trans-
formers are carried to it across the top of the dam.
(ii) Scheme DC 2 (See Plate 24)
The layout is generally similar to Scheme DC 1 except that the chute
spillway is located on the left side of the canyon. The concrete-
1 ined chute terminates in a ski,..jump flip bucket high on the left
side of the canyon which drops the discharges into the river below.
The design flow is 90,000 cfs, and discharges are controlled by a
3-gated, ogee-crested-contro 1 structure, simi 1 ar to tt1at for Scheme
DC 1, which abuts the left side thrust block.
The saddle dam axis is straight, following the shortest route be-
twe~n the contra 1 structure at one end and the rising ground beyond
·the 1 ow-lying area at the other. -
10-23
(iii) Scheme DC 3 (See Plate 25)
The layout is similar to Scheme DC 1 except that the right side main
spil ha.Jay takes the form of a single tunnel rather than an open
chut.e. A 2-gated, agee-control structure is located at the head of
the tunnel and discharges into an inc1 ined shaft 45 ·feet diameter at
its upper end .. The. structure will discharge up to a maximum of
90,000 cfs.
The concrete-1 ined tunnel narrows to 35 feet diameter and discharges
into a flip bucket which directs the flows in a jet into the river
below as in Scheme DCl.
An auxiliary spillway is located in the center of the dam and an
emergency spillway is excavated on the left abutment.
The 1 ayout of dams and power facilities are the same as for Scheme
DC 1.
(iv) Scheme DC 4 (See Plate 26)
9
The dam, power facilities, and saddle dam for this scheme are the
same as those for Scheme DC 1. The major differ.ence is the subst i-
tution of a stilling-basin type spillway on the right bank for the
chute and flip bucket. A 3-gated, agee-control structure is located
at the end of the dam thrust block and contra 1 s the discharges, up
to a maximum of 90,000 cfs~
The concrete-1 ined chute is built into the face of the canyon and
discharges into a 500-feet-long by 115-feet-wide by 100 feet high
concrete stilling bijsin formed below river level and deep within the
right side of the canyon. This arrangement forms the service spill-
way with central orifices in the dam and the left bank rock channel
and fuse plug forming the auxiliary and emergency spillways,
respectively, as in the alternative schemes.
The downstream cofferdam is located beyond the spillway, and the
diversion tunnel outlets are located farther downstream to enable
construction of the stilling basin.
(b) Comparison of Alternatives
As the arch dam, saddle dam, power faci1 ities, and diversi.on vary only in a
minor degree among the alternatives, a comparison of schemes rests solely
with a comparison of the spillway facilities.
As can be seen from a comparison of the costs in Ta~le 10. , the flip
bucket spillways are substantially less costly to construct than the
stilling-basin type of ~:heme DC 4 .. The left side spillway of Scheme DC 2
runs at a sharp angle to the river and ejects the discharge jet from high
on the canyon face toward the opposite side of the canyon. Over a 1 anger
10 .... 24
. 1 -
I . '
I , .....
I
I
I
I
I
I
I
I
I
I
I
t
I
I
I
I
-
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
(c)
period of operation, scour of the heavily jointed rock could be a consider-
able problem causing undermining of the canyon sides· and their subs.eque.nt.
instabi1 ity, together with the possibility of a deposition of material
downstream with a correspondi11g elevation of the tailrace.. Construction of
a spillway on the steep left side of the river could be more difficult than
on the right side because of the presence of deep fissures and large un-
stable blocks of rock which are present on the left side close to the top
of the canyon.-
The two-right side flip bucket spillways schemes, based on e.ither an open
chute or a tunnel, take advantage of a downstream bend in the river to
eject discharges parallel to thu course of the river. This will reduce the
effects of erosion but could still present a problem, as can be seen from
the outline of the estimated maximum possible scour hole which would occur
aver a period of time.
The tunnel type spillway could prove difficult to construct because of the
large diameter inclinedshaft and tunnel paralleling the bedding planes~
The high velocities~ encountered in all spillways, could particularly cause
troubles in the tunnel with the possibility of spiraling flows and severe
cavitation.
The stilling bas in type spillway of Scheme DC 4 reduces downstream erosion
problems within the canyon. However; cavitation could be a problem under
the high-flow velocities experienced at the base of the chute. This \"lOU1d
be somewhat alleviated by aeration of the flows, introducing air into the
water/concrete contact area at offsets along the chute invert. There is,
however, 1 ittle precedent for stilling basin operation at heads of over 550
feet; and even where floods of much less than the design capacity have been
discharged, severe damage has occurred.
Selection of Final Scheme
The chute and flip bucket spill ways of Schemes DC 1 and DC 2 could generate
downstream erosion problems which could, in the case of Scheme DC 2~ re-
quire considerable maintenance costs an.d cause reduced efficiency in opera-
tion of the pro.ject al a future date. Scheme DC 3 causes hydraulic prob-
lems and cavitation co.uld be severe. There is no cost advantage in this
type of spillway .over the open chute. In Scheme DC 4, the operating char-
acteristics of a high head stilling basin are little known, and there are
few examples of successful operation.
All spillways operating at the required heads and discharges wi 11 event-
ually cause some erosion. For all schemes, use of auxiliary release facil-
ities in the dam to handle floods up to 1:50-year frequency is considered a
reasonable approach to limit erosion and nitrogen supersaturation problems ..
The cost of the flip bucket type spillway in the scheme is considerably
less than that of the stilling basin in Scheme UC 4" The latter offers no
relative operati.onal advantage; therefore, Scheme DC 1 has been selected
for further study. · -
10-25
'"""""'''" ,_,~.,~.-·-··-_ ...
•• ~. _· < ~
-. .
10.14 -Final Review
The layout selected in Section 10.13 was further developed in.accordrince with
updated engineering studies and criteria. The major change compared to Scheme
DC 1 is in the central spillway configuration, but other modifications that were
introduced are described below.
The revised 1 ayout is shown on Plate 27. A description of the structures is as
follows.
(a) Main Dam
The maximum operating level of the reservoir was raised to E·levation 1455
in accordance with updated information relative to the Watanc\ tail water
level. This requir.es. raising the dam crest Elevation to 1463: with the con-
crete parapet wall cr~st at Elevation 1466. The saddle dam '!'las raised to
Elevation 1472.
(b) Spillways ,and Outlet Facilities
To alleviate the potential for nitrogen supersaturation problems, it was
necessary to restrict supersaturatea flow to an average recurrence interval
of not less than 50 years. In order to pass floods of greater freqency, an
alternative type of discharge facility was required.
In addition, it was considered probable that frequent and
nance would be required in the concrete-1 ined plunge pool
tral or•ifice spillyJays and just downstream from the dam.
cal area because of· the proximity to the dam.
costly mainte-
beneath the ceo-
This is a criti-
These two considerations led to the replacement of orifice spillways by
outlet facilities incorporating five fixed-cone valves, with a diameter of
108 inches~ capable of passing aAesign flow of 45,000 cfs.
The chute spillway and flipbucket are located on the right bank, as in
Scheme DC 1; however, the chute length was decreased and the elevation of
.the flip bucket raised compared to Scheme OC 1.
!"lore recent site surveys indicated that the ground surface in the vicinity
of the saddle dam was lower than originally assumed. The emergency spill-
way channel was relocated slightly to the south to accommodate the larger
dam.
(c) Diversion
The previous twin diversion tunnels were replaced by asingle-tunne·1
scheme. This was determined to provide all necessary security but will be
slightly less expensive than the two-tunnel alternative (see Section 10.9).
10-26
t ...
' I
I
I
I
t
I
I
I
I
' I
I
I
I
I
I
I
I
I
·I
I
' I
I
I
I
I
I
I
I
I
I
I
I
I
(d) Power Faci1 ities
The drawdown range of the reservoir was reduced, allowing a reduction in
height of the power intakew In order to locate the intake within solid
rock, it has been mm,_erl_into the side of the valley, requiring a slight
rotation of the water passages, powerhouse, and caverns comprisitlg the
power facilities.
10-27
LIST OF REFERENCES
,
(l) Acres American lncorpqrated, Report on 1980-81 Geotechnical
Investigations, February 1982.
I
I
(2) Woodward-Clyde Consultants, Interim Report of Seismic Studies for th~ 1.
Susitna Hydroelectric Project, December 1980.
(3) Woodward-Clyde Consultants, Final Report of Seismic Studies for the
Susitna H~roelectric Project, February 1982. ~
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
TABLE 10,2: OE:SIGN DATA AND DESIGN CRITERIA FOR
REVIEW OF' ALTERNATIVE LAYOUTS
River flows
Average flow (ave~.-30 years of record):
Probable maximum flood:
Max. flood with return period of 1:107 000 years:
Maximum flood with return pe~iCid of 1 :500 years:
Maximum flood with return period of 1:50 years:
Reservoir
Normal maximum operating level:
Reservoir minimum operating level:
Area of reservoir at maximum operating level:
Reservoir live storage:
Reservoir full storage:
Dam
Type:
Crest elevation:
Crest length:
Maximum height above foundation:
Crest width:
Diversion
Cofferdam types:
Upstream cofferdam crest elevation:
Downstream cofferdam crest elevation:
Maximum pool level during construction:
Tunnels:
Outlet structures:
final closure:
Releases during impounding:
Spillway
Design floods:
Service spillway ..... capacity:
-control structure:
-energy dissipation:
Secondary spi11wa}' -capacity:
-control. structure:
-energy dissipation:
Emergency spillway -capacity:
-type:
8 1 960 cfs
270,000 cfs
135,000 cfs (after ro~ting
through Watana
42,000 cfs {after routing
through Watana
1,455 feet
1,430 feet
21,000 acres
180,000 acre feet
1,100,000 acre feet
Concrete arch
1 ,455 feet
6)5 feet
20 feet
Rock fill
960 feet
900 feet
955 feet
Concrete lined
Low-level structure with
slide closure gate
Mass concrete plugs in
line with dam grout curtain
2,000 cfs min. via fixed-cone
valves
Passes PMf, preserving
integrity of dam with no
loss of life
Passes routed 1 :10,000-year
flood with no damage to
structures
45,000 cfs
fixed-cone valves
five. 108-inch diameter
fixed-cone valves
90,000 cfs-
Gated, agee crests
Stilling basin
pmf minus routed 1:10t000-year
flood
F'use plug
TABLE 10.2: (Continued)
Power Intake
Type:
Transformer area:
Access
Type of turbin.es;
Number and rating:
Rated net head:
Maximum gross head:
Type of generator:
Rated output:
Power factor:
Underground
Separate gallery
Rock Tunnel
Francis
4 X 140 MW
550 feet
565 feet approx.
Vertical synchronous
HVA
tr.9
I
I
I
I
I
I
I
I
I
I
I
f
I
I
I
I
I
I
I
I
I
I
I
I
I .
I
I
I
I
TABLE 10 •. 3: . SUMMARY OF COMPARATIVE COST ESTIMATES
PRELIMINARY REVIEW OF ALTERNATIVE ARRANGEMENTS
(January 1982 $ Million$)
Item DC 1 DC 2 DC 3 DC 4
Land Acquisition 22.1 22.1 22.1 22.1 Reservoir 10.5 10.5 10.5 10.5 Main Dam 468.7 468~7 468.7 468.7 Emergency Spillwa)' 25.2 25.2 25.2 25.2 Power Facilities 211.7 211.7 211.7 211.7 Switch yard 7.1 7.1 7.1 7 .. 1 Miscellaneous Structures 9.5 9.5 9 • .5 9.5 Access Roads &: Site Facilities 28.4 28.4 28.4 28.4 Common Items -Subtotal 783.2 783.2 783.2· 783.2 Diversion 32.1 32.1 32.1 34.9 Service Spillway 46.8 53.3 50.1 85.2 Saddle Dam 19.9 18.6 18.6 19.9
Subtotal 98.8 104.0 100.8 140.0 Total 882.0 887.2 884.0 923 •. 2 Camp &: Support Costs (16~) 141.1 141.9 141.£1, 147.7 Subtotal 1023.1 . 1029.1 1025,4 1070.9 Contingency (20~) 204.6 205.8 2Pj.1 214.2 Subtotal 1227.7 1234.9 1 ~30.5 1285.1
En(ineering & Administration
12.5%) 153.5 154.3 153.8 160.6 Total , 381.2 1389.2 1384.3 1445.7
. .
l. .
I
L -·-
I ... '
§
i ...
I
DEVIL CANYON
ROCK OUTCROP MAP
~: ·c.-~
----
• {
f
LEGEND.
LITHOlOGY:
~ ARGILUTE I GRAYWACKE
.CONT~t
. . .
---UMIT OF OUTCROP
CONTOUR l,.&He:S:
TOPOGRAPHIC CONTOUR INTER\JAL ·
-SOFEET
'~~
-------~_,____ o 200 400 FEET
SCALE
t
FIGURE 10 .. 3 .
••
I
I
I
·I
·I
I
I
I
1:1
I
I
I
I
I
I
I
I
tl. -
LEGEND -
-LrrHOl00'1~
D. M6IUrtE ... ~
-
INFE'M£0 .CJIIIEN1JIT)QN Of' iEDOIN;/
. ·I"'l.JmmN, »P!UEEfT OP WHEJE.l«JJED
' lr.'~·~d'-~.(.:,.11"1 F'EUIIt ~~-~ ----WIIEJIE •,,,\, .. r ~Ttlt JltMI 10 FEET
-=mu~~lllf'"'wtPE
CONTAC'fS:-
---~'fOP OF WOQC
--liTHillliMIIC,OUHED w.tffE·-WEMED
S'I'RUCTURE:
-
• StEMt...WIJTH -~ WHERE GREATER THAN IU FEET
~· ~~tO=·!lHOWN W&DE
GEOPttYSICAL SURVEYs:
1'sw·l5 ~EIISl:COON ~SEISMIC ltEfRAC'nOM
SW·JS Hf11.SifAMfON a W11.S0H
$..110"13 IMO,~~YDEalNSUL"WfTS
SLii•U tiii,WOGOWMU-CLYDE CONSULTANTS
.SEISMIC V£LOCrTY CHANGE
. ·~ogo SEJS&Ic· VELOCITY IN FEE:T ·pp SECOND
BOREHOLES: .IH•I
--~ ·-ZCINE
• SH£AR
CH·I OSilft -~COM: Klfb!lfG
aM·& .UI DIAMOND COR£ IC*INif
AM ·Gt All .._.MOLE
OTH£R:
DC•I_. I ~IDISIECI:ION WIT~ ROLOGfe
"of -~ DC•I
~ CIEDLOCUC FEATURE ODCIIIII!JED IN. ~ !IEC110H ._,
Fl
MAFIC DICE J)E!CRII!f::D IN !EcntlN
Fn.SIC. DICE DE3CAIItD IN 'sa:TJOH
MA1N SPILLWAY.-
APPAOACH
~NEL
I
I
l
I
i
i
l
I
I •.r F
~~ICDIKE
/~JI!:tro> j t90W
tY'-4-AZIMUTH ._.......,)8()• OF SECTION . ·_.
lOOKING UPSTREAl!t
CREST OF MAIN OW t:L.I4M
DEVIL CANYON
GEOLOGIC -SECTION DC..;3
~OIVERS_ I_ON LJ TUNNEL
··-·--··
-~-
FIGURE 10 .. 4
I
·~
I.
I
I
I
I
I
I
.I
I
I
I
I.
I
I
I
I
I
Ms.D4.000-
BORROW SI~E 0
I
\ \ . . . -
~ ; -9,,~ ·"---. .. __/'
~
..
DEVIL CANYON. r
GEOLOGIC MAP · -
LEGEND
tmfoc...oGY:
o~.UNQI~. a. AAaUTE AND ~-
·.;~ ;.•. fELSIC DIU. WIDTH . ...,.__ -~ ~
·~'l'••!' _ ~-10 ~~n. -
-
IWte OICE, .WIDTH SHOWN WHDI£ -~
. T*H--10 FEET.
CC)NTAC'J'S:.
-lJMIT OF OUT~
~! -
-
SHEAR,·Wimt SOOWN. WHDE -GIIIIEATD "MMil;
· lO RET, varTICAL ~ DIP SHOWN.
J_ ..J SHEAR__ _ _ WIDTH LE_ SS THAN_ 10 FEET, !..a MB':. 1 r V£trrJCAL, EXT£NT WHEilE KNO'Iti
· --fRACtURE ZONE. WI01'H SHOWN WHERE
CftUTER _THAN 10 .F.£ET • 'VER'tJUL UfiL.DSi: ·
OIP SHOWN · .
~f JOWTS,INCUNED, VERTJCAL, ~ INC.~
-__ ( SETS I ANQ II ONLY • -EXC£PT .FOR OHJr,
-,J{)IH'!S .)
OTHER-~
oc-t t ! GEOLOGIC SECTION Loc.lTlON
.1.\DCJ•I JOINT STATIO_N
SCAl.E.
I
I
I
I
I
:I,
I!
I
I
I
I
I
I
I
I
I
I
:I
•~
' .
... j,_'
')
.~·
N
JOINT STATIOl\'1
DCJ·2
N•fOO
tEl
. \1 .. · ~.. r---. z + c:::.-a.../ Q-::j-
COMPOSITE .JQSNT PLOT
SOUTH BANK .
N•479
w
s
.~~~
...,t
~-.~··~~~==~~~N~
./.. · •• ·~-4 ·~-----···-~-~-NT-_ .... S1i-~TION---...._ · ·.··
OCJ-4 ~------DC.I-1 N•IOO
.. ~ ... ~ ' . . .. ..
(
• . . .• ~-----.U. SITI(A ··~. RIV;;___.-· .. ---~~~~---· ••• ~
' DC.J·2 --..____,..,r-----.. r---....__... _ ___,_ __
~···
J
,..
~01NT STATION
!)CJ-3
ti•IOQ
DEVIL CANYON
JOINT PLOTS
H
COMPOSITE JOINT PLOT
NORTH 8ANN
,. .. 714
w
NOTES
I. alNTIUIS ARE PEftef;NT OF oiOIJiil'S Pfllt 1~ OF AAEA.
CONTCUR tNT'EJM'L.-l,~,T,IO,IIS.a~.
2. N EQUAL! NUMBER OF DATA POIITS.
3. COIIIIPOSITE PLOTS INCQIIHIORATE .AU.. ..KliNT O.TA. .!OINT
STAnCH !ILOTS · ®NTAIN DlTA •f'RQii SPECIFIC .JOiNT
STAnoNS. .
4. FOR <!OINT PLOTTING METHOO SEE
FIGURE 10.6
" -·~
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
10!50
1000
0
TYP. TUNNEL
SECTION
I-PRESSURE TUNNEL
900 ~------~--------~--------~------
880
30 35
rOR-:36,000 CFS AT DEVIL CANYON
DEVIL CANYON DIVERSION
HEADWATER EL./ TUNNEL DIAMETER
40
FIGURE to.aiAilliJ . ---------------------------------:'( -•' ~
I
I
I
I
I
I
~.
.
I
. I
I
I
I
I
I
I
I
I
I
~~----~------~------~--------------~~~~·
. l075
-
1050 I
-...,: 1000 u. -z
0 -...
~
IA.I
..I
&&I
2
<(
Q
950
~
~
,''
900
v v / ./
/ v
v /
I
I .
. I
'
I
\,1
J
. I I t ' -I
l
I l
'
I
2 3 4 5 6 1 8 9 10
CAPITAL COST # x 10 6
DEVIL CANYON DIVERSION
UPSTREAM COFFERDAM COSTS
FIGURE 10.9~~~~mt .
l'
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I -.,
I
s-
. ~
..
;I'
~
8 .)
......
~ . -U)
0 -
.!!(
t()
lri 7 -z U) -
~
0
TUNNEL (.) /-ci -0
6 ..
~--~
1-~
..;~ -5 ~--------+---~~~~--------~----~--r---~--~ u:>o I ~
t X
M -~
U)
0
--~ -+~COFFERDAM
. . .
~ 4 ~--------~--~--~~--------~------~r--~--~~
3 ~---------~------~~--------r-------~r-------~
2 ~-------~r---------~'~-------r----------r---------
0
TYPICAL
TUNNEL
SECTION
COFFERDAM CC STS
INCLUDE DAM l EIGHT
r::' AE:IOVE HEAC WATER
ELEVATION FO£; FREEBOARD
0 ----------~-----------------~--------._~----~ 0 20 25 30 35 40
TUNNEL DIAMETER (FT.)
DEVIL CANYON DIVERSION
TUNNELS COFFERDAM COST I TUNNEL DIAMETER
FIGURE 10.10 [il
I
I
I
.I
I
I
I
I
I
I'
I
I
I
I
I
I
I
I
••
2Q
18
16
14
-U)~ -
~ 12 -t-
CJ)
0
'-?
10
6
4
0
TYPICAL
I TUNNEL
SECTION
r--
0
.. .
-
'
I
' l
I ~ l
I
------~
i
I
I
' f
I .
~·
17' ~-
i .)/
20 25 30 35 40
TUNNEL DIAMETER (FT.)
DEVIL CANYON DIVEHSION
TOTAL: COST l TUNNEL DIAMETER
. FIGURE: 10.1( lii}. ''
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
. ..
11 ... SELECTION OF ACCESS PLAN:
This section describes the process of formulation :; . selection of an access
platl for the Susitna Hydroe1ectric Project. The methodology for comparison of
alt~rnative plans is outlined, and an evaluation of each basic plan is pre-
sented~ considering schedule, costs, and biological and social impacts. The
selected plan is described in detail, and recommendations for measures to reduce
impacts are presented. ·
Engineering studies conducted on the alternative routes consisted of development
of design criter~a, layouts of the alternative routes, preliminary field inves-
tigations, cost r.:~5timates of constructing the alternative routes and logistics
costs in transpor1:ing supplies and materials to the dam sites. Environmental
studies included Identification and evaluation of biological impacts for each of
the alternative .·cutes. The environmental studies included field investigations
and assessments for all the alternative routes. Social or socioeconomic studies
included a public participation program among the various studies. Public con-
cerns and pre-,-~rences, particularly those of the sector that wou1d be impacted
the most directly, were solicited and fully considered in the evaluation~
The evaluation of the alternative plans included development of selection cri-
teria, comparisons of the alternative plans, identification of conflicts among
the alternatives in the evaluation criteria, comparison of the conflicts in the
criteria, and the tradeoffs made in the evaluation.
11.1 -Background
(a) Existing Access Facilities
As discussed elsewhere, the proposed Devil Canyon and v!atana sites are lo.-
cated approximately 115 miles northeast of Anchorage and 140 mi 1es south of
Fairbanks.~ The Alaska Railroad, which links Anthorage and Fairbanksl
passes wi'thin 12 miles of the Devil Canyon site at Gold Creek. The George
parks Highway (Route. 3) parallels the Alaska Railroad for much of its
route, although between the communities cf Sunshine and Hurricane, the
Highway isorouted to the west of the Railroad, so that Gold Creek is situ-
ated approximately 16 miles south of the intersection of the railroad and
highway. A portion of the highway between Lane and Hurricane, known as the
Parks Highway, passPs through Denali State Parkfl At Cant\'/ell 51 miles
north of Gold Creek, the Denali Highway (Route 8) leads easterly approxi-
mately 116 miles to Paxson~ intersecting the Richardson Highway at this
point::
Tb the south, the Glenn Highway (Route 1) provides the main access to
Glennallen and intersects the Richardson Highway which leads south to
Valdez.
(b) Modifications to Plan of Study
The ori gina 1 P,OS proposed that a single route \'IOUl d be selected by May 1981
to be followed by detailed environmental investigations of this route.
0
Early in the study, three main access corridors were developed. Cons idera ...
tion of these plans on the basis of information received, c<1mments and con-
cerns from various state agencies and a recommendation from the Sus itna
Steering Committee, led to a decision to assess three ,atternative routes in
more detail throughout 1981 and recommend a selected route later in the
yea\r. Accordingly, this assessment included environmental studies, engi-
neering studies, aerial photography, drilling and geologic mapping of all
three alternatives rather than the single route initially envisaged.
11 ~2 -CJbject ives
The finally selected access plan must allow for the efficient and timely under ...
taking of construction and maintenance activities in order that the Susitna
Hydroelectric Development can be completed and electric power be reliably and
continuously provided to the Railbelt area of Alaska.
In meeting this basic objective, several specific objectives were developed as
a basis for· evaluation of the alterna .. ive access rt;lr~~s. These objectives are:
-Tc: allow tne construction of the Susitna project to pro('eed on a schedule that
would supply the necessary power to the Rail belt Area .~f A1aska when needed;
-To minimize cost. including construction costs of the access facilities
themselves, logistics costs for support of construction activities as well as
the logistics of subsequent operation of the completed project;
-T~ allow for ease of operation and maintenance to ensure reliability in-t~e
power supply;
-To minimize adverse biological impacts;
-To accommodate the preferences of local communities; and
-To accommodate the preferences of Native landowners;
11.3 -Approach
The approach uti ·1 i zed to arrive at an access recommend at ion vtas an adapt at ion of
the generic plan formulation and selection methodology described elsewhere in
the report. ·The methodology as specifically applied to selection of the access
plan is presented graphically in Figure ll.l.
To aid in understanding the selection process and the various studies conducted,
the following definitions are provided:
(a) Corridor
A strip ~f land generally 2 mile~ or greater in width leading between two
points -or areas.
11-2
,·}
-·
'I' i1 ~-. -_:
. :/"': .:
cl
I
I
I
I -
I
I
I
I
I .
I
I
:.
I
I
I
;I
-· '.1
I;
I
I
I
1:
I
I
I
I
I
I
I
I
I
I.
I
I
I
I
(b)
(c)
Route
A strip of land generally l/2 mile or less in width~ leading between t\'/0
points.
Segment
Portions of a route which when combined constitute one alternative route
between two points.
(d) Alternative Route
One of several routes which will be evaluated between two points.
(e) Plan
An access plan which will involve one or more or a combination of existing
and new al:ernative routes. The plan will also define the logistics in-
volved in the transportation of supplies and materials.
11.4 -Corridor Selection and Evaluation
The first step in the selection process involved identification o'f the three
general corridors shown on Plate 28 and described below:
Corridor ,,
1
2
3
Descri gti on
From the Parks Highway to the Hat ana site via the
north side of the Susitna River
From the Parks Highway to the Watana site vi a the
south side of the Susit:1a River
From the Dena 1 i Highway to the Hat ana site.
These corridors were selected based upon the use of existi1'1g transportation
faci 1 ities within reasonable proximity to the Watana and Devil Canyon sites~
A general environmental analysis was undertaken for each corridor.. The majcr
environmental constraints identified within each corridor are potential impacts
on the following:
(a) Corridor 1
-Fishery resources in the Susitna and Indian Rivers;
-C1 iff-nesting ·raptors · near Portage Creek and Devi 1 Canyon;
-Fur bearer habitat near Portage Creek and High Lake;
-Moose habitat on the Susitna River; and
-Caribou habitat between De vi 1 Creek and Deadman Creek.
Tl-3
c. ·. '1
1
I
I
i
.. ,
(b) Carr idor 2
I ) t c .
-Fishery resources in the Susitna and Indian Rivers;
-Cliff-nesting raptors near south side of the Susitna River;
-Waterfowl habitat in the Stephan Lake-Fog lake areas; and
-' Furbearer habitat in the Stephan Lake ... fog Lake areas.
Corridor 3
~ Caribou calving ar~a near Butte Lake;
~ Fur bearer h ;;bit at; and
-Some waterfowl habitat.
In addition, increased access will cause various impacts which are common to all
corridor·s. ·Archaeological resources could pose a constraint; a_t this time, the
locations of such resources that may exist are unknown. · ,.
Finally, socioeconomic impacts will vary both in magnitude and areas of concen-
tration, depending upon which access route or combination of access routes is
selected, and whether a road or railroad is used. With the socioeconomic
assessment of access schemes 5 there is more concern with the origin and type of
access than with the actual route, because these will affect communities
throughout the south-central part of the State.
With a road from the Parks Highway to the damsites (Corridors 1 and 2), effects
generally would be concentrated on the western side of the prcject area. An
easily accessible road corridor would provide for transportation of construction
materials, equipment, and labor as well as post-construction uses of the upper
Susitna basin' (such as recreation). The impact of a railroad from the S<;lme side
.would likewise be conceritrated on the western side. However, in every socio-
economic category, impacts would be the same or less than with the road. The
single exception would be in rail industry ~ctivities, which would experience
major changes.
With a road constructed from the Denali Highway to the damsitt~S (Corridor 3),
impacts along the Parks Highway-Alaska Ra~lroad corridor would depend upon
whether materials were to 1Je shipped by road or rail to Cantwell before being
transported along the. Denali Highway to the access road. Impacts would occur in
· the Cant we 11 area, however, regardless of transport at ion mode.
11.5 -Route Selection and Evaluation
Following identification of three major corridors, a number of access routes
were selected and evaluated based on engineering and economic criteria .. The
selected routes were then modified on the basis of an environmental· analysis.
'
(a) Design Criteria
Construction of the Susitna project will require a dependable safe and
efficient access route suitable for transporting personnel, consumable sup-
plies and large pieces of equipment for an extendea period, in adverse
weather conditions.
11-4
'I
I
I
I
I
I
I
I
I
I
I
I
I
.I
~·
I
·-
:I
I
~
·-1
I
I
I
··I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
(b)
(c)
The preliminary design criteria adopted for access road and rail alterna-
tives were selected on the basis of similar faci1tties in remote projects
of this nature. Basic par·ameters were as follows;
Maximum Grade
Maximum Curvature
Design Loadin~
-During Construction
-After Construction
Access Road
6 percent
·s degrees
Railroad
2.5 percent
10 degrees
80k per axle & 200k total not appl icab1e
HS-20 E-50
Following corridor definition., various segments that met engineering cri-
teria were mapped. These segments were then jointed to form alternative
routes which were compared on the basis of:
-overall length;
-average grade per mile;
-average deflection per mile;
Economic Criteria
-For the early stages of corridor and route selection, the alternatives were
compared on the basis of total centerline length of route, with minor ad-
justments for average grade and curvature. Preliminary capital costs for ·
construct ion ~"/ere estimated at $1,250,POO per mile.
Results
A total of 16 segments, combined into 30 routes were ldentified within the
three corridors. The alternatives identified as being most favorable in
terms of overall length, grade and alignment are as follows:
Type
over a 11
1 ength
Average
Corridor 1
Parks Highway
to Watana
north side
Road
64.9 miles
Grade · 2.4 percent
Deflection
per Mile 7° 06'+
Corridor 2
Parks Highway
to Watana
south side
Road
66.5 miles
2.2 percent
11-5 .
Corridor 3 Corridor 2
Denali Highway Parks Highway
to Watana to Watana
south side
Road Rail
39.1 miles 58.0 miles
1.3 percent 0.5 percent
1° 30'+
, .,...,..,...,.....,,.._,..~~c ... s·c . .-•. 1 .. ·. ,· .,. . ' ' . . ·.··:·.··.··
. ' ' . ";:;
' '
••••
(d) ·. Environmenta 1 Influences on A 1t~~rnati ve Routes
After the engineer·ing and etonomjc assessment identified the 3 roads and 1
rail route descri'Jed above» an initial screening was made which resulted in
several refinements to the alteY'native routes under consideratjon. A major
refinement i nvo 1 ved the deletion of a 1 arge portion of the road access cor-
ridor to the Parks Highway on the north side of the river (Corridor 1)"
The segment connecting the hi~1hv1ay and the Devil Canyon site routed around
Portage Creek was deleted mainly on the basis of potentially severe envi-
ronmental impacts on anadromous fish~ fur bearers, and raptors. The topo-
graphy in the Portage Creek area is furthermore such that the overall
length of road necessary to meet the established criteria was excessive.
In addition, the construction of the segment would be extremely difficult
due to the predominance of steep sidehill cuts of considerable height.
Another major refinement to the corridors was the routing to the west of
the northern portion of the Denali route (Corridor 3). This routing was
advocated on environmental grounds in an attempt to reduce potential im-
pacts on the caribou subherd calving area near Butte Lake.. A final refine-
ment consisted of realignment of the portion of the Corridor on the south
side of the river (Corridor 2) in the Stephan Lake-Fog Lakes area to reduce
potential environmental impacts to fur bearers and \~aterfow1 • ·
The main routes within the corridors remaining after the initial screening
are shown on Plate 29 and are briefly destribed below:
( i )
( i i}
(iii)
Parks Highway. to De vi 1 Canyon
This route follows the existing portion of the Alaska Railroad be ... ,
tween Gold Creek and the intersection of the railroad with the Parks
Highway just south of Hurricane. This route passes through Chulitna
Pass from the Parks Highway and then parallels the Indian River to
Gold Creek. The existing river valley is sufficiently wide to ac-
commodate a road. From Gold Creek to Devil Canyon, the route lies
south of the Susitna River,_ paralleling the river on a high ridge.
Devil Canyon to Watana -South Side of Susitna River
This route generally parallels the Susitna River and traverses east-
west from Devil Canyon to Watana. The topography is mountainous and
the route i nvo 1 ves the most difficult construction of ·the three
routes, requiring a number of sidehill cuts in rock and soil. This
route also includes the environmentally sensitive Stephan Lake and·
Fog Lake areas.
Devil Canyon to Watana -North Side of Susitna River
This route generally parallels the Susitna River and traverses east-
west from De vi 1 Canyon to Hat ana. This route is mountainous and in--
eludes terrain at the highest elevations of a11 routes, however con-
struction of the road would not be as difficult as the route between
the damsites south of the Susitna River.
ll-6
-" I
I
I
I
I
I
I
·I
I
I
.I
I
:1
:1
I
'1.' . '
I
I
I
I
I
I
I
I.
I
••
I
I
'I
I
I
I
I
I
I
I
I
I
(i v) .Denali Highwa,x to Watana
This route connects the Denali Highway with the t~atana site and runs.
in a north-south direction. This route is the easiest to construct
of the alternative routes. The terrain is relatively flat with a
few wetlands involved. This route would not require .any major
bridges~
11.6 -Description of Basic Plans
From the three routes remaining after the initial screening, eight pl.ans were
developed. These plans were evaluated in more detail than originally planned in
the original POS, as a result of information and assessments conducted during
the study program, the concerns of state agencies, and recommendations of the
Susitna Steering Committee (refer to Appendix D). The additional investigation
and evaluat·ions consisted mainly of environmental field work, geologic mapping
and subsurface borings •
' <
The plans are presented below and are also shown schematically in Figur·es 11.2
through 11 .. 5.
(a) Plan 1
This plans utilizes a roadway from the Parks Highway to ~latana Dam C.\long
the South side of the River. This access plan is based on materials such
as cement and steel being brought into the State through the port of
~~hittier. Food and other camp supplies would be imported through Anchorage
via container, and fuel directly from Kenai to Al)chorage via existing pipe-
line. These materials and supplies would then be carried by rail to a
railhead and storage area at Gold Creek. At Gold Creek, materials \~auld be
transferred to trucks for transport by road to the site. Other materials
and supplies \1/oul d be transported by truck from the Parks Highway. An a 1-
ternative for fuel supply would be rail haul from the refinery at North
Pole:~ Alaska.
(b) Plan 2 -All Rail
This plan \vou1d serve both damsites by a rail line. This alternative would
essentially preclude public access. Construction planning for this mode of
access would be based on trains being broken down and cars dropped on the
siding at Gold Creek. An engine and train crew would be stationed at Gold
Creek which would allovl shuttle cars from Gold Creek to the project s1te on
a daily basis. Passenger rail service would be required daily. If public
access is desired after construction, the rails could be removed and the
road bed graded into a single lane road with turnouts.
(c) Plan 3
This plan envisages the use of a combination of rail and road transporta-
tion. Construction activities at Watana would be served from a railhead
and storage area at Cant\'lell by truck across the. Denali Highway and a1 ong. a
11-7
I ,j
~ . . i
I
1
I newly constructed road from the Dena.l 1 Highway. Construct ion at De vi 1 Can-
yon would be served by toad from a railhead at Gold Creek and road access
from Gold Creek to the Parks Highway. This plan does not include a connec-
tion between the two darns.
(d) Plan 4 :1 j
This plan serves Watana by truck from a railhead at Cantwell ana Devil Can-
yon by rail from Gold Creek. In the plan, there is no connection between
dams.
(e) Plan 5
This plan ser:ves both dams by road from a railhead at Gold Creek. The
route is located on the south side of the river to Devi 1 Canyon with i
major bridge downstream from the dam site, then follows the north side of
the river to Watana. There is a road connection to the Parks Highway from
Gold Creek.
(f) Plan 6
This plan is identical to Plan 4 except that a service road for maintenance
purposes is inc ·1 uded on the north s 1 de of the river between the two dams.
(g) Plan 7
This plan is thf; same as Plan 3 except that a service road would be pro-
vided along thB north side of the river as in Plan 6~ -
_(h) Plan 8
This plan is the same as Plan 5 except there _is no road connection to the
Parks Highway~ A. newly constructed road would service Devil Canyon from
Gold~ Creek on the south side of the river. A major bridge would be re-
quired downstream of Oevi 1 Canyon and a new road on the north side of thE:
river would connect ·the two dams~ This alternative plan precludes public
access.
I
I
••
I
I
I
I
I
I
11.7-Additional Plans I
Following selection and evaluation of the eight plans described above, presenta-
tions were made to the Alaska Power Authority and the Susitna Hydroelectric Pro-I
ject Steering Committee. These present at ions and subsequent discuss ions re-.· .
sulted in the addition of the three plans described below.
(ill Plan 9 I
This plan is the same" u-s Plan 8 excet:t access between Gold Creek and Devil 1 ___ ...
Canyon is by rail along a similar route, and the railhead is located at • -
Devil Canyon instead of Gold Creek.
"
I
I
11-8
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I-
I
(b) Plan 10
-
This plan is identical to Plan 9 except that ~he road connecting Devil Can-
yon and Watana is on the south side of the Susitna River.
(c) P1 an 11
This plan utilizes a railhead at Cantwell~ road access via the existing
Dena 1 i Highway, a road from the Dena 1 i Highway to Watana and a road from
Watana to Devil Canyon on the north side of the River.
These plans are shown schematically in Figures 11.6 and 11.7.
Plans 9 and 10 suggested by the Steering Committee as a means to reduce accessi ...
bility to the area thus avoiding the introduction of adverse environmental im ....
pacts into the Susitna Basin.
Plan 11 \>Jas added as a possible way to provide access from only one area while
also alleviating the socioeconomic impacts the west side communities would feel
from any access road from the west.
11.8 -Evaluation Criteria
The specific objectives of the selected access plan are described in Section
11.2. The criteria used to assess the degree that any given plan satisfies
these objectives are described in the following paragraphs.
(a) Construction Schedule
It is essential that the selected access plan be adequate to meet the over-
all project scheduling requirement's. The load forecasts de~cribed in
Section 5 together with the examination of the existing system· and future
generating options indicated a requirement for first power from Watana in
1993. A delay in the on-line date by one year would mean that another
source of fossil fuel generation would have to be constructed, combined
\'lith retirement of some fossil fuel generation a year later, into there ..
serve category. In terms of present 1110rth, a delay of one year would in-
crease the cost of the project by approximately $50 mi 11 ion.
Analysis of the construction schedule requirements for rlatana demonstrates
that all-weather access route to the site is required by mid-1986 if the
on-line date of 1993 is to be maintained. For the purposes of these
studies, it has been assumed that an FERC licence to construct the project
will be received at the beginning of 1985, and the start of permanent work
on the project Will coincide with this date. In order to meet all the mid-
198"6 requirements, it is obvious that an access route to the site would
have to be constructed v1ithi n approximately 18 months.
A preliminary evaluation of the construction period for completion of the
access plans is presented below.
ll-9
Plan
1
2
3
4
5
6
7
8
9
10
11
Origin for.Watana Access
Parks Highway
Gold Creek
Dena 1 i Highway
Dena l i Highway
Parks Hi ghvray
Denali Highway, Go1 d Creek
Dena 1 i Highway, Parks Hi gh\'tay
Gdl d Creek
Gold Creek
Gold Creek
Denal i Hi gh~-Jay
-Approximate Construction
Period (,years)_
3-4
3-4
1
1
3-4
1
1
3
3
3
1
It is apparent from the above that only Plans 3, 4, 6, 7 and 11 could be
constructed within the 18 month peri o.d required to maintain the over a 11
project schedulee Since this would severely limit the selection process, a
scheme was developed to provide initial access to the Watana Site within
the framework of regulatory and scheduling restraints. This scheme, des-
cribed in more detail in'Appendix D involved construction of a pioneer road
to Watana from either Gold Creek or the Parks Hi gh\'lay. The pioneer road
would consist of a gravel based road with period passing turnouts and \WUl d
be constructed on existing ground insofar as possible to avoid significant
cuts or fills. Temporary floating Barley bridges would be used at river
crossings, replaced by ice crossings in the winter. The analysis indicates
that the pioneer road scheme will be sufficient to provide continuous
access to the site within J8 months, and will be sufficient to support
con!:t,ruction activities. until the permanent access route is completed ..
Certain additional licensing and permiting requirements are associated with
this scheme; these are discussed in Section 11.12.
The pioneer road scheme can be implemented with Plans 1, 2, ~' 8, 9 and 10,
therefore all 11 plans can be considered equivalent in terms of their abil-
ity to meet initial project requirements.
(b) Construction and Logistics Costs
For the purposes of this evaluation, construction costs include the cost of
constructing the access facilities, adjusted for any differences in ·c~st of
constructing the Susitna project itself which relate to the particular
access plan under construction. Logistic costs are +.he costs associated
with transporting, labor, fuel, equipment, materials and supplies to con-
struct the two power developments.
(c) East of Operation and Maintenance
This criteria relates to the relative ease of operation and maintenance of
the two developments after construction is complete. Initial planning en-
visa.ges operation of both developments from Watana for several years aft.er
Devil Canyon is brought on-line,. after which time, both projects will be
11-10
til ·l ~i . -_: '
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
••
I
I
I
I
~ .•
I
I
I
I
I'
I
I
I
I
I
I
I ..• ;
operated remotely from a ~entra1 location. Maintenance of. two projects uf
this size and complexity wi 11 obviously be an important consideration· •.
Duplication of maintenanc~ facilities and staff at both. sites \~auld tesult
in a substantial incr~ase in the annual costs of the overall development.
Th~'2 most economic scheme) given the sequence of development$ would be to·
est~ablish an operation and maint~;nanc~ facility .at Watana, \.Yith a reliable
means -of access to De vi 1 Canyon 32 mfles dovmstrearn,~ In this regard,
acci'eSS plans With a road connection between the tv1o sites have been
eva·luated as being superior ir1 terms of ease of operation and maintenance
than plhns without a road connection.
(d) Flexibility in Construction Logistics and 'Transportation
This criterion is used to evaluate the extenf to which an access plan con-
tributes to the maintenance. of a reliable and flexible logistic support
systt:m duri.ng construction of Hatana and De vi 1 Canyon.
For the Susitna project, a fundamental consideration is whether or not to
provide a road connection to a major highway. For this evaluation, the
following alternatives have~een considered:
'
- a road connection either to the Parks Highway or Dena 1 i Highway;
-rail access only from Gold Creek;
Plans 1 through 10 described in Section 11.3 all include a railhead and
storage area at Gold Creek. Accorindgl_y, plans incorporating a road con-
nection with the Parks or Denali Highways obviously provide· greater flexi-
bility and reliability in case .of a transportation description invo1'ling
the Alaska Railroad, compared to plans \~ith "rail only 11 access. Specific
considerations are as follows:
-Any breakdm·m in the rail system would result in a loss of all ground
transportation, in the absence of an alternative road system •. The in-
creased risk of delays has an associated cost penalty. An analysis has
been undertaken· to quantify the risks associated with rai 1 access onl..v.
The methodology for this risk analysis is presented in Appendix D.,·
·~
-The availability of two possible manes of transportation will undoubtedly
be reflected in lower and more competitive bids for constructi'onj supply
and service contracts, si nee contractors waul d othervri se include some
contingency to cover trnsportation disruptions. Although significant,
this aspect is difficult to quantify.
(e) Environmental Considerations
Exclusive of socioeconomic considerations, the objective is to develop an
access plan which minimizes adverse changes to the natural environment.
The cri ter·i a used to assess the degree to which any plan meets this objec-
tives are described below.
11-11
0
{ i) Effects of Bi.9. Gam~
A primary concern associated with' the selection of an access plan is
the potential effect on the Nelchina caribou he~d, specifically the
subpopul ation of approximate_ly 1,000 animals that inhabit the north ...
western sect i-on of the .. Up.p£r __ Susitna. Basin.. The impacts of hunters
· on moose. and bear at~e also cons ide red but as secondar-y concern_s.
These impacts can be greatly 1 essened by se 1 ect ing a route other
than the access from the Denali Highway. •
( i i) Effects of Fisheries
In the case of resident fisheries, there are relatively isolated
lakes (Butte Lake, Big Lake) and streams in the-northwestern section
of the Upper Susitna Basin, and the Fog Lakes area that would re-
ceive additional angling pressure if road. access was provided.
These impacts can be lessened by avoiding access from the Denali
Highway and the route on the south side of the Sus itn a River between
the dams ites. ·
Since Devil Canyon acts as a natural barri.er to anadromous fish
migration, there is no concern regar ... d ing the effect of improved
access on this resource upstream of Devil Canyon. However, Indian
River and the Susitna River below Portage Creek,-are important for
salmon. Any access plans that follow or -cross these rivers could
affect salmon directly through habitat disruption (i.e., sedimenta-
tion) or indirectly through increased fishing pressure. These im-
pacts could be les.sened by avoiding road access paralleling the
Indian River. ·
(iii) Effects on Furbearers
Wet 1 ands, important to furbearers, have been identified between the
Parks Highway and Gold Creek, near Deadman Mountain, near Deadman
and Big Lakes and the Upper Deadman Creek. In add it ion, the Fog
Lake -Ste-phan Lakes wetlands complex is a valuable furbearer habi-
tat. A red fox denning complex has also been identified south of
Deadman jvfountain. Any access road crossing through these areas has
the potential for negative impacts on fur bearers. Impacts on fur-
bearers would be least by selecting access from Gold Creek to Devil
Canyon on the south side of the Susitna River and on the north side
of the River between the dam sites. ·
(iv) Effects on Birds
Heavily forested areas between the Parks High\·lay and Devil Canyon
along riverbanks are productive avian habitat. Construction through
these areas ~tlould disturb this habitat.
(v) · Effects on Wilderness Setting
The Upper Susitna Basin is presently in a state of wilderness to
11-12
I
I
I
I
I
I
I
I
I
I
I
I
I
I ,,
'I
I
I
I
·.-
~ ••. --,_. c-~ •h
,.: '
i'~
I .
I
I
I
I
I
:1
I
I.
I
I
I
I
I
I
I
{f)
-
semi-wilderness. Although continued intrusion with ATV~s from
Dena 1 i Highway, potential deve1 opmeot of native 1 a nds and the estab-
lishment of the Indian River remote and disposal site have the po-
tential of changing the character of sections of the basin, improved
public access and construction of the Susitna Hydroelectric Project
will produce.~_major alteration in the remoteness of the area. Nat_-
ural resource agencies an~ the local public have expressed a desire
to maintain the status quo to the maximum extent possible. People
from the urban centers of Anchorage and Fairbanks have expressed a
stronger desire to pro vi de road etc cess and open the area for recrea-
tion development. The status quo of the area would be retained to
the greatest extent by pro vi ding only rail access to the damsites ..
(vi) Effects 6"n Archaeol ogi ca 1 Resources
,· Archaeological resources are likely·present along all-access routes.
The route from Denali crosses a substantial area of high archaeolog-
ical potential. The thin soil and lack of vegetation result in a
high potentia 1 .for impacts to resources a 1 ong this route. The other
access routes are believed to be less sensitive. Avoidance of the
Denali access link lessens the probability of the disturbance of
archaeological sites.
Social Considerations
(i) Native Landowners
Native organizations have selected land surrounding the impoundment
areas and south of the Susitna River between De vi 1 Canyon and \~atana
damsites. To allow for increased opportunity to develop either
lands on the south side of the river, the native landowners have ex-
pressed a strong desire to have a Susitna access road along the
south side of the river from Watana to Go 1 d Creek, ultimately con-
necting to the Parks Highway. It is considered that the basic
native preferences waul d be met by pro vi ding road access to both
damsites.
(ii) Local_ Community Preferences
Since the local communities are likely to receive many of the dis-
benefits with. few of the benefits of a Susitna development, the ob-
jective to accommodate local community preferences has been included
in the access plan selection process. The criteria used in assess-
ing the degree to which this objective is met is divided into four
areas due to the differences in community preferences (refer to
Appendix D).
-Cant\'ie 11
The community of Cant\'lell desires economic stimulus and is in
favor of the economic .changes that could result from having a
major construction project in the area. The desired stimulus
could be achieved by providin.g road access to the Dena)i Highway
\'lith a rail head at Ca ntwe 11.
11-13
-t
.:. Morth, of Talkeetna
The communities along the railroad north of Talkeetna are opposed
to deve\ opment in the area and strongly prefer a maintenance of
the status quo. These communities have expressed a desire for
-rail access only, although existing conditions in these communi-
ties would probably be disrupted least with a plan involving road
access only from the Dena 11: Hi gh\~ay.
This conclusion is based on the consideration that if rail access
only is provided, the practicality of a self contained family-
status community at either of the sites would be greatly
diminished and a single-status on1y camp facility would like.ly be
established. If this \'/ere to be the case, workers wou1 d tend to
locate their families in the nearest communities, thus increasing
the impacts in these communities.-
-Ta 1keetna/Trapper Creek _
Although attitudes are somewhat divided, the-majority of residents
of the communities of Talkeetna and Trapper Creek prefer a main-
tenance of the status quo. This can be most easily accomplished
by providing access via Denali Highway.
-Wi llow/Wasi la Area
0
The residents in this area are more in favor -of economic develop-
ment than in other areas.
-Indian River Land Disposal· Sites
In 1981, a total of 75 remote state land parcels \-Jere awarded by
1 ottery in the Indian River area. Of these, 35 ~vere staked in the
summer of 1981. The 35 land holders were contacted by letter
through APA Public Participation Office. Of the 12 responses re-
ceived to date, 11 favored retention of the remote status of the
area and one favored road access to the area. This area would be
most affected by road access fi om the Parks Hi gh~;1ay and 1 east a f-
fected by access from the Dena 1 i Highway.
(g) Agency Concerns
" Correspondence, meetings and interaction with the various agencies involved
with the Susitna Hydroelectric Project Steer·i ng Committee occurred through-
out the study. Agency comments have been considered in the evaluation.
T~e concerns of the agencies have generally related to environmental
issues, with the emphasis on biological and land use impacts .. Therefore,
evaluation in terms of the environmental criteria discussed previously is
considered to generally include agency concerns. It is considered that the
resource agencies favor a rail only access plan with a major opposition
towards road access from the Denali Highway.
11-14
9.'
I
I
·-1-·, ' .
I
I
J·
I
I
I
~~·c.-
• .
I
I
I
I
1
I
I
I
I
• ·-. ,~...{&
.... I
I
I ~ ... ,
I
••
I
I
I .-·
I
I
I
I
I
I
I
I
I
I
-~ (I
{h) Transmission
Access plan selection has been coordinated with the transmission line
studies. The transmission line studies to d,1te have identifi.ed two corri-
dors, one north of_the Susitna River and one south of the Susitna River
from Watana to Gold Creek. Although corridors run along the-river, ther·e
is flexibility to expand the corridor to include the access road when the
decision on which access route will be constructed is made. Due to more
stringent engineering criteria of lines and grades for road alignments~ it
was decided that the selection of a transmission line route would occur
subsequent to the access road selection.
The results of the transmission studies has also established that if the
northern Denali acc~ss route is selected, the transmission line would not
follow that route due to excessive cost and adverse visual impacts.
( i) Recreation
In meetings, discussions, and evaluation of recreation plans, it has become
apparent that the various recreation.plans are sufficiently flexible to
accommodate any access route selected. No single route was identified
which had superior recreational potential associated with it. Therefore,
compatability with recreational aspects vJas essentially eliminated as an
evaluation criteria.
11.9 -Evaluation of Access Plans
The 11 access plans evaluated on the basis of the criteria described in Section
11.8 have bee.n grouped in accordance with the following categories 1n order to
clarify the presentation.
Category
Plans providing access from both
Parks and Denali Highways
Plans providing access from Parks
Highway only
Plans providing access from
Denali Highway
Access from Gold Creek only
Plan Numbers
3 and 7
1 and 5
4, 6 and 11
2, 8, 9 and 10
In addition to the specific considerations outlined in the following paragraphs,
a major concern for all access plans is the creation of access to areas pre-·
viously inaccessible or relatively inaccessible. Such access would lead to im-
pacts to furbearers through increased trapping pressure and to big game through
hunting pressure. In addition, detrimental effects could occur to all \•dldlife
· through disturbance and destruction of habitat by ATVs. Cultural resources
would also be vulnerable to amateur collectors and ATV traffic.
11-15
(ej · Access to Both Parks and Denali Highways (Plans 3 and 7_l
· (i) Cost
The costs of the 11 alternative access plans are summarized in Table
11.1 .. Given the preliminary nature of the fielddata used to
develop construction costs, construction cost differences of less
than $10,000,000 (approximately 5 to 10 percent of the cost of the
alternatives examined) should not be considered significant.
Maintenance costs are a small portion of construction costs, and
large variations in maintenance costs will have negligible influence
on over a 11 costs. The 1 ogi st ics costs are based on current freight
ra-tes and vary by less than 10 percent for all plans. The personnel
shuttle costs and contingency risk costs are necessarily approx;.:.
mate but are adequate for comparison purposes. When comparing the
total costs,-the plans were considered equal if the total costs were
within $40 million, and a definite cost advantage was considered if
there was a $50 million difference.
On the basis of the foregoing~ Plan 3 is comparable to the minimum
cost alternative of any of the plans. Plan 7 has approximately a
$60 million cost disadvantage compared to P~an 3.
(ii) Ease of Operation, ~laintenance and Construction Flexibility
Access Plan 3 does not meet the ease of operation and maintenance
criterion because it does not have a connecting road between \~atana
and Devil Canyon. Access Plan 7 does meet the ease of operation
criteria by having a connection service road between the t~t1o sites.
Plans 3 c.nd 7 both satisfy the criteria for flexibility for con-
struction 1 ogi st i cs and transportation by having a road access con-
necting to a major highway.
(iii) Biological
The primary biological concerns for these two plans relate to the
effects the road would have on furbearers, big game, and cultural
resources ..
A roadway from the Parks Highway \'lould cr'oss productive furbearer
wetlands habitat between the high~'lay and Gold Creek .. The Denali
segment of both these plans also crosses aquatic fur bearer habitat
near Deadman Mountain~ Deadman and Big Lak.es, and Upper Deadman
Creek. In addttion, a red fox denning complex south of'Deadman
Mountain within one mile of the proposed road is likely to be af-
fected.
The primary concern relative to big game for both these plans is the
Denali segment, whi~h would pass through an area that has frequently
been used by ~ither major portions or all of the Nelchina herd and
includes the calving and summer ranges of the northwestern subgroups
11-16
I
I
I
I
I
I
I
,I
· ..
I
I
I
I
I
I
I
I
' I
"" .;.
f.
I
I
J
I
I
I ,••
I.
I
I
I
I
I
I
I
I
I
I
I
I
of the Nelcttina caribou herd. The r·oute also lies across the late
summer migration route of cari. bou moving toward Butte Lake and Go 1 d
Creek and parallels a traditional spring migration route southv1ard
t~. the Susitna River. "
The_direct effects upon this group of caribou, shou1d Plan 3 be
implemented, include disturbance to cows and calves during the road
construction period, a disturbance and possible impediment to
caribou migration as a result of increased traffic in the ar'ea, and
the possibility of direct mortality from road kills. Ho~1ever, the
presence of the road should not interfere with migration, since
caribou are known to cross roads.-Moreover, interference with the
calving.areas could cause a major adverse impact on the females who
show an affinity to traditional calving grounds.
Of greater importance than these factors, however, are the indirect
consequences to this group of caribou of increased access to its
range. An access road across this alpine tundra ~t/ould provide the
opportunity for all terrain vehicles to push a network of unplanned
trails throughtiut the range of this subherd. This new access would
cause disturbance and increased mortality to these caribou from
their contact with vehicles, campers, and hunters. Thusi there is a
chance that this route could lead to partial abandonment of impor-
tant caribou habitat.
The actual magnitude of impact is difficult to assess since it de-
pends on .the somewhat unpredictable behavior of both caribou and
man. With an increased emphasis on management of the area and
stringent _hunter control, it is technically possible to lessen the
potential extent of impact. It is expected, however, that resource
a-gencies would be apprehensi.ve about the success of any mitigtation
plans and \>Jould strongly resist any road access from the Denali
Highway.
(iv) Social Considerations
.
Without mitigating measures, access plans with a road~1ay originating
from the Parks Highway could significantly impact the westside com-
munities in terms of demand for increased services, changes in popu-
lation, housing availability, government expenditures and revenues,
·labor demand, and unemployment. There v1ill also be significant ef-
fects on construction, retail trade, and tourism.
Many of these changes wi 11 occur as construction workers attempt to
t,elocate to the communities near the construction site. Depending
upon commuting modes to the camp, there could be a large increase in
vehicular traffic in the area.
These access plans also include a road from, the Denali Highway. As
such, many of the impacts which would be felt in the \vest side com-
munities of Talkeetna., Trapper Creek, and Sherman would a1so occur
11-17
in Cantwell. With a road from the north, it is expected many of the
worKers would settle in Fairbanks., thereby reducfng some of the im-
pacts which th-e west side communities would experience.
These p~l ans meet the" preferen<;:e of the pub 1 i c in Cant\'Je 11 as some
changes wi 11 occur but wi 11 not meet the preferences expressed by
those in the west side communities who desire no cnange.
However, road access connecting the Oenal i and Parks Highway wou1 d
create extensive publfc access following construction thus creati~
the maximum change in the status quo of the area. ·
As discussed under Section 11.13, it is considered that mitigation
measures can be implemented to lessen the effects on the west ~ide
communities of Talkeetna and Trapper Creek. Hith road acces·s from
the Parks Highway, change in the remoteness of Go 1 d Creek and the
Indian River Land Disposal sites will occur regardless of
. mitigation.
(b) Access from Parks Highw.ay Only (Plans 1 and 5}
(i) Costs
Access Plans 1 and 5 are both comparable to the minimum cost altern-
ative (Table 11.1).
(ii) Ease of Operation and Construction Flexibility
Both Access Plans 1 and 5 satisfy the ea.se of operation crit\~ria by
having a road directly connecting both sites •. Both Access PlcUlS 1
and 5 satisfy the flexibility criteria by having a road connection
with a major highway.
Access Plans 1 and 5 involve a shorter haul distance compared to any
alternative having access vi a Dena 1 i Highway.
Anchorage has been identified as the most viable port of entry for
the majority of the materials and supplies (_). When comparing
Access Plans 1 and 5, with plans having access from the Denali High-
way, 1 ogi st i cs and cost advantage over any access from the Denali
Highway. With the majority of materials and supp1 i es coming from.
Anchorage, the access route from the Denali Highway \'IOUld involve an
additional haul of approximately 52 miles to ~Iatana when compared to
an access from the Parks Highway. The additiona1 52 miles of haul
to ~Jatana, for a Dena 1 i access alternative, vwul d be a disadvantage
in long-term operation and maintenance.
(iii) Biological Considerations
The primary conc~rns with. access from only the Parks Highway v1ere
discussed in (a) above. Briefly~ the concerns are the potential im-
pact"'to ~furtn:a~er habitat between the highway and Go1 d Creek and 'po-
tential degradation of fisheries habitat in the Indian and Susitna
11-18
I
I
I
I
.I
I
I
I
I
I
I
I
•=
I .
. "
I
I
' I
I
I
I
I
I
I
.I
I
I
I
I
I
I
I
I
I
I
I
I
Rivers. Of lesser concern is the disturbance of moose and bear pop ...
ulations and removal of their habitat caused by the northside con-
necting road in Plan 5. · ·
In addition to these, Plan 1 includes a ~onnection on the south side
of the Susitna. River bet\'-teen the two damsites. This road would pass
near and through extensive \'let land areas in the Stephan Lake-FQg
Lake area. These wetlands provide habitat for furbearers and water-
fowl and support a large, year:-round concentration of moose. Be-
cause this .area is currently relatively inaccessible, potential im-
pacts include removal of habitat and increased mortality through
hunting and trapping. ·
(iv) Social Considerations
Evaluation of these plans from a socioeconomic aspect reveals that
Plans 1 and 5 wi 11 result in the greatest impact to the west side
communities. Because access is provided from the west onlY, the
majority of the impacts v10uld be felt in the west side communities.
There would be a greater tendency for people to relocate in the com-
munities and perhap? in Anchorage and a .lesser tendency to live in
the Fairbanks area. There would be some impacts to the Cantwell
area, but fe\'/er than with a road from Dena 1 i. Impacts waul d be the
same as discussed in (a) above.
In terms of public preference, these plans least meet the desir·es of
people living in the project area. The plans would cause the great-
est change in the Talkeetna-Trapper Creek area (where residents have.
expressed negative attitudes toward social change) and would mini-.
mize impacts to the Cantwell area (where residents have expressed a
. desire for change). The Indian River land disposal site and Gold
Creek would experience the greatest change with the selection af
this plan.
(c) Access for Denali Highway (Pl.9JIS 4., ti and 11)
( i) Costs
Table 11.1 indicates that Plan 4 is.comparable to the least cost
alternative {Plan 5). The cost of Plan 6 is approximately $40
mill ion greater .than that of Plan 4 and the cost of Plan 11 is
approximately $35 million greater than that ofPlan 4.
(ii) Ease of Operation and Construction Flexibility
Plan .4 does not satisfy the ease of operation criterion due to the
absence of a road directly connecting the two dam sites. Plans 6
and 11 both have a road directly connecting the dam sites, therefore
both pl.ans satisfy the ease of operation criterion.
Plan 4 on1y partially meets the construction flexibility criterion.
-Plan 4 i;tcludes a road connection to a major highway for the \~atana
ll-19
(d)
deveJopment-but not for the De vi 1 Canyon development~ Access Plans
6 a·nd 11 both satisfy the flexibility criteria by having a ~:annec
tion to a major highway.
(iii) Biological Consid.etations
These three plans a11 involve road access from Denali Hi:9hvJqV to
Watana damsite. Th~ potential biological and cultural impa ~ asso-
ciated with this route were.discussed under (a) above. Basi tlly~
impacts could occur to portions of the Nelchina caribou herd through
increased hunting mortality and potential interference with migra-
tion and calving. Increased access and trapping pressure could also
impact furbearers. In addition, because of treeless topography and
shallm'l soi 1, disturbance and remova 1 of any cultural resources
could result.
,.
-Plans 4 and 6 involve construction of a rail connecting from Gold
Creek to Devil Canyon. No major environmental problems were identi-
fied along this portion of the route. The connection road on the
north side of the Susitna River _between the t~tto dams was discussed
under {b) above, the only environ~ent~l concern being the crossing
of moose habitat.
(iv) Social Considerations
Plans 4, 6 and 11 involve the major access point of origin on the
Denali Highway, rather than the Railbelt Corridor. Workers' fami-
lies would tend to locate more com:nunities, including Cantwell and
Fairbanks. Due to the rail access from Gold Creek, there waul d
still be changes in the west side communities, but fewer than with a
road originating from ·the Parks Highway~ Plan 11, involving access
from Den a 1 i Highway only, waul d cause the greatest number of changes
in the Cantwell and Fairbanks area and fewer changes to the west
side communities. These changes would be the same as described in
(a) above.
Access from Gold Creek Only {Plans 2, 8, ~. and 10)
Table 11.1 indicates that the total cost of Plan 8 and 9 are respectively
$15 and $30 million greater than the least cost alternative, Plan 5. The
substantial savings in construction costs are offset IJ.t increased personnel
shuttle costs and an allowance for contingency ri.sk .. The cost comparison
a 1 so shows that the total costs of Plans 2 and 10 are $55 mi 11 ion and S40
million more than that of the least cost alternative.
( i) Ease of Operation and Construction Flexibility
Access Plan 2 meets the criterion for ease of operation since the
dams are directly connected with a rai 1 route. Access Plans 83 9 ·
and 10 partially satisfy the -ease of operation and maintenance cri-
teria. These plans have a road directly connecting the two dam
sites, however, they do not have a connection to a major high\'lay.
11-2.0
I
I
I
I'
I
I
I
I
I
I
I
"_,7~--•• ~.
' I
I
I
I
I
t
I
' I
I
I
I
I
I
I
I
I -.-
1
I
I
I
I
I
I
I
·.-·1
( i i )
This reduces the flexibi1 ity in operation and maintenance of the.
site~ as discussed in Section 11.8. ·
Access Plans 2, 8, 9 and 10 do not satisfy the flexibility criteria
for construction as they do not h.ave a road connection to a major
highway. ·
Biological Considerations
These plans all preclude access from the Parks Highway or Denali
Highway; t.herefore, the impacts associated vtith increased access are
substantially reduced.
Plans 2_and10, which involve connections between vJatana and Devil
Canyon on the south side of the Susitna River, have as the major po-
tential environmental impacts, the disturbance of wetland areas near
Stephan and Fog Lakes,-as discussed under (b) above.
The overall reduction in access and the fact there is no access con-
necting with the Denali Highway to the north indicates these plans
would result in the _least number of impacts to biological and cul-
tura 1 ·resources.
(iv) Social Considerations
Thes·e plans all involve access from the west only, the only differ-
ence being road or rail, and if rail, the distance into the basin
the railroad extends. As such, impacts ¥muld again be concentrated
on the west side communities. These impacts ~JOuld likely be concen-
trated in the Gold Creek area as well as Talkeetna and Hurricane be-
cause of their location at rail-highway intersections. The Cantwell
and Fairbanks areas would be less affected as there would be no
northerly access.
The public has expressed a preference for a rail access and a main-
tenance of the status quo. · Although raii access would best maintain
the status quo of the Upper Susitna Basin in general with the rail
access, si gni.fi cant changes could occur in the Talkeetna/Trapper
Creek area as discussed in Section 5.1 (e).
These plans v;ould not meet the public preferences expressed by Cant-
well residents.
11.10 -Identification of Conflicts
From the evaluation pres.ented in Section 11.8, it is apparent no single plan
meets all the objectives or satisfies all the criteria established as part of
the ,study. The basic conflicts identified were:
. (a) Social and Biological tonsiderations vs
Construction and Operation Logistics
Rail or road access from a railhead at Gold Creek Nithout· road access from
ll-21
a major highway would limit social and biological changes in the immediate
project area and retain the status quo to the greatest extent possible.
This option is in dire~t conflict with the requirement to provide flexibil-
ity in construction logistics and transportation and to provide ease of op-
eration· and maintenance •. The selection of such an option would increase
the risk of high costs, schedule delays, and safety problems and decrease
project reliability ..
(b) Social vs Biological Considerations
Social and biological objectives are not in basic conflict since limited
acce5:; to tbe project area is most desirable in both cases.. If, however,
the assumpt ·ion is made that road access to a major hi gh\1ay \~Ji 11 be
provided, then a conflict arises. From the social/local public preference
perspecttve, access from the Denali Highway is preferred. This plan v10uld
create the ~conomic stimulus desired in Cantwell, reduce the potential for
change in the Trapper Cteek/Talkeetna area, while retaining the remoteness·'
of the Indian River land disposal site and the railroad communities north ·
of Talkeetna. The Denali access, however, is in conflict with biological
objectives since it would allow access by hunters and ATVs to a large
portion of the ll!"'per Susitna Basin and create potential impacts on the
Nelchina CaribrJ,J:: other big game species including moose and bear, the
fisheries in isolated lakes and streams and forbearer habitat. I-11
addition, the potential for disturbance of archaeological sites in this
area is greatest. Although mitigaion measures can be employed to reduce
these potential biological impacts; it is considered likely that gover-nment
resource agencies would be appr·ehensive about the success of any control
programs and would thus be opposed to any access fr9m the Dena]i High\vay.
The selection of a Denali access plan could result in unacceptable delays
in license approval or a subsequent rejection of this plan requiring a
reassessment of access plans from the west.
Table 11.2 broadly sunmarizes the conflicts in the evaluation,;
11.11 -Comparison of Access PI ans
(a) Access from Railhead at Gold Creek (Plans 2, 8, 9 and 10) vs
Access from ~1ajor Highway \r1al}s 1, 3, 4, 5, 6, 7, 11)
Considerable cost, schedule, safety and reliability risks are associated
with construction of a major project without road access to a major high-
way. On the other hand, road access to a major highway will create addi-
tional change in the status quo of the Upper Susitna Basin. If the deci-
sion is made to t:levelop a large scale hydroelectric facility in the Upper
Susitna Basin, it is considered essential that the orderly development and
maintenance of the facility should be afforded a higher priority than main-
tenance of the status quo. Thus, access plans originating at a railhead at
Gold Creek only are not recommended.
These considerations led to the rejection of plans not providing road
access to a major highway.
11-22
c;
,,
I •'
I
t
I
I
I
I
I
.I
I .·
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
~.
(b)
(c)
(d)
Plans eliminated in this comparison;· 2, 8, 9, 10
Plans remaining! ~-~ 33 4, 5, 6, 7, 11
Access From Both Parks Highway and Denali Highway
(P1 ans 3 ·!I 7) Vs Access from Only One Highway
(Plans 1 , ·. 2 , 4, 5 , 6 , · 7 , -8 , l 0 , 11 )
The plans which optimize transportation flexibility and ease of operation
involve the initial construction of a road from Denali Highvtay to \~atana
damsite., .. To~allow for improved logistics during the peak construction at
Watana and throughout the construction of De vi 1 Ca nyoii, road access. \-toul d
also b~ created to the Parks Highway. The disadvantages of these plans are
t_hat they wou1 d create the maximum change in the status quo producing both
the biological impacts associated with the Denali link and the social
impacts associated v1ith the Parks Hi gh~ttay 1 ink.. These impacts are further
intensified with both roads since the connection of the Parks and the
Denali Highway would encourage hunters and tourists to drive the complete
loop.
These plans are also more costly than the minimum cost alternatives. It
is considered that the social and biological impacts that would result
from these plans cannot be justified by the added transportation flexibil-
ity and ease of operation benefits associated with road access to both the
Parks and Denali Highways.
These conclusions resulted in the rejection of the plans providing road
access to both the Parks and Dena 1 i Hi ghv1ay.
Plans eliminated in this comparison:· 3, 7
Plans remaining: 1, 4, 5, 6, 11
Road\'lay ·Connecting the Dam Sites Directly
(Plans 1, 2, 5, 6, 7, 8, 9, 10 11) vs
No Roadway Connecting the. Dam Sites Directly (3, 4)
Plans incorporating a road connecting the dam sites directly are clearly
superior in terms of ease of operation and maintenance to plans which do
not directly connect the dam sites. The access p 1 ans which do not connect
the damsites directly do not have advantagesi n any of the other, or oDm-
bined criteria to \varrant not eliminating these alternatives from further·
consideration.
These conclusions resulted in the rejection of plans not connecting the
damsites directly.
Plans eliminated in this comparison: 3, 4
Plans remaining: (. 1, 5, 6, 11
Access to Denali Highway (Plans 3, 4, 6, 7, 11) vs
Access to Parks Highway (Plans 1, 5)
The main concerns associated with the Denali access are the potential
effects on the Nelchina caribou herd, increased access to a large area of
11-23
(e)
a 1 pine tundra vrith the associ a ted effects of disturbance by ATVs and dis-
turbance of potential cultural resources.
Although there are some fisheries and fur bearer concerns in the lndi an
River area associated with a Parks High\'lay access, from the biologicai per-
spective, Parks High~vay access is preferred to a Denali Hi ghvJay access.
From a social perspective, the Denali route is clearly superior to the
Parks Hi gh~tlaY route. The Denali route \'ioul d promote the economic stimulus
desired in Cantwell while reducing the influence on the community of
Trapper Creek and Talkeetna which has expressed· a desire to rna i ntai n the
status quo. It is considered,_ however, that even with a Parks Hi gh\lfay ac-
cess, mitigation in the form of se1f-contained construction camp facili-
ties, regulation of commuter schedules and control of transportation modes
can reduce or avoid many of the potential changes in Talkeetna and Trapper
Creek.
With any access plan from the \'lest, a major railhead vtould be located at·
Go 1 d Creek creating suffi ci ·ent 1 ocal changes. With road access from the
Parks Hfgh\'lay-to Gold Creek, changes \'/ill also ocCllr at Indian River-and
disposal sites ..
Based on the above discussion, it is concluded that the Parks Highway ac-
cess is preferab 1 e to the Dena 1 i access plan. This conclusion is based on
the assumption that:
-if a Denali route Here selected, it would be Plan 6 which \vould still re-
sult in significant social changes in the Gold Creek area;
-government resource agencies will be opposed to the Denali route with a
likely 1-to-2 year delay in schedule or denial of permit resulting;
-changes in local communities can, to a large degree, be mitigated through
controls imposed on contractor and construction workers; and
-controls \'/Oul d be very difficult to impose upon hunters and ATV operators
who would utilize the Denali's route after construction.
The foregoing consideration5 resulted in the elimination of plans involving
access from the Denali Highway.
Plans eliminated in this comparison:
Plans remaining:
Comparison of Plan 1 vs Plan 5
3' 4, 6' 7' 11
1' 5
P1 ans 1 and 5 both commence on the Parks Hi gh\'tay near Hurricane and p\"oceed
through Chulitna Pass and along the Indian River to Gold Creek. From Gold
Creek, both Plans proceed east on the south side of the Susitna River to
the Devil Canyon site. At De vi 1 Canyon, Plan 1 proceeds east on the south
11-24
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
side of the Susitna River to the Watana site. Plan 5 crosses the Susitna
River at Devil Canyon and proceeds east on the north side of the Susitna
River to the Watana site. Access Plan 1 has potential for greater environ-
mental impacts than Plan 5. Access Plan 5 has a slight cost advantage over
Plan 1, ~lso Plan 5 fs slightly easier to construct due to the difficult
terrain in the segment between Devil Canyon and Watana south of the Susitna
River. The only advantage Plan 1 has over Plan 5 is in Native landovmer
preference\.• It is, therefore, concluded that the environmental cost and
construction considerations out\~·eigh the Native lando\lmer preference, and
therefore, Plan 1 is eliminated from further consideration.
11.12 -Recommended Access Plan
Based on the above discussion, it is recommended that:
-The selected access plan for the construction and operation of the Susitna
Hydroelectric Project should comprise a road commencing near t~P 156 on the
Parks Highway, proceeding southeast crossing the Sus itna River at Go 1 d Creek,
turning northeast to Devil Canmyon Dam site along the southern side of the
Susitna River, crossing the Susitna River at Devil Canyon, an proceeding
along the north side of the Susitna River to Watana Dam site (see Plate 30).
-To allm'l for continued access for project construction by mid·-1987, a pioneer
road (limited access) between Gold Creek and Hatana Dam site be constructed
commencing· in mid-1983. The application for permits to construct this pio-
neer road be submitted to the State of Alaska and the Bureau of Land Manage-
ment by August 1982, independent of the FERC license application,.
-To mitigate against agency concerns in regard to the pioneer road concept and
to avoid the possibility of public access to the project area in the event
that the project is not built, road access between the Parks High\'lay and Gold
Creek not commence until after FERC license approval. If the project does.
not proceed after the pioneer road is constructed, the road as such should be
rendered impassable to future vehicular traffic.
-To minimize potential impacts to furbearers and fisheries resources in the
Indian ~·iver and Susitna River areas, special construction techniques be
. utilized (including adequate bank stabilization, revegetation and restora-
" tion) when crossing wetland areas or \·/hen constructing in proximity to the
Indian or Susitna Rivers.
.... To minimize the effects of public access during the op-eration phase of the
project consideration be given to controlling public access across Devil Can-
yon Dam.. If access is provided east of Devil Canyon Dam site, restrictions
should be placed on the use of ATVs and hunting.
-To assist in minimizing changes in the local communities of Talkeetna~
Trapper Creek, Sherman and Curry, it is strongly recommended that subsequent
decision on construction camp facilities, commuter modes, v10rk incentives,
and general policies incorporate a special effort to minimize the effects of
construction on these local communities. Specific mitigation recommendations
are included in Section 11.13.
11-25
The foregoing is based on the fo11m·ling assumptions~
The pioneer road concept wi 11 be approved by g.over nment regula tory agencies
since the pioneer road would not connect to any existing road before the
issuing of a FERC license, thus not making the prior commitment to allowing
public access to the Upper Susitna Basin~
-Although the native landowners have expressed a strong preference for road
access from Parks Highway to both darnsites along the south side of the Susit-
na Riv~r, their basic desires would be met by providing road access, from any
direction~ to their existing land. holdings~
-Public access will be prohibited during the construction phase of the pro-
ject. Also, the selection of Plan 5 offers some flexibility in regard to the
degree and type of pub 1 i c access subsequent to 1993.
-Biological and social impacts will be mitigated through adoption of the
recommendations presented in Section 11.13.
If permits to commence construction of the pioneer road are not obtained by
mi d-1983, it may be necessary to accept a 12-to 18-month de 1 ay in the on-1 i ne
schedule or possibly revert to one-of the less acceptable access plans which do
not require a pioneer road.
11.13 -Mitigation Recommendations
The plan recommended by Acres does not satisfy a11 the evaluation criteria out-
lined in Section 11.2. In or~er to reduce potential impacts to biological and
cultural resources and to alleviate socioeconomic impacts to the communities of
Talkeetna, Trapper Creek, Sherman and Curry'" the following mitigation measures
are recommended:
Permit only on-duty construction workers to have access to both the pioneer
road and access road.
-After construction of the pov1er development is complete, maintain a control-
1 ed access route beyond the Devi 1 Canyon Dam. It is anticipated a coof1eY·a-
ti ve agreement could be r~eached with BLM and ADF&G concerning. the number of
people permitted access to the areas and the cost of any control measures.
The construction camp should be as self-contained as possible, thus limiting
the number of workers who could otherwise bring their families to a nearby
community and commute daily.
Provide incentives to encouraged workers to work the longest time possible
between leaves to minjmize commuter traffic. Although the final schedule
will not be known until labor agreements are established and construction
commences, 1 anger work periods between breaks can be advocated.
-Provide planning assistance if requestd to the communities of Talkeetna~
Trapper Creek, Sherman and Curry to aid them in preparing for the effects of
increased populations.
ll-26
I
I'
I
'I \
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Evaluate various commuter management policies and select the one which
reduces impacts .to the local communities. Socioeconomic impact assessment
studies currently under way for the Susitna Project will provide important
input data for evaluating possible commuter management policies.. ·
-Utilize excavated cuts and other construction techniques to. prohibit utiliza-
tion of the pioneer road after construction of the access road. Areas used
for the pioneer road which do not follow final road alignment should be
reclaimed.
11 .. 14 -Tradeoffs Made "'n the Se 1 ecti on Process ·
(a) Basis of Selection.Process
From the natural resource and local public preference perspective, mainten-
ance of the status quo is probably most favored. However, it is unrealis-
tic to consider that a project the size of Susi~na can be implemented with-
out changing the existing character of sections of the Upper Susitna .
Basin ..
Access to the dam sites is a complex and controversial issue. As such, it
has received considerable attention from the study team, APA, resource
agencies and the public .. Although the studies have determined that there
is no single access plan that satisfies all the project objectives and
evaluation criteria, it has·been possible to develop an access plan which
provides a reasonable tradeoff of preference. These tradeoffs are essen-
tially based on the following compromises:
-All entities must present a degree of flexibility, otherwise a satisfac-
tory compromise is impossible.
-Whenever a specific objective is partially crnnpromised, considerable ef-
fort is made during subsequent decisions to compensate.
-Any compromises made are clearly outlined such that decision makers t~e
_viewing the final recommendation are aware of negotiations to date.
· (b) Tradeoffs Made in the Selection Process
( i) Engineering
Concessions made include:
-no road access from Denali Highway which \vould incluoe a complete
1 oop connect i ng Parks Highway with Dena 1 i Highway;
-no pioneer road to Parks Highway .Prior to the issuance of a FERC
1 i cense;
11-27
0
commitment to be prepared to make the pioneer road impassible if
FERC license not granted;
-restrictions to be placed on vmrker commuting schedules and mode;
worker incentives to be provided to minimize effects on local
communities;
Objectives retained-include:
-road access to both dam sites to allow for ease of construction,
operatiorr and maintenance of the project;
-maintenance of schedule through retention of the basics of the
pioneer road concept.
(ii) Biological
Concessions made include:
-road access from Parks Hi ghNay affecting Indian River area and
providing partial public access to the upper basin:
Objectives retained include:
-no access from Denali Highway which was considered to have the
greatest potential for environmental impact;
-no route on the south side of the Susitna River between the dam
sites, thus avoiding the sensitive Stephan Lake and Fog Lakes
area;
-emphasis on construction mitigation when developing road link be-
tween Parks Highway and Go 1 d Creek;
-retention of a degree of control on future pub 1 i c access by ac-
cepting the Parks Highv1ay plan \-Jhere, due to the terrain, private
vehicles are basically restricted to the access corridor bet\'leen
Parks Hi gh\1/ay and the De vi 1 Canyon dam sit e. The degree and type
of access east of Devil Canyon can be somewhat controlled by regu-
1 ati on of access across the Devil Canyon dam.
The alternative of not connecting to a major highvJay \~as considered
to have the least net adverse biological impact. The ease of opera-
tion and maintenance and the construction flexibility criteria, as
explained previously, was considered to outweigh this advantage.
The mitigation measures and road management will reduce the adverse
biological impacts associated with an access connection to a major
highway, to a minimum.
11-28
ll
I
I
I
I-
I
I
I
I
I
·t
I
I
I
I
I
I
-I
I
I
I
I
I
••
I
I
I
I
I
I
I
I
I
I
I-
I
( i i i ) Soc i a 1
Concessions made include:
-road access to the Upper Sus itna Basin;
-road access from Parks Highway which creates greatest potential
for change in the Indian River land disposal si-te ..
Objectives retain0d include:
-Through the implementation of a relatively self-contained con-
struction camp~ restrict ion of private vehicles fr·om the construc-
tion site~ implementation of mass transit modes for commuting wor-
kers!) incentives to encourage workers to remain on site and con-
trolled public access east of Devil Canyon following construction~
it ·is considered that changes in the local communities of Trapper
Creek/Talkeetna area will be minimized;
-Although the western communities favored a rail access, they also
favored maintaining the status quo. It is our opinion that the
recommended p 1 an with its associated rni t i gat ion should produce
less change in the Talkeetna/Trapper Creek area than an all-rail
access plan. ·
Overall consensus of the local community preference favored access
fy··om the Denali Highway. _The advantages of the Parks Highway ac-·
cess over the Denali access in reducing the biological impacts is
considered to outweigh the local community preference. In addi-
tion to the lessened biological impacts, the recommended plan
better meets the preferences of Native 1 and owners~
The recommended plan does not fully meet the preferences of the
Native 1 and owners. They would prefer the access road between
uevil Canyon and Watana be located on the south side of the
Susitna River. The advantages of the road being located on the
north side of the Susitna River, include, reduced biological im-
pacts~ the. actual construction of the road is easier than if lo-
cated on the south side, and the construction cost of the road is
less. These advantages are considered to outweigh the Native
1 and owner pre: erence of having the road located on the south side
of the Susitna River.
11-29
-f'!"··.·.· -· .•. ...,....., •. ,""'C""··. ~··-~·~' •••
I J".
I
I
t'LAN
DESCRIPTION:
I
I
r I FAr.F OF NEW ROAD
,J . . TION
CQ.S_l ($ x l,OOO,OOO)
INlENANCE
($ X 1,000,000)
lOGISTICS COST
X 1,000,_0JJO)
TOTAL
($ X 1,000,000)
rl-K;:,uJ\!NEl SHUTTLE
($ X 1,000,000)
CONTINGENCY RISK
~X 1,000,000)
AL COSTS
X 1,000,000) .
ifRUCTION SCHEOlU F
~_J,O_R. B_RI DGES
I
I
I
I
I
I
I
I
1 z
ROAI1\IIAV: · PARKS RAIL: GOLD
HIGHWAY TO OCVIL CREEK TO OCVIL
CANYON & WATANA CANVOI4 & WAt ANA
ON SOUTH SIDE ON SOUTH SlOE
Or SUSITNA OF SUSITNA
62 58
.
158 140
5 4
215 210
378 354
0 25
0 40
378 419
3-4 3-4
2 2
TABU: 11' .. 1: SUSUNA ACCESS PLANS
..$ ~4 ' 6
ROA[)IAV: DENALI ROAOOAV: DENALI ROAI)tlAY: PARKS ROA{)I{AY: D::NALI
HIGHwAY TO HIGHWAY TO HIGHWAY TO OEVJ.L . HIGHWAY 10
WATANA, PARKS WATANA,. RAIL, l~OLO CANYON ON SOUTH WATANA, RAil, GOLD
HIGHWAY TO CREEl< TO OEV IL SIDE OF SUSITNA, CREEl< TO OCVlL
DEVIL CANYON CANYON ON SOUTH OEVIL CANYON TO CANYON ON SOUTH
ON SOUTH SIDE sroc or SUSlTNA. WATANA ON NORTH SIOC OF SUSITNA.
OF SUSITNA. NO NO CONNECTING SID£ OF SUSITNA. CONNECTING ROAD
CONNECHNG ROAD ROAD ON NORTH SlOE OF
SUSHNA~
70 60 68 102
151 119 143 179
6 5 8 8
231 230 214 230
388 354 365 417
0 10 0 0
0 15 0 0
388 379 365 4'!7
1 1 3-4 1
0/1 0 2 0
~
I H '':i. lU 11
• .
ROAIJrlAY; DENALI ROAD4AV: GOLD RAILi GOLD RAIL: GOLD ROAD'!AV: OCNALI
HIGHWAY TO CREEK TO DEVIL CREEX TO OCVIL CREEK TO OCVIL HIGHWAY TO WATANA
WATANA, PARKS CANYON ON SOUTH CANYON ON SOUTH CANYON ON SOUTH CONNECTING ROAD
HIGHWAY TO DEVIL SIDE OF SUSlTNA, SIDE OF' SUSlTNA, SIOC OF' SUSHNA .. BETWEEN WATANA
CANYON ON SOUTH DEVIL CANYON TO .ROA~AY OCVlL ROAO/IAY OCVIL AND DEVIL CANYON
SIDE Of SUSITNA. WATANA ON NORTH CANYON TO WATANA CANYON TO WATANA ON NORTH SIDE
CONNtCTING ROAD SIDE or SUSITNA. ON NORTH SIDE ON SOUTH SlOE Ot SUSlTNA.
ON NORTH SIDE OF SUSITNA. OF SUSITNA .. or SUSITNA.
111 54 . 58 53 86
209 93 108 123 145
9 7 .~· 5 11 .
231 214 216 214 258
449 314 '329 342 414
0 25 l? 25 n
0 40 40 40 0
449 379 394 407 414'
1 3 3 3 1
0/1 1 1 1 0
{' ,-_, .. ~ ::...
I
I
I
I
I
I ,
I
I
I
I
I
I
I
I
I
I
I
I
I
TABLE 11.2: IDENTifiCATION OF CONFLICTS
Criteria
Costs
Minimize Costs
Ease of Operation and
Construction Flexibilit}'
Ease, of Operation and
Maintenance
Construction Flexibility
Biological ~
Minimize Biological Impacts
Social,
Accommodate Preference of
Native landowners
Accommodate Local
Community Preference
1 -Does not satisfy criteria
2 -Intermediate
3 -Satisfies criteria
1 2
3 3
3 2
3 1
2 3
3 1
1 2
3 4 5 6
3 J 3 1
1 1 3 3
3 2 3 2
1 1 2 1
2 2 2 2
2 3 1 3
7 a 9 10 11
1 J 3 3 1
3 2 2 2 3
3 1 1 1 3
1 3 3 3 1
2 1 1 1 1
2 2 2 2 3
0
J.
DEFINE OBJECTIVES
· SELECT ACCESS
ROUTE TO HYDROPOWER
SITES THAT ALLOWS
CONSTRUCTION AND
1 OPERATION WHILE
BEST MEETING
OVERALL CRITERIA
STATED IN .([J
2.
...,__,.... DESIGN PARAMETERS
ROADWAY AND. RAIL
ENGINEERING CRITERIA
2.
ESTABLISH CANDiDATES
A TOTAL OF 33 r-
ROUTES ARE
ESTABLISHED IN
THE 3 CORRIDORS
1
2A .
PUBL1C PARTICIPATION
PRESENT THE OPTIONS -
TO THE PUBL;C AND
INVITE COMMENT
3.
SCREENING PROCESS
~ TECHNICAL
ECONOMIC
ENVIRONMENTAL
PUBL!C PREFERENCES
TRANSMISSION IMPACT
4 ..
PLAN FORMULATION
3 ROUTES ONE lN
. ........-_ .... EACH CORRIDOR
AS A RESULT OF
THE SCREENING
PROCESS iN IID
ARE ESTABLISHED
POHT FACILITIES
ROADWAY OPTIONS
RAIL OPTIONS
LOGISTIC REQUIREMENTS
8 PLANS, WHICH
UTILIZED THE
3 ROUTES ARE
ESTABLISHED
5.
EVALUATION
ADDITIONAL STUDIES
SOILS DATA
ENGINEERING
r+ CONSTRUCTION COSTS ........
_. II ALTERNATIVE PLANS
ARE EVALUATED TO
THE FOLLOWING
CRITERIA
LOGISTICS COSTS
TRANSMISSION IMPACT
-.f ENVIRONMENTAL.
LABOR ORGANIZAT~ON
..-.CONCERNS
AGENCY CONCERNS
AS A RESULT OF
r+-AGENCY CONCE'RN.S,
3 ADDITIONAL PLANS
·ARE ESTABLISHED
~ NATIVE LANDOWNERS
PREFERENCES
~ LOCAL CQMMUNITY
PREFERENCES
-
·---
ENGINEERING
ECONOMIC
ENVIRONMENTAL
SCHEDULING
DESIRED LEVEL OF
ACCESS
AGENCY CONCERNS
SOCIAL PREFERENCES
TRANSMISSION
ACCESS' PLAN SELECTION METHODOLOGY
~IGURE JU IJR·
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
••
I
I
I
PARKS
HWY.
PARKS
HWY.
' ' '
CANTWELL · DENALI HWY.
PROPOSED
,ROAD
GOLD~ --__... D.C~ ---
CREEK SITE
ANTWELL
GOLD
CREEK
PLAN I
DENALI HWY.·
PLAN 2
--.-\_WATANA
SITE
WATANA
SITE
FIGURE ,,.21A~~m I L----------------------------~~~----------------
I
: ••.
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I'
~' --
1
I
PARKS
HWY.
CANTWELL. DENALI HWY.
HURRICANE
1!>.._
GOLDJ --
CREEK
PLAN 3
PROPOSED
-j
D.C.
SITE
ANTWELL /DENALI HWY.
HURRICANE
G. OLD D.C.,
CREEK SITE
,PLAN 4
I
I
I
I
~~
ROADS~ . I
I
I
tWATANA
SITE
G
I
I
I
PROPOSED~
ROAD
J
I
1
~
,, lwATANA
SITE
FIGURE 11.5 !A~Im!·
I
I ®
I
I
I
I '
I
I
I
I
I
I PARKS
HWY.
I
I
I
I
I
I
I
CANTWELL DENALi HWY.
HURRICANE
' '
PROPOSED ROADl_ -
. _.... --.......
GOLD.J-
CREEK
---:cf . LWATANA
SITE SITE
CANTWELL
HURRICANE
GOLD
CREEK
PLAN 5
DENALI HWY.
PROPOSED
ROAD
l
I
I
I
I
I
1
I
-..tWATANA
SITE
--------,-~-.-,
PLAN 6
FIGURE 11.41 ~~~l~ l
I
I @
I
I
I
I
I
I
I
I
I
I
I
I
I
0
I
I
I
I
>-"-'""'"""'CANTWELL L D~NALI HWY.
HURRICANE
PROPOSED ROADS
---·
•
I
I
I
I
I
I
I
GOLy--
CREEK
---r
D.C. -lWATANA
SlTE SITE
PLAN 7
DENALI HWY.
HURRICANE
PROPOSED\ ROAD ·
.,.,..._ -.._._._
GOLD_j---
CREEK •
__. --
D.C) l WATANA
SITE SITE
PLAN 8
FIGURE IL5.
I
I ® I
I
I
I
I
.I
I
I
I
I
I
I
I
I
I
I
••
CANTWELL
HURRICANE
DENALI HWY.
]
PROPOSED
ROAD --..._.
GOLD
CREEK
,-'~ ~~~~~H+~r '\_WATANA
CANTWELL
HURRICANE
GOLD
CREEK
PLAN 9
DENALI HWY.
PROPOSED
ROAD
SITE
- -_. -tWATANA
SITE
PLAN iO ··.
FIGURE 11.6! ~~~m1
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
PARKS}_ HWY. .,
CANTWELL
HURRICANE
GOLD_,
CREEK
DENALI HWY.
J"' D.C.
SITE
PROPOSED
ROAD
---
PLAN II
I
l
I
1
I
J
-..'-WATANA
SITE ·
FIGURE 11.71 A~~~~ I
I
I
I
I
I
I
I
I
I
I.
I
I
I
I
I
I
I
I
I
12 -HATANA U.£VELOPMENT
This section describes the various components of the 'tJatana Development, includ-
ing diversion facilities, emergency release facilities, the main dam, primary
out1et facilities, reservoir, main and emergency spi'llways, the power intake,
penstocks and-powerhouse complex including turbines, generators, mechanical and
electrical equipment, switchyard structures,· and equipment and project lands. A
description of permanent and temporary access and support facilities is also
included.
12.1 -General Arrangement
The evolution of the Watana gene~~al arrangement is described in Section 9. The
Watana reservoir and surrounding area is shown in Plate 31. The site layout in
relation to main acce,s5 facilities, borr·m'l areas and camp facilities is shown in
Plate 32. A more detailed att"angement of the various site structures is pre-
sented in Plate 32A.
The Watana dam will form a reservoir approximately 48 miles long, with a surface
area of 38,000 acres, and a total volume of 9,515,000 acre.:feet at a norma1
maximum operating elevation of 2185. During operation, the reservoir will be
capable of being drawndown to a minimum elevation of 2045.
The dam will be an earthfill structure with a vertical central impervious core.
The crest elevation of tht= dam will be 2210, with a maximum height of B85 feet
and a crest length of 4,100 feet. The total volume of the structure will be
approximately 62,000,000 cubic yards. During construction, the river will be
diverted around the main construction area by means of two concrete-lined diver-
sion tunnels, each 40 feet in diameter, on the right bank of the river.
A power intake located on the right bank will comprise an approach channel in
rock leading to a multi-level reinforced concrete gated intake structure capable
of operation over the full drawdown range. From the intake structure, six pen-
stocks, consisting of concrete-lined tunnels, each 17 feet in diameter, \·rU1
lead to an underground powerhouse complex housing six Francis turbines with a
rated capacity of 170 MW and six semi-umbrella type generators each rated at 180
~WA. Access to the powerhouse camp 1 ex \•Ji 11 be by means of an un 1 ined access
tunnel. Turbine discharge will be conducted through six draft tube tunnels to
two surge chambers dO\•mstream of the povJerhouse, then by means of two 30-foot
diameter concrete-1 ined tailrace tunnels. A separate transformer gallery or
chamber just upstream from the po1t1erhouse cavern will hGuse nine single-phase
15/345 kv transformers. The transformers will be connected by 345 kV single-
phase, oil-filled cable through two cable shafts to the switchyard at the sur-
face.
A tunnel spillway located on the right bank will be designed to discharge all·
flm·1s resulting from floods having a return frequency of 1:50 years or less.
This structure will be equipped with six fixed-cone valves at the dO\'/nstream end
to minimize undesiranl€ nitrogen supersaturation in the river downstream from
the dam during spillway operations. Flows resulting from floods with a fre-
quency greater than 1:50 years but 1 ess than 1:10 ,OUO years wi 11 be discharged
by a chute spillway a1 so on the right bank. The spillway control structute at
12-1
the upstream end wi 11 be cantrall ed by three fixed 1t1hee 1 gates 1 ead ing to a ·
reinforced-concrete-line chute section and then to a flip bucket at the down-
stream end. An emergency spillway on the right bank will provide sufficient·
additional capacity to permit discharge of the PMF without overtopping the dam.
An emergency release facility will allow lowering of the reservoir over a period
of time to permit emergency inspection or repair.
12.2 -Site Access
(a) Roads
At \~atana the main access ro~rJ wi 11 enter the site from the north. In
addition to the main acces~, s.~veral additional roads will be required to
the camp, village, airstrips tc.nk farm, haul roads to the borrow areas, and
construction roads to the dam al'd all major structures* These roads with
the exception of the haul roads c:.re snown on Plate 35.
The construction roads will be 40-foot \~!ide gravel surfaced roads with
small radius curves and grades limited to 10 percent. Major cut and fill
work will be avoided. A gravel pad approximately 5 feet thick will be re-
quired for the roads. This gravel pad will provide a drivable surface and
also will protect against the sporadic permafrost areas.
(b) Bridges
No major ~emporary bridges will be required for the construction of the
t~atana development. The crests widths of the upstream and downstream cof-
ferdams \•Jill be planned to provide suitable access to the south bank of the
Susitna River during construction.
I
, •.
I
·I
I
I
I
I
I
I
The camp l eted main dam crest wi 11 provide permanent access across the I
Susitna River.
One area which may require a small temporary bridge is Tsusena Creek near· 1 its confluence with the Susitna River. Currently it is envisioned that
this crossing can be accomplished with one large or multiple culverts.. ·
(c) Airstrip II
A permanent airstrip will be constructed approximately 2.5 miles north of
the main construction camp (see Plate 35). The runway \vill be 6,000 feet I
in length and will be capable of accommodating the C-130 Hercules aircraft,
as well as small jet passenger aircraft. A road will serve the airstr'p
connecting to the camp, village, and damsite, A sma11 building will t 1 constructed to serve as a terminal and tower and a fuel truck/maintenar .. -:
facility will be constructed.
A temporary airstrip ~vi 11 also be constructed to support the early phases
of mobilization and construction. This temporary runway \-Jill be 2,500 feet
in length and will be located in the vicinity of the main construction
camp. The airstrip will be capable of supporting other type aircraft ..
12-2
I
I
I
I
··I·
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
The temporary airstrip will eventually be incorporated into one of the main
haul roads for Borro~tl Area D. This will occur after the permanent airstrip
is in service~
(d) Access Tunnel
An access tunne 1 wi 11 be provided to the underground po\'Jerhouse and associ-
ated works. The main access tunnel will be approximately 35 feet wide and
28 feet high. The tunnel will allow permanent access to the operating
development and will also be uti1ized during construction as the main con-
struction ·tunnel. Construction adits will branch off to the various com-
ponents of the development during construction.
12.3 -Site Facilities
(a) General
The construction of the Watana development will require various facilities
to support the construction ·activities throughout the entire construction
period. Following construction, the operation of Watana will require cer-
tain facilities to support the permanent operation and maintenance of the
PO'--Jer fac i 1 ity.
The most significant item among the site facilities will be a combination
camp and village that will be constructed and maintained at the project
site. The camp/village will be largely a self-sufficient community housing
and maintaining living facilities for 6,000 people during constructien of
the project. After construction is complete, it is planned to dismantle
and demobilize the facility and to reclaim the area. It is additionally
planned to utilize dismantled buildings and other items in the camp/village
as much as possible for use during construction of the Devil Canyon devel-
opment. Other site facilities include contractors• work areas, site power,
services, and communications. Items such as power and ccmmLnications will
be tequired for construction operations independent of camp operaticns.
The same wi 11 be true regarding a hospital or first aid room ..
Permanent facilities required will include a permanent town or small ~om
munity for approximately 130 staff members and their families. Othelr' per-
manent facility items will include a maintenance building for use during
subsequent operation of the power plant.
A conceptual plan for the permanent town has been developed, however~ it is
~ecommended that preliminary design and fin?, ~esign be defered until near
the end of construct ion \-Jhen more in format i 1·: .1s to the physical parcrmeters
of design is available and, more importantly, the human requirements and
preferences are better defined.
Fuel oil has been selected as the means of heating the camp/village facil-
ities.
(b) Temporary Camp and Vi 11 age
The proposed location of the camp and vil"tage will be on the north bank of
the Susitna River bet~tJeen Deadman and Tsusena Creek, approximately 2.5
miles northeast of the Watana Dam. The north side of the Susitna was
chosen because the main access will be from the north and south-facing
slopes can be used for siting the structures and the location. The loca-
tion is shown in Plate 35.
The camp will consist of portable woodframe dormitories for bachelors with
modular mess halls, recreational buildings, bank, post office, fire sta-
tion, warehouses, hospital, offices, etc. The camp will be a single status
camp for approximately 5,000 workers.
The village, accommodating approximately 550 families, will be grouped
around a service core cant ai ni ng a school, gjtflln as i urn, stores, and recre a-
tion area.
The village and camp areas vJill be separated by approximately 1.5 miles to
pro vi de a buffer zone between areas. The hospita 1 wi 11 serve both the main
camp and village.
The camp location wi 11 separate 1 iving areas from the work areas by a rni le
or more and keep travel time to work to less than 15 minutes for most per-
sonnel.
The camp/vi 11 age wi 11 be constructed in stages to accommodate the peak work
force as presented in Table 12.1. The facilities have been designed for
the peak work force plus 10 percent for turnover. The turnover will in-
c 1 ude a 11 owances for over 1 ap of work er·s, v ac at ions, and vis i tor-·s. Th~ con-
ceptual layouts for the camp and village are presented on Plate 36 and 37.
(i) Site Preparation
Both the camp and the vi 11 age areas wi 11 be c 1 eared in select ar·eas
for topsoil, and the topsoil will be stockpiled for future use in
reclamat1on operations. At the vil1age site, selected areas w1ll be
left with trees and natural vegetation intact.
Both the main camp and the vi 11 age site have been se 1 ected to ~:n~o
vide well-drained land with natural slopes of 2 to 3 percent.
A granul.ar pad varying in thickness from 1 to 8 feet wi 11 be p 1 aced
at the main camp, covering most of the areas inside the perimeter
fence. This will provide a uniform working surface for erection of
the high density housing and service buildings and will serve in
certa1 n areas to protect the permafrost \'/here it underlies the camp.
In the village area, a granular pad will be installed only as neces-
sary to support the housing units and to provide a ~uitable base for
construction of the temporary village center buildings .
..
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
.I
I
I
I
I
a·
I
I
I
I
I.
I
I
I
All roadways within the camp/vi11 age areas will be flanked by road-
side ditches; with culverts carrying water across the intersections.
In general, drainage wi 11 tie through construction of a surface net-
work of ditches. Peripheral ditches will intercept overland flows
from adjacent non-cleared 1 and and carry them around the camps"
Runoff will ~ltimate1y be directed to existing drainage channels
1 ead ing to Tsusena Creek for the vi 11 age, and the Sus itna River for
the main camp.
( ii) Facilities
Construction camp buildings \'lill consist largely of trailer-type
factory-built modules assembled at site to provide the various
facilities required. The modules will be fabricated complete with
heating, lighting and plumbing services~ interior finishes, furnish-
ings, ·and equipment. Trailer modules will be supported on timber
cribbing or blocking approximately 2 feet above grade. Larger
structures such as the central utilities building, warehouses and
hospital will be pre-engineered, steel-framed structures with metal
cladding.
The larger structures will be erected on concrete-slab foundations.
The slab will be cast on a non-frost susceptible layer at least the
thickness of the annual freeze/thaw 1 ayer.
Permawalks will connect the majority of the buildings and dorr:ai-
tor i es ~ Th.e permawa 1 ks will be heated.
The various buildings in the camp are identified on Plate 37 ..
A detailed description of the ·nature and function of the buildings
is presented in Appendix 08.
(c) Permanent Town
The permanent town \'li 11 be 1 oc a ted at the north end of the temporarJJ v il-
l age (see P1 ate 35) and be arranged around a sma 11 lake for aesthet i\:c pur-
poses.
The permanent town vtill consist of permanently constructed building:s :and
not factory built prefabricated type modules. The various buildings in the
permanent town are listed below:
Single family dwellings;
-~lultifamily dwellings;
-Hospital;
School;
Fire station;
-A town center will be constructed and will contain the following~
. a recreation center
. a gymnasium and swimming pool
. a shopping center
12-5
{d)
The concept of ~uilding the permanent town at the beginning of the con-
struction period and using it as part of the temporary village was consid-·
erect.. This concept was not adopted, since its intended occupancy and use
is a minimum of 10 years away, and the requirements and preferences of the
potential occupants cannot be predicted w1th any degree of accuracy.
Sits:: Pov1er and Uti 1 it ies -.
( . ) 1 :t
( i i)
0
Power
Electrical power will be required to maintain the camp/village and
construction activities. A 345 kV transmission line will be con-
structed and will service the site from 1987 onward. The 345 ·kv
transmis~ion line will be operated at 138 kV while it is bringing in
power to the site. After the Watana development is complete and in
operation, the transmission 1 ine will supply power to the Intertie
from Watana and will operate at 345 kV. Since the transmission line
will be required after· construction is complete, the only cost of
the line attributable to the camps will be lhe interest costs in-
volved in constructing the line 6 years earlier than required for
permanent operation.
During the first two years of constructio~ (1985 and 1986), the
pm'/er supply will come from diesel generators. ·These generators
will remain on site after 1987 as standby power supply since site
contractors will provide for their own construction pm·1er after this
time. The peak demand -during the peak camp population year is esti-
mated at 13 MW for the camp/vil 1 age and 7 MW for construction re-
quirements, thus totaling 20 MW of peak demand.
The distribution system in the. camp and village construction \'li1l be
34. 5 k v. .
Power for the permanent town will be supplied from the station ser-
vice system at the power plant.
Water
The \'later supply system will provide for potab1e water and fire pro-
tection for the camp and village construction and selected contrac-
tor's work areas. The estimated peak population to be served will
be 6,800 (5,000 in the camp and 1,750 in the village).
The principal . source of water will be Tsusena Creek, vlith a back up
system of wells drawing on ground water. The water wi 1l be treated
in accordance with the Environmental Protection Agency 1 s (EPA) pri-
mary and secondary requirements.
A system of pun1ps and constructed storage reservoirs wi 11 provide
the necessary system demand capacity. Distribution \vi11 be by duc-
tile iron pipe system contained in utilidors. The utilidors will be
I
·a·". . .
;I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
.I
I
I
plywood box sections, most of which will be integrdl with the perma--
walks. The distribution and location of major components of the
water supply system are presented in Plate 35. · Details of the util-
idors are =presented in Plate 38.
(iii) Haste Water
A waste \'later collection and treatment sys tern will serve the camp/
village. One treatment plant 'r'lill serve the camp/village, while
gravity flow lines with lift stations will be used to collect the
waste water from all of the camp and village facilities. The uin-
camp•' and 11 in-village 11 collection systems will be run through the
permawalks and utilidors so that the collection system will be pro-
tected from the elements.
At the village, an aerated collection basin will be installeo to
collect the sewage~ The sewage will be pumped from this collection
basin through a force main to the sewage treatment plant.
An aerated collection basin will be needed at the village to balance
out the highly cyclic waste water flows.
The chemical toilets located around the site will be serviced by
sewage trucks, which will discharge directly into the sewage treat-
ment plant.
The sewage treatment system will be a biological system with
lagoons. The system will be designed to meet Alaskan state water
1 aw secondary treatment standards. The 1 agoons and system wi~ 1 be
modul at to all ovJ for phased growth and contraction of the camp/
• 1 1 v1. age.
The location of the treatment plant is shown in Plate 37. The loca-
tion was selected to avoid unnecessary odors in the camp as tne
winds are from the SE 4 percent of the time, which is cons ioe~ed
minimal.
The sewage plant will discharge its treated effluent through a force
main to Deadman Creek. All treated sludge will be disposed iG a
solid waste sanitary landfill.
(e) Contractor 1 s Area
The onsite contractors \vill require office, shop, and general work ar·eas.
Office space for the contractors has been provided and its location is
shown on Plate 35.
Partial space required by the contractors for fabrication shops, mainten-
ance shops, storage or warehouses~ and \•Jork areas within the camp confines
has been designated and is shown on Plate 37. Additional space requir-ed
for the aforementioned items will be located bet\veen the main camp and the
main access road.
12-7
12~4 -Diversion
(a) General
Diversion of the river flo~tl during construction wil1 be accomplished \~ith
two 38 foot diameter circular diversion tunnels. The tunnels will be
concrete-lined and located on the right bank of the river. The tunnels are
4,050 feet and 4,140 feet in length. The diversion tumiels are shown in
plan and profile on Plate 39.
The tunnels are designed to pass a flood with a return frequency of 1:50
years, equivalent to peak inflow of 81,100 cfs. Routing effects are small
and the peak fl 0'!/ the tunne 1 s \vi 11 d i sci1 arge a peak flow of 80,500 cfs,
The maximum water surface elevation upstream of the cofferdam is El 1!53ti.
-A rating curve is presented in Figure 12.1.
The upper tunnel or tunnel No. 2 will be converted to a permanent low level
I
I
I
I
I
I
outlet after construction. The local enlarging of the tunnel diameter to I
45 feet is to accommodate the low level outlet regulating devices after the
tunnel is used as a diversion tunnel.
(b) Cofferdams I
As discussed in Section 9 t~2 nature and riverbed will require a slurry
wa11 th~ough the alluvium mat2rial to bedrock. The soil/bentonite slurry
\vall will be constructed through the ·c"!osure dam and alluvium material to
bedrock in order to minimize the amount of seepage into the main dam exca-
vation. The abutment areas above riverbed ~tJill be cleared and grubbed~
with excavation of all overburden material to sound rock prior to placement
of any cofferdam fi 11 .
The upstream coffer·d am will be a zoned embankment founded on the c 1 a sure
dam (see Plate 40). The closure dam \'lill be constructed to Elevation 1475
. based on a low water Elevation 1470, and \vill consist of coarse material on
the upstream side grading to finer material on the do¥mstream side. \'Jhen
the closure dam is completed the soil/bentonite slurry wall will be con-
structed to minimize seepage into the main dam foundation excavation-A
dewatering system will be established in the main dam excavation to control .
inflow from ~he abutments and runoff.
Above Elevation 1475 the cofferdam \vill be a zoned embankment qms1st:1ng of·
a central impervious core, fine and coarse upstream and downstream filters,
and rock and/or gravel supporting shell zones ''lith t"ip-rap on the upstream
face.
The downstream cofferdam wi 11 be a closure dam constructed from approximate
Elevation 1440 to 1472. It will consist of coarse material on the dO\'In-
stream side grading to finer material on the upstream side, with a soil/
bentonite slurry wall can be constructed in the finer material as described
above for the upstream cofferdam,
12-8
I
I
I
I .,
I
I
I
I
I
I
I
I
I
I
I
~.
I
I
I
I
I
I
I
I
I
I.
I
I
I
The upstream cofferdam crest elevation has a 9 foot freeboard allowance.
This includes 3 feet for settlement and wave runup and 6 feet for upstream
reserve 1 r ice protection 4 Large chunks of ice will be present in the
river during the spring flood. The 6 foot freeboard prevents ice from
overtopping the cofferdam, causing damage.
(c) Tunnel Portals and Gates Structures
A reinforced concrete gate structure \'li 11 be located at the upstream end of
each tunnel, housing two closur·e gates (see Plate 41).
E~ch gate will be 40 foot high by 16 foot wide separated by a center con-_
crete ~ier. The gates will be of the fixed roller vertical lift type oper-
ated by a wire rope hoist. The gate hoist will be located in an enclosed,
heated housing. Provision wi1l be made for heating the gates and gate
guides. The gate in Tunnel No. 1 will .be designed to operate ~'lith the res-.
ervoir at elevation 1,540:~ a 50 foot operating he_ad. The gate in Tunnel
No. 2 will be designed to operate with the reservoir at elevation 1,540, a
120 foot operating head. The gate structures for Tunne 1 s No. 1 ail d 2 \vi 11
be designated to withstand external {static) heads of 130 and 460 feet,
respectively. The downstream portals will be reinforced concrete struc-
tures with slots for stoplogs (see Plate ).
{d) Final Closure and Reservoir Filli!!_g_
As discussed in Section 9 one of the diversion tunnels will be converted to
a low level outlet or emergency release facility during construction,
It is estimated one year will be required to construct and install the
permanent low level outlets in the existing diversion Tunnel No. 1. This
will require the that lower Tunnel No. 2 pass all flows during this period_
The main dam will be at an elevation sufficient to allow a 100 year recur-
rence period flow (90,000 cfs) to pass through Tunnel No. 2. This flow
will result in a reservoir elevation of 1625. During the construction of
the low level outlets, the intake operating gate in the upper Tunnel i'~o. 1
will be closed. Prior to commencing operation of the 1ow level outlets,
coarse trashracks will be insta11ed in the upstream intake structure in
slots provided.
Upon completion of the low level outlet in the upper tunnel the inta~e gate
will be opened and the low level outlets will commence operation. Upon
commencing operation of the low level outlets the lower Tunnel No. 2 will
be temporarily closed wi.th the intake gates ·and construction of the per""m-
anent plug will also commence upon filling of the reservoir. It is esti-
mated it will take 12 months to completely place and cure the plug.
During this time the main dam crest elevation will have reached an eleva-
tion sufficient to start reservoir impounding and have sufficient storage
available to store a 250 year recurrence period flood corresponding to a
reservoir elevation of 1,890 feet.
. During this time "the low level o~tlets will be passing the minimum summer
and winter flows of 6,000 cfs and 800 cfs.
12-9
T~.e filling of the reservoir will take 4. years to comp1Elte to full reser-
voir operattr~~ elevation of 2185 .. After 3 years of fi 11 i ng the reservoir
\'till be at flevation 2150 and will a11ow operation of the po~Jerplant to
commence ..
The filling sequence was determined from the main dam e 1 evat ion at that
time during construction, the starting reservoir pool elevation at that
time during construction. and the capability of the reservoir storage to
absorb the tnflow volume from a-250 year recurrence period inflow without
overtopping fhe main dam~ The 250 year recurrence period flood volume was
selected to be consistant with the recurrence period flows and risks used
for the design of the diversion and entire project.
This information is presented graphically in Figure 12.2.
Once closure of the upper d1version tunnel is made trashracks will be in-
stall.ed at the upsj:-."eam entrance to the tunne 1. The racks wi 11 serve to
prevent debris from being drawn into the intake and damaging the high pres-
sure slide gates.-The trashracks will be a permanent installation with no
provision for remoyal except with the assistance of divers.
The trashracks will have a bar spacing of about 3 feet and will be designed
for a maximum differential head of about 40 feet. To limit the maximum net
velocity through the trashracks to abo.ut 12 ft/s the racks wi 11 be semi-
circular. Provision will be made for the monitoring the head loss across
the trashracks.
12.5 -Emergency Release Facilities
The upper diversion Tunnel No. 1 will be converted to a permanent low level out-
let, or emergency release facility. These facilities wi 11 be used to pass the -
required minimum discharge during the 4 year reservoir filling period and will
also be used for draining the reservoir in an extreme emergency case.
The facility will have a capacity of 30,000 cfs at full reservoir pool and will
be capable of drawing the reservoir dam in 14 months. The reservoir drawrlo\vn
time incorporating the low level outlets presented graphically in Figure 12.2
for various "start 11 times during the year.
Ouri ng operation, energy wi 11 be desi gr.ated by means of two gated concrete p 1 ugs
separated by a 340 foot length of tunnel (see Plate 43).
Bonnetted type high pressure slide gates will be installed in the expansi0n
chamber tunnel plugs for the upper diversion tunnel, once closure of the tunnel
is made. The gates will be arranged in groups of 3 gates in series, each group
consisting of 1 upstream emergency gate and one operating gate in the upstream
plug and one operating gate in the downstream plug (see Plate 43).
The slide gates will be 7.5 feet wide and 11.5 feet high and will be ofwe1ded
stee 1 construction. The 9 ates wi 11 be designed to withstand a tot a 1 static head
of about 740 feet~ however will only be operat~d with a maximum head of about
12-10
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I.
I
I
I
I
I
I
I
I
I
I
I
I·
I
I
I
I
I
560 feet. To reduce energy dissipation problems, all three groups of gates \-Jill
be operated only v1hen the head is less 460 feet. ·
Uuring operation, the operating gate opening in the upstream plug will be equal
to the opening of the corresponding gate in the downstrean plug. This will
effectively balance the head across the gates~ The maximum operative head
across a gate will not exceed about 320 feet.
Each gate will have a hydraulic cylinder operator designed to raise or lower
gate against a maximum head of 560 feet. Three hydraulic units will be in-
stalled, one for the emergency gates, one for the upstream operating gates and
one for the downstream operating gates. The nominal operating pressure will be
2,000 psi. Each gate will have an opening/closing time of aoout 30 minutes. A
grease system will be installed in each gate for injection of grease between the
gate leaf and gate body seats to reduce frictional forces when the gates are
operated. Both local and remote operation of the gates will be possible.
As no facilities will be installed for dewatering th~ area around the emergency
slide gate, the design of the gate will be such that the hydraulic cylinder as
\'le 11 as the cylinder packing may be inspected and rep a ired without dewatering
the area around the gate.
An air vent will be installed at the downstream side of the operating gate in
the downstream plug. Air will be dr-awn from the access shaft. The vent \'fill be
heated as necessary to ensure that freezing will not occur.
To prevent concrete erosion, the conduits in each of the tunnel plugs wi11 be
steel 1 ined.
The upstream gate operating chamber will be separated from the downstream cham-
ber by a water-tight bulkhead door. In the event of a major failure of the up-
stream emergency gate, water would be contained \vithin the upstream gate-oper-
ating ch ~her.
12.6 -Nain Dam
(a) Comparison with Precedent Structures
The main dam at Watana, as currently proposed \vith a height of 8B5 feet
will be among the highest in the world. The highest embankment darns com-
pleted in North America are ~lica Creek in British Columbia {794 feet} and
Oroville in California (771 feet). Two dams under construction in the USSR
will exceed 1,000 feet, but the only dam completed to a height in excess of
800 feet is Sulak in the USSR. A list of embankment (earth and rockfill)
dams in excess of 500 feet completed, under constl~uction or proposed is
given in Table 12.2.
The Watana site is located in a seismically active area and the major· de-:
sign features of 24 embankment dams between 350 and 795 feet in height con-
structed in seismic areas are summarized in Table 12.3. The characteris-
tics of the \~atana design which will be developed in this section are
12-11
included in this table for comparison~ Special site conditions, depth to
bedrock, availability of materials, size of reservoir, s·ite location~ for
example, all have an impact on the design and such factors account for some
of the extremes quoted in the table.
A further comparison is given in Table 12.4 which includes the princip~l.
geometrical parameters of the core and outer slopes for high dams in
seismically active areas .. Considering these various parameter:
-The freeLoard ranges between 13 and 62 feet, with seven of the eleven
cases quoted being 1 ess than the 25 feet proposed for Wat an a.
-The crest width ranges between 33 and 111 feet. Wide crests are usually
the result of non-structural requirements, i.e., a roadway across the
dam. Neglecting extreme widths, seven of the ten cases quoted are
between 30 and 40 feet compared with the 35-foot width proposed for
Watana.
-The core \'!idth ratio ranges between 0.29 and 0.56, with only Gne example
higher than the 0.50 ratio proposed for Hatana.
-The upstream slopes range between-2.u:l and 2.7:1. The Japanese dams
tend to have flatter slopes (\vithin the range 2.5 to 2.7), while the
North American dams are in the range of 2.0 to 2.6. The Watana s1opeof
2.4:1 is among the steepest, but is flatter than the next two highest
dams, ivlica at 2.25:1 and Chicoasen at 2.2:1. However, special features
are included in the Watana design, primarily the use of free-draining
pro·:essed gravel in the upstream shell, to minimize the effects of earth-
quakes on the stability of the upstream face of the dam.
-The downstream slopes range between 1.8:1 and 2.7:1. Ten of the 15 cases
quoted are equal to or steeper than the 2:1 slope proposed for \~atana,
while only one case is flatter than 2.2:1.
Complete details of the core materials used in all the dams list~d in the
accompanying table~ are not available in the literature. However a number
of large dams· have been constructed in Canada using similar glacial depos-
its as core material. The mean grad{ng curves for these materials includ-
ing that used for the core of the t11i c a Creek dam, the existing dam ge!1er-
ally comparable to Watana in size~ materials, and location, are compar·ed
with the mean grading curve for the core material proposed for the tiatan a
dam in Figure 12.5. It is apparent f)•om this figure that the ~latana core
material is \<Jell within the t'ange of materials used successfully for other
large dams in North America.
In summary, the proposed Watana design is generally conservative with re-
spect to precedent design. Howeve.r, special features tthich are discussed
in more detail later in this section are incorporated in the Hatana section
to provide additional safeguards against seismic loading.
12-12
"
•••
I
I.
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I.
I
I·
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I ..
I
I
(b) Excavation and Foundation Preparation·-General
The geology of the Watana site is described in Section 9. In summary, the
existing conditions at the damsite comprise alluvial deposits in the river-
bed up to HO feet deep overlying bedrock, while the lower slopes of the
valley are covered \vith ta 1 us and there is overburden on the upper s 1 opes.
The bedrock is jointed and \veathered at the surface with weathering along
joints extending to considerable depths. Locally in shear zones and drain-
age gullies the rock is weathered throughout to depths in excess of 40
feet. The frequency of joints and fractures generally decreases with depth
but fractured and weather zones have been identified 1ocally at depths up
to 200 feet. , Zones of permafrost occur, particularly in the south abut-
ment.
The dam foundation must satisfy the following basic requirements:
-The foundation under the core must be stable and capable of supporting
the weight of the core under all loading conditions, must not erode under
the seepage gradients which wi 11 deve 1 op under the core, and must not
allow excessive seepage losses under the core.
The foundation under the upstream and downstream shells must be stable
and capable of supporting the weight of the darn without excessive move-
ment under all loJding conditions.
-The core material must be pr·evented from moving down into the foundation
(e.g. into cracks ~r joints) and then through the foundation under the
transition zone into the downstream shell or beyond.
-The contact bet\veen the core and its found at ion must remain watertight
despite the distortions that will occur in the dam because of its uwn
weight and the thrust of the reservoir.
-Any seepage through the foundation must be contra 11 ed and discharged so
that excessive seepage pressures do not develop in the downstream port ioo
of the cm"e, in the materia 1 s beneath the she 11, or do~m stream of the
dam.
The excavation and foundation preparation necessary to meet these objec-
tives are outlined in the following paragraphs.
(c) Excavation
(i) River Excavation
The properties of the river alluvium are not well defined but it
includes sands, gravels, cobbles, and boulders up to 3 feet or more.
Such materials are not suitable as a foundation for the core:; pri-
marily because of. their relatively high permeability. Such alluvial
deposits have been left in place under both upstream and downstream
12-13
..
I
shells of many of the world's largest dams (see Table 12.3). How-••.
ever, at Watana .these granular materials could undergo 1 iquefaction
under seismic loading with potentially catastrophic results. In-
sufficient data is available to demonstrate thatthere is no pos-•...
sible risk ot liquefaction of the alluvium, but further investiga-
tions may provide data to support the concept of removing the allu-
vium only under the central portion of the shells. However, in view 1 of the high seismicity of the area it is proposed that the river .
alluvium be removed over the whole foundation area.
( i i) . Under Core and Fi 1 ters I
The core and filters must be founded on sound rock to ensure that no
material can wash through open joints. This will require excavation
of overburden and talus on the slopes and v1!2athered rock in the
valley bottom and on the abutments. The talus thickness on the
abutments perpendicular to the slope varies from zero to an esti-
mated 20 feet and weathered rock to 40 feet or more in some areas,
Weathered rock is here defined as closely jointed or fractured rock
with weathering and infilling. of the joints. The final foundation
will be sound hard rock with only minor weathering, which can be
grouted to ensurt. that core material cannot be washed through joints
in the rock~
The maximum rock slope along the abutments is determined to some ex-
tend by the valley shape. In general, lH:lV slope or flatter is
ideally preferred although steeper slopes have been used. At Watana
damsite, the natural slopes at lower elevations are relatively
steeper but still genera 11 y less than lH !2V. It is therefore pr·o-
posed that the overall core foundation slopes will be no steeper
than 1H:2V below elevation 1800 and lH:lV above elevation 1800. The
cross cut slopes will be lH:lOV.
Local irregularities in the rock surface are undesirable because of
the potential for differential settlements or strains in the core
that could cause cracking and potential piping through the core~
Such irregularities also make it difficult to compact the core
material to form a tight core-rock bond and they must be eliminated
either by additional rock excavation or the addition of concrete to
achieve an acceptable slope. Such slopes would normally be on the
order of lli: 2V in the 1 ov1er sections of the dam where cant act ores-
sures are higher, flattening to 1:1 at higher elevations. ·
The depth of excavation required to remove unsuitable rock will vary
considerably over the core contact area. In scme area very little
excavation may be needed, while in highly \'v'eathered zones excavation·
may extend to 50 or 60 feet. On the basis of available data., it is
estimated that the average excavation under the core and filter-s
will be 40 feet.
12-14
I
I·
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I·
·I
I
II
(iii) Under Upstream and Downstream Shells
The shells will be founded on competent rock. Loose or detached
rock or rock ribs ana highly ~eathered rock will be removed to ex-
pose sound rock. Weathering along joints and local irregularities
in the rock surface will be acceptable. The actual thickness of
rock to be excavated vtirl vary across the site; but it is estimated
that the average will not exceed 10 feet.
t •'~J Dental Excavation
Dental excavation over and above normal excavation is expected in
zones of intense shearing or highly irregular surface£. Whereas th~
need for such excavation has been identified by investigations com-
pleted to date, the magnitude has not been properly assessed because
of heavy vegetation, tundra cover, and general lack of outcrops and
access problems.
(v) Excavation Methods
It is expected that the excavation of the overburden material within
the dam foundation will be performed as a multi-level operation us--
ing wheel~d loaders working with dozers. ~oulders that cannot be
removed by excavation equipment will be ,blasted. On the steep work-.
ing areas wi 11 be formed with material excavated from the slopes
above. These working areas will be progressively lowered removing
overburden and weathered rock in one operation.
The excavation of the foundation will need to be complete for safety ,
reasons from about Elevation 1800 down to the riverbed before plac-
ing of fill is commeliced. The excavation on the upper slopes will
then require to be kept sufficiently in advance of grouting and f il1
placement to avoid interference of these activities by the blast~ng.
Excavation of unrippable weathered rock and trimming of the r~ck
surface to acceptable slopes will require blasting which will l"e-
strict other activities such as surface grouting.
Numerous access roads will be required throughout the dam area to
reach the various working levels. Dental excavation will be none by
small backhoes and final cleanup of the area under the core and
filters will be carried out to a high standard by hand with high
pressure water jetting prior to grouting. The rock surface under ·
the core and filters should be clean enough for detailed geolugical
mapping and for grout leaks to be observed and caulked if necessary.
The foundation must be free from snow and ice before fill material·
is placed.
Selected alluvial material from the riverbed may be used in the
downstream shell of the dam but the remaining material, generally a
mixture of weathered rock and overburden, will be wasted or used for
road, cofferdam or temporary facility construction. Spoil areas
will generally be below final water level in the reservoir a~ea.
12-15
(d} _§;outing and PressureR?lief
A combination of consolidation gr·out.ing and cutoff curtain grouting under
:~he core and a downstream pressure relief (drainage) system are proposed
tor the ~atana site. Those systems will result in:
Improved stabi1ity of the foundation;
Reduct ion in rock mass permeability and hence seepage through the founda-
tion;
Reduction in the risk of movement of soil particles through joints in the
rock; .and
-Contra 1 and safe discharge of any seepage flows through the grout
curt a in.
It is proposed that the curtain grouting and drilling for the pressur·e re~
1 ief systPm. are carried out from galleries in the rock found at ion in the
abutments and beneath the dam: Det a i 1 s of the grouting, pres sure re 11 ef
·and galleries are shown on Plat~ 46.
-
The purpose of grouting is to improve foundation and abutm~nt rock condi-
:ions with respect to load bearing and seepage considerations. Tte n~od,
e:tent, and detail of grouting is dependent on site geological condit10ns,
type, and character of rock, reservoir head~ and locaticn of specific
structures. The diorite bedrock at Watana is competent as far as load
carrying capacity is concernerL However, numerous shear zones from \1 few
inches to several feet in width, have been i~entified in a general NW-SE
direction. Occasionally, the width of shear zones may be ~everal tens of
feet locally. Most of these zones, which are found both in the river than-
nel and in abutments, contain gouge material and under appropriate condi-
tions, may be susceptible to piping .. These features are discussed in ffiOl'e
detail in Section 9.
The permeability tests in boreholes indicate the rock ma~s permeability at
the Watana site to be generally in the range from 1 x lOb em/sec to 1 x
lu-4 em/sec, indicating a maximum seepage rate through the foundation
of t~e order of 4 cubic feet per second. However, these permeab i1 ity
values may not properly account for shear zones. For example, in Bm~ehole
BH-2 on the north abutment, circulation was lost during drilling when the
boring encountered a shear zone. Also, because of heavy vegetation 3 talus
cover and limited access: it is possiole that there may be other shear
zones not yet identified. A properly co~1ducted primary grouting program of
an exploratory nature will be required i'i1der the dam and in the abrtments
and, depending on the results of this rrogram, additicaal grouting includ-
ing multiple 1 ine curtains rnay be required. ·
12-1 f
I
I .-
1
I
I
I
I
I
I
I
I
'
I
I
I
-1
I
I
I
I
I
I
I
I
I
-I
I
I
I
·t
I
I
I
I
I
(i) Consolid,a~ion Grouting
The rbck under the core, upstrean filter, and downstream filter will
be consolidation grouted to provide a zone of relative1y impermeable
rock under the entire contact. Locally, the rock may be sound and.
free of any discontinuities resulting in virtually no, grout take;
rievertheless, the Joints and shear zones are generally steeply dip-
ping and any particular vert ica1 plane is 1 ikely to intersect these
zones which are estimated to be lS to 20 feet apart. Consolidation
grouting is estimated to require 30 foot deep holes on a 10 foot by
10 foot grid.
(ii) Curtain Grouting
The design of g~out curtains under dams is largely empirica1, though
based on data from boreholes. At the Wa.tana site, only borehole
DH-21 extends to a significant depth below the river to elevation
876 -feet, approximately oOO feet be1oVJ dam foundation level. Shear-
ed and highly fractured zones are indicated at an aver~g8 of 50 feet
intervals to the bottom of the hole and the upper zones ~hou 1 d be
grouted to reduce seepage losses. The average rock permeability de-
creases significantly around 200 feet depth~ A grout curtain is not
-expected to be 100 percent effective in eliminating seepage, but
rather to increase the length of the seepage path. Flow net analy-
sis indicates that a positive ct..toff of 350 feet deep increases the
potential flow path by a factor of 1.7, decreasing the average
hydraulic gradient from about. 0.9 to 0.5.
_For the purposes of this study, a double }~ow grout curtain to a
depth of 0. 7H, where H is the head of water behind the dam at ::1 par-
ticular location, \'lith a maximum depth of grout curtain of 35u feet
has been assumed. Grouting \·lill be carried out from a series of
underground galleries which will also serve the drainage syste~n
pressure relief.
It is likely that in some areas the grout take at depth will be very
lov1. Primary holes will be considered as exploratory holes a!na will
be core dri11ed. On the basis of the core and vtater pressure tests
in the exploratory holes, the depth of secondary holes can be ae-
cided. The exploratory holes may also identify areas that need
additional grouting.
All h0les will oe water pressure tested in stages and the grouting
program will be determined using these results. Grouting will be
c-arried out using split spacing with the primary holes at 40 feet
spacing. The secondary, tertiary and quaternary holes \·/Ould hr·ing
the final hole spacing to 5 feet if required.
~ Permafrost in the area to be grouted will have to be thawed befor"e
water pressure testing and grouting can be done. The greatest depth
of permafrost so far recorded was in Bh-8 where the hole froze up to
12-17
0
. .
175 feet depth, Additional boreholes may be required for the thaw-
Rock wi 11 be
ho 1 es.
ing of sufficient rock to form an effective curtain~
thawed by circulating clean river· water through -drill
The effect1veness of the initial thawing and gtouting may be diffi-
cult to assess but the permanent galler·ies under· the dam will enable
additional grouting to be carried out at any time during the follow-
ing reservoir filling and subsequent thawing of the foundation and
abutments.
It is desirable for
uities as possible.
from 80° to 60° and
drill the boreholes
the grout holes to intersect as many discontin-
The dip of the main joint sets and shear range
its is therefore considered most efficient to
vertically or at an angle of 45°.
A major shear zone approximately 600 feet wide trending in a NW
direct ion intersects the left edge of the dam and reservoir area and
the curtain should extend into the abutment to provide.a positive
cutoff of this zone. The ground surface rises to the south of the
dam and the surface express ion of any shears to the south of the
major zone will be outs ide the reservoir area and are un 1 i kely to
cause appreciable seepage.
The extent of the grout curtain through the shear zone will be de-
termined by t:xploration from the grout gallery.
Artesian water pressure was observed at the shear zone in BH-12 in-
dicating that materials with high permeabi1ities exist in the shear
zone which must be effectively grouted.
No major shears have been found on the right abutm 1t where the rock
is of good quality. The grout curtain will extend ·rom the spillway
intake structure 400 feet into the abutment with the depth of the
curt a in set at a minimum of 200 feet.
The spillway control structure is located on the dam centerline-and
the grout curtain will extend beneath the structure with drilling
and grouting from the gallery formed within the concrete· roll·Nay.
Drainage will be provided behind the grout curtain with holes
drilled from the gallery.
(i·H) Drai!§ge and Pressure Relief
-I .
I
I
I
I
I
I
I
I
.....
II
I
I
I
I
Drainage features are included beneath the dam foundation and the I
abutments to intercept seepage through the grout curt a in and re 1 i eve
pressure. Ccm~on drainage and grouting will be ~onstructed with
grouting from the upstr·eam side and drainage from the downstream ·•
side of the galleries. The use of galleries is recommended for the
following redsons:
12-18
I
I
.,
I
I
I
I
I
I
I
I
I
I
I
I
I ( i v)
I,
I
I
·I·
I
-Curtain grouting from the gallery can be carried out independently
of the construction of the darn. This can shorten construction
times.
-The grouting can continue longer into the winter than would have
been possible with surface grouting.
.... Permanent access is av a i 1 ab 1 e under the dam for inspection.
Additional grouting or drainage holes may be drilled after
construction of.the dam which is an important consideration where
there is permafrost. The tha\'ling effect of the reservoir may
require r·emed i a l grouting after impounding has commenced.
-
-Higher grout pressures can be used if required because of the
overlying weight of embankment.
-Drainage holes drill€d dmvnstream of the grout curtain \vill be
discharged into the gallery enabling flow from individual holes to
be monitored. This system will prevent the outlets of the
drainage.holes freezing which is an essential requirement.
Gallery drainage is more effective because pressures are relieved
at a lower l eve 1 .
-The galleries may be used for the installation of instrumentation
and provide access for the repair and replacement of
instrument at ion.
:>
-Tunneled galleries provide the great advantage as an explor-ation
tunnel for the rock of the dam found at ion. The tunnel gives the
best opportunity for understanding the nature of the rock along
the grout and drainage curtains which will be invaluable in the
f au 1 ted and sheared zones.
The drainage/pressure relief holes will be drilled after all gt'OUt-
-ing is complete. They will be 3 inches in diameter spaced at
approximately 10 foot centers. Generally the ho 1 es wi 11 be open but
any penetrating fractured or sheared rock may require perforated
casing to prevent caving.
Construction Methods -Grouting and Pressure Relief
The schedule of work is of particular importance in this phase of
the work. The excavation for the g a 11 er i es must be carried out be-
fore consolidated grouting because the grouted rock mo.y be disturbed
by the blasting for excavation. It will also be preferable to com-
plete excavation of the dam foundation in a particular section be-
fore excavation of the gallery so that the surface rock profile may
be confirmed before tunnelling.
12-19
Hock temperatures will be measured any any areas of permafrost
thawed prior to .grouting. Grout holes 't/ill generally be 1-1/2' inch
in diameter~ Large hole sizes wil1 be drilled where explotatory
cored holes are required or do\vn-the-hole hammer equipment is used.
All holes will be washed and pressure tested before grouting.
Grouting will be done with Type II Port1 and cement with 2 percent
addition of bentonite (by weight of cement). The water/cement ratio
and grouting pressures will be varied accordirtg to the conditions
encountered. Grouting will be carried out in stages using packers.
Some redrill ing between stages will be required. To allow greatest
flexibility of the ~chedule, most curtain grouting, which will in-
clude up-hole grouting will be done from the galleries. In the in-
clined galleries special platforms will be required for drilling and
grouting equipment.
Primary grout holes will be treated as exploratory holes and core
drilled with further core drilled holes as required to test the
effectiveness of the grouting. The grouting progr ain will be modi-
fied according to the rock conditions encountered as the work pro-
ceeds.
(v) Gallery Construction
The layout of the galleries are shown on Plate 46. The. horizontal
and inclined tunnels will be excav(lted by conventional drill and
blast methods. Vertical shafts will be raise bored providing a
smooth excavated profile with little support required. It is ex-
pected that the majority of the gallery length \vill not require any
support but from avail able geologic data it is estimated that aoout
25 percent will require rock bo 1 t and shot crete support.. Stee 1
arches vlill be required at the portals ano at tunnel junctions or in
highly fractured Ot" shear·ed zones. A concrete slab will be c~st in
the tunnel invert to provide an even working surface and to fo~"r.l the
drainage channel.
Measuring weirs will be cor structed in the drainage channels i~
order that the volume of sewage water may be monitored. The seepage
water will be discharged from the gallery just above tail water l eve 1
through drainage tunnels extending to the downstrearn toe of ··he dam.
The drainage outlet of these tunnels \vill be located under taih1ater
level to prevent icing up of the outlet. Inspection access will be
provided at the down stream toe of the dam but from a separate porta 1
above water level.
Lighting for inspection of the galleries and ventilation wil1 te re-
quired. The fresh air intake during the winter must be heated to
prevent freezing of seepage water within the tuPne1 s. The ventil a-
t ion will only be required occasionally when personnel are in the
tunnels. Elevators will be installed in the vertical shafts to-
gether with emergency stairs and cable hoists installed in the in-
c 1 i ned tunne 1 s for· movement of equipment.
12-20
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
{e) Main Dam Embankment
The main dam v1ill consist of a central compacted core protected by fine and
coarse filters on the upstream and do~tmstream slopes. ·The downstream outer
shell wi11 consist of roc'f< fill and alluvium gravel; and the upstream outer
shell of clean alluvium gravel .. A typical cross section is shown on Plate
45~
(i) Comparison of Vertical and Inclined Cores
The design of the embankment is dependent on the type of core chos-
en, either a vertical core or an inclined core, and its location,
upstream or central in the embitnkment.
The advantages to each type of cote are as fo l1mvs:
-Vertical Core
Provides better contact with the foundation;
Provides slightly more thickness of core for the same quantity of
the core material; ana
Settlement of the core will be independent of the post-
construction or seism.ic displacement settlement of the do\vnstream
shell.
-Inclined Core
Can place bulk quantity of downstream shell before p1acing core
material; and
Can carry out foundation treatment duri119 placement of she1~
material.
The major disadvantages for each type of core are as follow-s:
-Vertical Core
Placement of core material controls placement of filters ana shell
mat~rials; and
Possible arching of a thin core oy transferrins weight to c;~jacent
filters and shell materials during settlement or seismic d~sp1ace
ments.
-Inclined Core
Excessive ~est-construction settlement or seismic displacment of
downstream shell may cause rupture of core; and
Locatiorr of core may effect upstream slope by ma~ing it flatter
for stability reasons.
'1? 2 .. 1~:.--I
A central vertical core was chosen for the embankment based on a re--
view of precedent structures discussed above and.the nature ~f the
proposed impervious material.
The proposed impervious material is a combination of glacial outwash
and tills with a \'lide grain size distribution~ This material is
nonplastic and would tend to crack rather than deform under tensile
stress and-hence may be susceptible to erosion. For a sloping core
the possibility exists of cracks developing in the core for a non-
plastic material because of lateral settlement or displacement dur-
ing a seismic event. It also becomes difficult to avoid high ten-
sile and shearing stresses in an inclined core. Settlement data
indicates that the magnitude of water load settlements in rockfill
dams may increase at a rate greater than direct pro port ion to the
height of the dam. For these reasons a central vertical core will
be used in the \~a tan a Dam cross sect ion.
(ii) Earthquake Resistance Design Features
Bee a use of the apparent lovl p 1 ast i city of the materia 1 to be used in
the impervious core and the requirement for an earthquake resistant
design, the following design features will be incorporated into the
main dam cross section:
-The cor--e-foundation contact will be widened near the ends of the
embankment to ensure seepage control during normal operating con-
ditions and any seismic event.
-Thick filter zones will be placed upstream and downstream fr..,m the
impervious core to prevent breaching of the core from either post-
construction settlement and cracking or from any cracking result-
ing from a seismic event.
The filters will be designed to be self-healing in case of trans-
verse cracks in the core resulting from either post-construction
settlement or a seismic event.
-The downstream filters will be designed to be capable of hana1 ing
any abnormal flmvs which could result from transverse crack~ng at
the core from post-construction settlement or a seismic eve~t.
-The proposed ~tlidth of the core will prevent arching of the core
caused by transfer of load to the shell or filter materials.
I
I
·I·
I
I
I
I
••
I
··,
I
I
-Compacted river alluvium gr·avel and rock fill vlill be usect to con-
struct the downstream outershell. Compacteo processed clean rivet
alluvium gravel of high permeability will be c~~d to construct the
upstream outershell to minimize settlement displacem~nt ana the
build up of pore pressures during a seismic event and to ensure I
rapid dissipation of any pressures which may occur. , ·
12-22
. . _, . . .
,,
I
I
f -. . . . , . . . . . .
I
I (iii)
I
I
I
I
I
I
I
I
I
I
I
I
:I
I • ,
I
I
··~ ..
Freeboard and Static Settlement
~--"
't'he minimum required crest elevation of the ~·1atana Dam, not includ-
ing static and seismic settlement, was determined for each of the
following conditions:
Norma1 maximum reservoir
el ev at ion
Storm surcharge
Hater e 1 ev at ion
Wave runup allowance
Dry freeboard allowance
Elevation top of cbre
Roadway thickness
Minimum crest elevation
1:50 Year
Storm
2185
b
2191
6
3
2200
3
2203
1:10,000 Probable
Year Storm Maximum Flbod
2185
8
2193
6
2199
3
2202
2185
17
2202
3
2205
These elevations refer to the maximum section of the dam and are
based on a normal operating reservoir level of 2185 feet. The gov-
erning minimum crest elevation excl.',..lding static and seismic settle-
ment is 2205 feet at the maximum sect ion and at the abutments,
This is the lowe5t elevation acceptable for the dam crest and allow-
ances must be made for static settlement of the dam following its
cnrnpletion, settlement on saturation of the upstream shell, and pos-
sible slumping because of seismic loadin9. It has not been· pcssible
to perform detailed calculations at this time to determine the like-
ly settlements since no test data is available.
For purposes of this feasibility study, it has been assumed that
seismic slumping could be up to 0.5 percent of the height of the dam
and the design crest elevation at the center of the dam is, t~nere
fore, shm'ln at 2210 feet, 5 feet abuve·, the required minimum e:~va-
t ion of 2205 feet. An allowance of 2 feet has been made at b~th
abutments and hence the design crest elevation rises from 2207 at
the abutments to 2210 feet at the center. Thus, under normal nper-
ating conditions the minimum fr"'eeboard relative to the maxim~t.ri oper-
ating poo 1 e 1 ev at ion of Llo5 wi 11 be 22 feet at the abutments and 25
feet at the center of the dam.
If for any reason the crest settles below 2210 feet; more material
should be added to maintain the safety margin-of 5 feet to al 1 cw for
seismic slumping .
12-23
...
An additional allowance must also be made for post-construction
settlement of the dan under its own \'Ieight and because of the
effects of saturation on the upstream gravt:1 fill when the r·eservoir
is first filled, This allowance is not shO'tll1 on the drawings since
it is not a permanent requirement~ However~ for initial estimating
purposed;, 1 percent of the hei~ht of the dam has b.een allowed~
Hence, at the end'of construction the dam crest at the center-of the
dam would be at 2,210 feet plus 9, or £..,219 feet. The additional
height constructed into the dam wou1d be achieved by steepening both
slopes above approximately elevation 1850.
Further margin agq.inst overtopping of the mafn dam is provided by
the emergency spillway. Under normal operation this spilhvay is ·
sea 1 ed by a fuse p 1 ug or dam across the entrance channe 1. This p 1 ug
is simply a gravel dam with special design of the core and strict
contra l of the she 11 mat_~r i a 1 s to ensure that it wi 11 et·ode rapidly
when ov~rtopped, allowing flood flows to be discharged freely down
the emergency spillway.
'
The location and typical cross section through the fuse plug are
shown on Plate 53. The fuse plug has a total length of 310 feet and
a height of 31.5 feet. A pilot channel 20 feet wide with an invert
1.5 feet lower than the crest, is provided at the center of the fuse
plug, to start the washout at a predesignated location. The. loca-
tion of this pilot channel is chosen so that the flow of water dur-
ing washout vd 1l be smoothly channeled into the spillway chute •.
{iv) Typical Cross Section
The typical cross section of the main dam is shown in Plate 45. The
central impervious core slopes are 1H:4V wlth a crest width of 35
feet. The thickness of the core at any section will be slightly
more than 0. 5 times the head of water at that sect ion. Nin imwn
core-foundation contact will be 50 feet requiring flaring of tae
cross section at each end of the embankment.
The upstream and dmvnstream fi 1 ter zones increase in thickness from
15 or 30 feet near tne crest of the ctam to a rn .dmum in excess of 60
feet. They are sized to provic .. protection against possible piping
through transverse cracks in the core that could occur bt::cause of
settlement or resulting from disp1ao"'"::;-2nt during a seismic event.
The shells of the dam will consist of compacted alluvium gravels.
To minimize pore pressure generation and ensure rapid dissipation
during a seismic event, the saturated upstream shell will consist of
compacted clean alluvial gravels pr'Ocessed to remove fines so that
not more than 10 percent of the materials is less than 3/8-inch in
size. The dmvn£tream shell will consist of compacted unprocessed
alluvial gravels and rockfill from the excavations for unctarg}~ound
v10rk s i nee it ltd 11 not be effected by pore pressure generation
during a seismit event.
12-24
I
I
I,
I ,:
I
1:
I
I
I
I
I
I
I
IJ
I
I
I
I
·I
I
'I
. I
I
I
I
I
••
I
I
I
I
I
I
I
I
I
Slope protect ion on the upstream s1 ope wi 11 consist of a 10 foot
zone of oversize material up to 24 inches in diameter, placed and
compacted by suitable equipment.
The typical crest detail in shown in P1ate 45 .. Because of the ttar-
rowing of the crest dam, the filter zones are reduced in width ana
the upstream and downstream coarse filters are el iminat>:?d, A 1 ayet
of filter fabric is incorporated to protect the core material from
dam age from frost penetration and dessication, and to act as a
coarse filter where required .
(v) Core Material Properties
The core material wi1l be. obtained from Borrow Ar~t:a D, locatec on
the right bank of the river, upstream from the dam. The area con-
sists of a series or glacial deposits separated by alluvial an.: la.-
custrine deposits. A generalized surficial stratigraphic colu~1 in
Area D, based on all investigations to date including seismic lines
and deep drilling, is given in Table 12.5.
Typical gr·ading curves for each unit are presented in Figures 12.6
through 12.10, which also indicate the range and average moisttire
content of each unit. It is proposed to blend material from the
various units as required to provide core material with a max~~um
particle size of 6 inches Qnd within' specified limits of moi~ture
content, gradation (as shown in Figure 12.11) 3nd plasticity~ The
composite gradation curve from Borrow Area 0 is shown in fig~n::
12.12.
The Atterberg limits will be within the following ranges:
-Plasticity Index-0 to 20
-Liquid Limit -10 to 45
Permeability tests indicate a permeability on the order of 10°
em/sec, which is within the normal range for glacial deposits ~sed
in simi 1 ar dams.
Modified Proctor Compaction tests on material passing 3/4-ir . .:::~i sit:~ve
indicate an optimum moisture content oA' 7.5 percent with a ma;{·hnum
dry density of 135.5 pcf. Standard Proctor compaction test r~sults
on material passing No. 4 sieve indicate optimum moisture cont~nt of
10.4 percent with a maximum dry density of 127.6 pcf. The tes~ re-
sults are plotted in Figures 12.13 and i2.14.
The natural water cont~nts of samples tested range from 7 to 21 per-
cent with occasional samples of finer grained material up to 4u per~
cent. Blending and processing of the core material will be ne~es
sary vlhile pockets and layers of very ~1/et material ',vill be left in
the borrow areas or otherwise wasted.
12-25
Consolidated undrained test results (see Figure12.15) at 95 percent
~1odified Proctor density and 2 percent above optimunl moisture con-
tent, indicate the angle of shear strength resistance {VJ) equal to
37~, with a cohesion intercept (c) of zero~
Pinhole dispersion tests inaicate that the proposed core matet'ial is
a non-dispersive mttterial.
Consotidation tests indicate a compression indec (Cc) of o .. Ob and
test results are shown in Figure 12.16.
(vi) Excavation, Placement~ and Compaction of Gore Material
-.
The borrow area wi 11 be exc ~v a ted to a depth of approximate 1 y 30
feet working vertical faces. Processing and blending of the mater·i-
al will be done during excavation. Oversize material (greater than
b inches) \'/ill be removed by grizzlies or raked out of' the fill dur ...
i ng spreading. Frozen mater i ~1 \<Ji 11 have to be 1 eft in place or .
loosened by blasting and ripping for haulage to waste area. ~1ois
ture conditioning will be done in the borrow area.
Material will be placed in 8-inch compacted lifts at a maximum mois--
ture content of 3 percent above optim~m moisture content, and com·
pacted to 95 percent of the maximum density obtained from the ::;~an
·ctard Proctor test. Type of roller, number of passes, thickness of
1 ift and moisture content can be adjusted basr=d on field tests and
equipment to be usea.
(vii) Fine and Coarse Filter Materials
. Fine and coa)~se filter matflria1 wi11 be obtainE:d from narrow ;'-\raa E:,
The material wi11 be proce:;.;se· to provide ths:; required gradat ivns.
Frozen material will> wh~~0 ~ossible, oe allowed to progressively
thaw insitu, with a system of surface ditches to accelerate .. :a"ainage
of the thav1ed material. Where this is not practical for scheduling
reasons or other considerations, the foreign material will be loos-
ened by ripping or blasting and hauled to a disposal area. ~nisture
conditioning will be done in the borrow area.
Criterion 1: The 15 percent (015) of a filter material must :Oe not
more than five times the 85 percent size (085) Df a
prot2cted so i 1 .
Criterion 2: The 15 percent size (015) of a filter material should
be at least five times the 15 i)ercent size {015} of. a
protected soil .
Criterion 3: The 50 percent size (D5U) of a filter material must be
not more than 25 times the 50 percent size (050) of a
protected so i 1 ~
12-26
I
I
I
I
I
I'
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
'cl
I
I
I
I
I
I
I
The required gradations of the fine and coarse filter material to
satisfy the above criteria are sho~'fi T Figure 12.11, whi1e compos-
·.· i te gradation:; for Borrm'l /l,rea E a 1M1 in Figure lc .ltL
Permeab i 1 i ty of the fine f i1 ter and ... ;se filter is estimated to be
greater than 1 em/sec and 10 em/sec, respectively. Permeability
will be verified by large scale field or laboratory tests.
The fine and coarse filter material are assumed to have an angle of
shearing resistance (0) of 35:. for the purposes of these studies~
Actual properties will be determined from large scale triaxial tests
ard/or modeling the gradation for standard triaxial tests for final
de~ign.
(viii) EY.:avation, Placement,; and Compaction of Filter Material
The borrow areas will be developed utilizing scrapers and dragli·,~s
which will supply the required amounts of fine and coarse fi1ter
material construction. Material will be processed by screening and
blending Gsing wet screening methods. Oversized material will have
to be removed and either used as an aggregate source or possibly
used in the outershell of the dam~
The method of placement and compaction will depend on the r.c~;.i 1 ts of
full scale test fills to be done prior to construction using the
proposed equipment ana materials. It is assumed that 12-inch lifts
with four passes of a large vibratory roller will provide there-
quired compaction.
( i x) A 11 uv i a 1 Fill l'tlat erial
The alluvial fi 11 wi 11 be obtained fro'il Borr0\'1 Areas E and I, The
upstream shell of the dam will be constructed using processed allu-
vial gravel and the downstream shell of unprocessed alluvium fill
material mixed \'lith rock from the various excavations, when a\rail-
able. Any oversized material (greater than 18 inches) will be
either used in the rip-rap zones or crushed for concrete aggreg·ate.
The gradation of the available alluvial fi11 material will be as
shown in Figures 12.16 and 12.17, while the: required grading limits
for the upstream shell are shmvn in Figure 12 .11. The do~tm·st.~-.cam
shell material will not require proces· ing. ~·!aximum size of r·iver
gravel will be 18 inches in the greate5L dimension.
Permeability of the processed alluvial fil1 is estimated to be
greater than 100 em/sec.
An angle of shearing resistance of ~ equal to 35Cl has been asstnned
for the alluvial fill material.
12-27 •
I
Actual properties ~.,n1 be ·determined from 1arge scale triaxial tests ·I
and/or modeling the gradation for standard triaxial tests for final
design.
(x) Excavation, Placement, and Compaction of Alluvial fill Naterial
The alluvial f i 11 materia 1 wi 11 be obtai ned from the main dan foun-
dation excavation and from dO\'!nstream from the main dam. ~~ethod of
excavation will be by scraper operations above the water table and
dragl ine operation below the water table to a maximum depth of ~0
feet. The mater i a1 wi 11 have to be processed to remove the undet·-
s ized and oversize material for the upstream she11.
All material in the si·:..ll:; must be \1/ell compacted to minimize post-
construction settlement and seismic slumping. The method of p1are-
ment and compaction will depend orythe .results of full scale test
fills to be done prior to construction using the proposed equipment
and material~ It is asst:n'ed that 24-i(,.:t, lifts for alluvium fill
material with four· passes of a large v1bra:-ory roller will provioe
the required compaction.
(xi) Rip-Rap Material
The rip-rap material (slope protection) will be obtained from tne
oversize material from the various borrow areas~ Quarry A and an,y
other rock excavations. The rip-rap material will be placed an the
upstream slopes and in c~t'ta1n areas on the dovmstream slopes of the
dam exposed to wave action.
The gradat·ion of the rip-rap matefial, Figure 12~11, is based on a
6-foot vJave heigr.t us'!ng a r;omorgraph, Figure 5-6, from Eivllll0-2-
2300. The maximum size of rip-rap material will be 24 inches .. It
is assumed that 36-inch lifts for the rip-rap zone with four passes
of a large vibratory roller will f.:!Ovidt! the required compact1tm~
(f) Stability Analysis
(i) Methodology
Static and dynamic stabi 1 it_y· analyses have been performed to cnnfirm
the stabi1 ity of the upstream 1d downstrearn slopes of the Watana
dam. Ttie analy5HS indicates stable slopes under a11 conditions fo'."
a 2.4 horizontal to 1.{; vertical upstream slope and a 2.0 horizontal
to 1,0 vertical down~t""eam slope.
The static analyses have been done using the STr.UL computer p~~O!}r'am
developed to handle general slope st~bil h.y prob1ems by adaptation
of the ~lodified Bishop method and a finite element program fJr sta-
tic ana1ys·is of earth and rwckfill dams (FEADAf'l) to determir1e the
initial stress•"s in the dam during rormal operating conditions. The
r~esults and c~nclusions from both the static and dynamic analyses
ar~ given in Appendix D.
12-28
I
I
I
I
I
I
I
,I
I
II
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
••
I
I
I
I
I
I
I
1:
,I
I
The dynamic analyst:; have been done using the QUAD 4 finite element
program which incorporates strain dependent shear modulus and damp-
ing parameters. The design earthquake for the dynamic analyses was
developed for"' a Benioff zone event.
An asses$f!ient of the static and· seismic response of the l,;latana dam
for the static and postulated seismic loading involves the follow--
irg:
-Finite Element Model
The finite element model consists of 20 lt1yers of elements \'lith
approximately 550 nodes and 520 elements. uif·Ferent soil para-
meters as described in the following sections have been chosen for
the core, transition material, and the shell material. The tran-
sition material will comprie the fine and coarst:: ·filte_r zones.
-Static Analysis
The slope stabi-1 ity and analyses were done using the STABL com-
puter program for the general solution of slope stability problems
by a two-dimensional limiting equilibr1um method. The calculation
of the factory of safety against instability of a slope is per-
fo.rmed by an adaptation of the f'lodified oishop method of slices
v1hich allows the analysis of trial failure surfaces other than
those of a circular slope. Soil properties used in the analysis
are given in Table 12.6.
The following conditions \>Jere analyzed~
Condition ~
Construction
Normal Operating
Rapid Or awdovm
Normal Operating with
Maximum Pool
Required
f•1i nimum Factor
of Safety
1.3
1.5
1.0
1.3
Calculated 'factor
rof Safety
U/S Slope DIS Slope
2.2 -2.2
. 2.0
1.8 -2.0
2.0 2.1
1 7 --1,7
1.. 7
The calculated factors of safety as shown in the above table indi-
cate no general slope stability problems under static loadi11g.
Further analysis, using the finite element program for static
analyses of earth and rockfill dams (FEADA~I), determined the ini-
tial stresses in the dam during normal operating conditions. The
program calculates the stresses, strains, and displacements in the
dam simulatin~ the actual sequence of construction operations.
Approp~iate nonlinear and stress-dependent, stress-strain proper-
ties for the soils were taken from information compiled in Table 5
in Duncan et al (1980). Table 12.6 presents the values used in
the-an a lysis. T\vo an a 1 yses were performed to show the effects of
relatively soft vs stiff core materiaL
12-29
... Oyna!llic Ana1ys is
The dynamic analysis was ctone using the QUAO 4 computer ptogr<!m.
The initial values of shear modulus .and damping ra:,io to b~ used
in the analyses were derived from typical values available in
tianerjee et a1 (1979} and are as follows:
ZGI\IE
Core t-1ater i a 1
-Soft
-Stiff
Transition Naterial
Shell filater i a 1
90
120
150
180
Uampi ng Shear
Type Curve
sand
sand
sand
sand
The· design earthquake time hi story wa·s deve 1 oped by Wood 1t/at~d
Clyde Consultants and is shown ·in Figure 12.18. The significant
featur-es are as fo 11 ows :.
-Magnitude 8.5 Richter;
-Location 40 kilometers below site (Benioff Zone);
-Maximum acceleration of 0.55g;
-Uuration of strong motion -45 sec; and
-Significant number of cycles-25.
The preliminary dynamic analysis had peak output values occur-
ring about 24 seconds into the earthquake acce1erat ion time his-
tory. Based on these results, the three iterations for the pro-
posed dynamic analysis were performed using the following sec-
tions of the e~rthquake time history:
Iteration 1:
Iter·ation 2:
Iteration 3:
From 10 to 30 seconds
From 10 to 30 seconds
From 10 to 70 seconds
Conclusions: (Infor-mation to follow on completion of computer
anaysis in mid-February 1982).
(g) Instrumentation
Instrumentation vlill be installed within all parts of the dam to provide
monitoring during construction as 11ell as during operations. Instruments
for measuring interval ver~ical and horizontal displacements, stresses and
strains, and total of fluid pressures) ~s well as surface monuments and
markers \vill be installed. The quantity and location will be decided
during final design~ Typica.l instrumentation is as fo1lo~t/s:
12-30
I
I
u
I
I
I
I
I
I
I
·I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-1
I
-I
I.
I
I
I
I
I
I
I
-Piezometers
Piezometers are used to measure static pressure of fluid in the
pore spaces of soi 1 rockfi 11 and in the rock found at 1 on.
-Internal Vertical Movement Devices
Cross-arm settlement devices as developed by the USBR •
. VariO!JS versions of the taut-\ii re devices have been developed to
measure internal settlement .
. Hydraulic-settlement devices of various kinds.
-Internal Horizontal Movement Devices
. Taut-wire arrangements.
. Gross-arm devices .
. Inclinometers .
. Strain meters.
Othe~ Measuri~g Devices
. Stress meters.
. Surface monuments and alignment markers .
. Seismographic records and seismoscopes .
. Flow meters to record discharge from drainage and pressure
relief system.
12.7 -Relict Channel TY·eatment
(a) Site Conditions
Ear 1 i er stud-ies i dent i fi ed a buried ch anne 1 r·unn 1 ng from the Su si tna ~1 ver
gorge immediately upstream from the proposed damsite to Tsusena Creek~ a
distance of about 1.5 miles. Boring by the Corps of Engineers penetrated
454 feet of glacial deposits overlying bedrock ltlhich v1as encountered at
elevation 1,775 fe-;t, while the surface elevat1on of the lov1est saddle is
approximately 2,005 feet. Additional investigations during the curre~t
study further de 1 i neated the channe 1 and fu 11 detai 1 s are given 1 n the Task
5 Geotechnical Report. The channel represents a potential source of leak-
age from the Watana reservoir. Along the buried channel thalweg, the high-
est bedrock surface is some 450 feet below reservoir level, while along the
shortest leakage path between the reservoir and Tsusena Creek the highest
rock surface is some 250 feet below reservoir level. The maximum hydraulic
gradient along the bured channel from the edge of pool to Tsusena Creek, is
approximately 9 percent, while between existing riverbed levels it de-
creases to about 6 percent. There are surface lakes within the channel
area and while some drill holes encountered artesian water, others pene-
trated highly permeable zones resulting in complete loss of drilling fluid.
Zones of permafrost have also been identified throughout the channel area.
12-31
(b)
Although the glacial history of the area is not clearly understood, a se-
quence of events has been postulated in the Task 5 Geotethnical Report,
based primarily on the investigation of the Borrow Area D adjacent to the
buried channel. The generalized sui"ficial stratigraphic column is given in
Table, 12.5 and the gradation of the soils in the various units are given 1n
Figures 12~55 through 12.18.
Of particular relevance to the buried channel problems are the alluvium at
the base of the channel, encountet~ed in one deep borehole beb·:een 292 feet
and bedrock at 454 feet below ground, and the unconsolidated outwash, allu-
vial and fluvial deposits. The deep alluvium offer·s a potential leakage
path, its high permeability being indicated by loss of drilling fluid,
vlhile the unconsolidated, primatily sandy deposits may be subject to lique-
faction following saturation.
Potential Problems
The major potential problems associated \'lith the buried channel are leak-
age, both surface and subsurface flows; piping at downstream outlets to
Tsusena Creek; the impact o: permafrost and the long~term effects as heat
fr-om the reservoir thaws the ground thl~ough the channel area; and i nstab: 1-
ity of soil slopes on saturation, thawing, or seismic loading leading to a
breach of the rim of the reservoir.
(i) Surface Flows
During the study of alternative layo'Jts for Vfatana) the maximum op-
erating reservoir level was higher than the cr1tical ground e1eva-
t1on of 2005 in the bm~ied channel :J.reci.. Th~~e layouts, therefore,
incorporated a saddle dam about 40 feet high and 2!1500 feet long
across the critical section of the channel. TI1e foundation condi-
tions for such a s add 1 e dam are not we 11 defi n-::d at this time hut
because of the variable nature of the glacial deposits, the effects
of permafrost and potential for liquifaction vlithin the foundation
were addressed. It \'las concluded, however, that in any event there.
was a strong possibility that settlement of such a dam could n.ot be
adequately controlled and there vmuld be a real risk of transvers€
cracking occurring through the dam. With the reservoir level-above
ground sJrface, any such cracking could lead to surface flows and -
subsequent channeling through. the unconsolidated deposits.
{ii) Subsurface leakage
No field permeability tests have been conducted, but it is antici~
pated that the total subsurface leaking wi 11 be relatively small and
economically insignificant. For example, if the average permeabil-
ity of all material in the channel were lo-2 em/sec, the total
1 eak age flow would be' 1 ess than 100 cubic feet per second. By in-
spection of the grading curves, the actual permeability is certainly
less than lo-2 em/sec, except possibly in the channel bed allu-
vium, and a more realistic leakage flow would be about 10 cubic feet
12-32
I-
I
I
I
I
I
·'I
'I
I
I
I
I
I
I
I
I
1·-, . . . .
..
I
-·
I
I
I
I
I
I
I
I
I
I
I
I
I ,.
••
I~ I '. .
per second. The capital value of this leakage is about $4 million.
However, any leakage may be concentrated in the discharge zone in
Tsusena Creek, and there is potential for· piping which could lead to
large-scale.erosion cutting back to the high ground forming the rim
of the reservoir.
( i _i i) Permafrost
Thawing of permafrost wi 11 result in higher seepage rates and pos-
sibly settlement of the surface as excess water drains from the
thawed soils.
(iv) Liquefaction
Filling the reservoir will lead to the saturation of some of the
glacial deposits within the buried channel area, including the upper
slopes of the Susitna River valley., and produce the potential for
liquefaction of these deposits under seismic loading. Under extreme
circumstances, 1 iquefact ion could lead to mass movements of soils
into the reserve ir and breach of the reservoir rim in the area of
the freeboard dike.
For this situation to occur, it would require a large, continuous
deposit of loose, saturated, granular material with sufficient
ground surface slope so that the soil above the liquefied zone wcula
move under its own height. Although such a scenario is consid~red
most unlikely, the investigations to date are not sufficiently de-
tailed to preclude the possibility. In view of the potentially
catastrophic failure that would result from a breach of the r~ser
voir rim, further investigations must be carried out prior to con-
struction to confirm the stratigraphy and provide adequate data to
properly assess the need for and design of remedial treatment. :
.(c) Remedial Measures
Since the stability of the section of the buried chdrmel forming tne rim of
the \4atana reservoir is essential for the feasibility of the Watana ~evel
opment as outlined in this report, practical solutions to all possib1e
scenarios, including extreme combinations of the problems outlined above,
must be identified.
(i) Surface Flows
To eliminate the potential problems associatea 'tJith settlement and
breach of a saddle dam allowing surface flows through the buried
channel area, the maximum operating level of the reservoir has been
lmvered to 2,185 feet leaving a width of at least 1,500 feet of
11 dry11 ground at the saddlE: above this elevation1 A freeboard dike
~ith a crest elevation of 2010 is required to provide protection
against extreme reservoir levels under probable maximum flood condi-
tions. The shor-test distance between the toe of the dike and the
edge of the 2185-elevation reservoir pool is at least 450 feet, ana
12-33
I
and under Pt•JF flood, the \'later 1 eve 1 wi 11 just reach the toe of the
dike. I
(ii) Subsurface Flows
Progressive piping and erosion in the area of discharge into tne
Tsusena Creek \vi 11 be centro 11 ed by the placement of properly graded
granular materials ~TTke ·a f'i Yter blanket over the zones of emer-
gence. Fie1d investigation will be car-ried out to define critical
areas, and on 1y such areas wi 11 be treated. Gont inuous mon i tor·ing
of the out 1 et a) .. ea will be necessary 1 s i11ce it may take many years
for equilibrium vlith respect to permafrost to become estab1 ish~d in
the buried channel area~
If the permeability of the base alluvium is found to be excessive~
grouting of the upstream inlet zone could be carried out to reduca
the total leakage.
(iii} Permafrost
Thawing of permafrost wi 11 occur that may have an impact on subsur-
face flows and ground settlement. No specific remedial work is
necessary; but flows, ground \'later elevation, and groJnd surface
elevation in the buried channel area must be monitored and· any nec-
essary maintenance work carried out to maintain freebcard an<1 con-
trol seepage discharge.
(iv) Liquefaction
To guarantee the integrity of the reser·voir rim through the channel
area requires that either:
There is no potential for a liquefaction slide into the l"'eser·voir
\vhich cou1d cut back and br·each the rim, or
If there is such potential, there is a sufficient volume of stable
material at the critical section that even if the upstrearn ::i:ateri-
als were to slide into the reservoir, the failure zone couln not
cut back to the reservoir rim.
Any remedial treatment required will depend on the location .and e>:.-
tent of critical zones and cou1d range from stabilization by compac-
tion (vibroflotation) or grouting techniques, either cement ~·~ chem-
ical grouting, or in the limit, removal of materia14
The stratigraphic column indicates that the two lower till deposits
I and K have been overconsolidated by glaciation, and it is unlikely
that these deposits could 1 iquefy under any circumstances~ The
overlying Unit H is a medium fine sand ',-~fit~ silt and is probanly the
most susceptible to liquefaction of all the materials sampled .. This
unit has been identified up to 40 feet thick in places, with the top
of the layer ~stirnated to be about 100 feet below ground sur--face at
the deepest point, as shovm on Plate 6.34 of the Task 5 Report. All
12-34
'1·.··.' : ll .. I'
I
I
I
I
I
I
I·
I
I
I
I
'I
I
I
·I
d,
I
I
I
I
I
I
I
I
I
I,
I
I
'I
I
I
·.,I
,I
·•.··•·· ·'
. '
. ·~"-·-.··
materials above this unit, are normally consolidated water/1ain till,
nutwash, alluvium, and fluvial deposits which could include zones of
critical materia 1-s.
. .
There are insufficient data available to identify the fu11 extent of
such critical materials; henc;e,_.Jt __ _i§ -~ot possible to precisely de-
fine the remedial work necessary-·at-lTfis time. Available a1tel'"f1a.-.
tive methods include:
-DensificQ.t ion
Layers within about 100 feet of the surface could be compacted by
vibrof1otation techrdques to eliminate the risk of 1iquefacti(ln . ·
and prov id€ a stable zone.
Stab i 1 i z at i on ,
Critical layers at any depth could be grouted, either with .c.e·:nent
for fine gravels and coarse sands m" by chemical grouting ftt.r fine
sands and s i 1 ts.
-Removal
This could range from the replacement of critical material n~~ar
the valley slopes with high-quality~ pro-cessed material, whi.z:h
would stabilize th~ toe of a potential slide and so prevent the
-initiation of failure that might otherwise cut back and caus~
major failures, to the excavation blending and replacement of
large volumes of material to provide a stable zoneA
The ultimate tr-eatment \'lill be based on an engineering and cost
analysis study of the appropriate alternatives during the design
phase of the project when the site conditions have been more closely
defined. However, to confirm the technical and over.p.ll finaoc~al
feasibility of tr~e project at this time, it is necessary to C.;,lnsider
a solution to tht: worst conditions deemed possible.
On tht~ basis of avail able data, such conditions are:
... That the a 11 uv ium Unit H enc0untered between elevations 2lu:JJ and
2140 in drill hole DR22 is a homogeneous loose, silty~ fine: sand~
That it is of 1 arge areal extent and cant iouous frorn beyond the
'· saddle out to the Susitna valley slopes ..
-That is of such thickness that a fa i 1 ure plane could be contained:
fully vdthin its boundaries,.
With these conditions, liquefaction of the unit under seismic load-
ing following saturation fr0J11 the reservoir could result in the
overlyi.ng materi.aJ sliding on the 1 iquefied zone into the res~l~voil~.
12-3.5
' . -. _,
--j_""-~ • ..:, •• _ ... : •.• :,:..,_ , -~.t "-··-·~~:~'~,...a,t,1'~;,_u . ..:o.·.~~ ~~-·c··-.{'~ ..... ~_:.~:~.~..,, '""~~ _· ... ~ '.
(d)
Catastrophic. failure would d~velop i..f the back scarp of the failure
;. surface through the overlylng materia1s broke ground surface on the
downstream side of the saddle belovJ reservoir water leve1.~
The most positive solution·to such a situation wou1d be the replace-·
ment of the critical zone with material that wou1d not 1iquefy •.
This would involve~ in effect; .the rearrangement of the in-place ,: ..
materials to create an underground dam sect ion founded on the den~e.. ·
till layer beneath the critical alluvium. Such an operation would
invo 1 ve the excavation of a trench up to 135 feet deep with a sur-
face width up to l::OOQ.feet. Selected materials would be compacted
to form a central zone, with 2 horizontal to 1 vertical slopes.
Surplus and unsuitable materials would be placed on both sides of
this central 11 dam 11 to complete backfilling to.ground surface. The
central zone would be desioned to remain stable in the event that
all material upstream did s1 ide into the reservoir~. Preliminary
estimates indicate that such a structure would need to be 5 ~OOO feet
long, with a total cut volume of about 13 million· cubic yards, of
rlhich 4-1/2 mill ion cubic yards could be used in the compacted
center zone~ The cost of such \vork is estimated to be about SlUO
million. The need of such e~penditure is considered to be most
likely and is deemed to be covered by the.overa11 project contin-
gency sum~
Further Investigations
0
Additional site investigations are necessary in the relict channel area to
more closely define the follov.;ing;
Confirm and/or refine the stratigraphy throughout the area.
-Thickness, extent, density, continuity, and permeaoility of the alluvium
identified in DR22 immediate1y above bedrock. The .investigations shou1d
include pumping tests and dye injection tests to check the continuity of
this unit along the buried channel, since this is deemed to have the
·greates~ potential for leakage. · ·
-Density of the lov1est till layers I and J which have been subjected to
overconsolidation by glaciation to confirm that theyv;ould not liquefy
under earthquake loading.
-Density, gradation, extent, and continuity of the sandy silt alluvium,
Unit H.
-Extent of any other units which may be subject to -1 iquefaction.
-Conditions in the outlet area of the relict channel into Tsusena Creek.
-Ground \l!ater regime throughout the channel at'ea v1ith part icu1 ar refer~nce
t~ the source of artesian or confined aquifers and the drainage outlets ·
from such aquifers.
12-36
·1.··.,.-·.1 . 1
. .1
I
>~ .•• ;:'~~
. .
I
I
I \.
I
I
I
:1
I
I
~I
f
J
I
l
( •.. -.. ·
I .·
I
::'1
'I
I t'
I
I
:I
I
I'
'.•1·. . ,
10
I
·I
,I
I
I
I
I
(e) Construction Methods
For the purpos~ of thts feasibility study, it has been assumed that treat~
ment of the relis_t channel.· will involve ti;e following:
... Construction :o( the freeboard'tfike at the c:.~est ·of the saddle involvit1g
stripping of orqanic material and construction of the dike with impervi-
ous c(~e and gravel or rock-fill shells; and
-Stripping, grading, and placement of a filter blanket over the outlet
area of the channe1 in T.susena Cret;k, This b1<H1ket is estimated ta be 7
feet thick and cover an area of 460 acres •.
Allowance for the cost of these items is included in the estimates .. How ...
ever, it is considered likely that the actual area of downstream blanket
required wilT be less than allowed, but that other~emedial work in the
sadd1e area vdll probably be necessary~ ·
The allo~t/ance for the downstream blanket is therefore considered a ~=ason
able estimate to cover all work required in the channel area excl udi;ng the
extreme situation involving major excavation as discussed above.
1,.8-Primary Untlet Facil·ities
The primary function of the outlet facilities will provide capability to ~is
cil argE floods with_ recurrence frequencies of up to 1:50 years after they h.ave
~ ·been routed through the ~~atana reser-voir. Downstream erosion will be mi~ima1
and the dissolved nitrogen content in the discharges will be restricted as much
as possible to avoid harmful effects on the downstr-eam fish population .. J. sec-
ondary function of the facilities will be to enable the rapid drawdown cf the
reservoir by up to 150 feet during an extreme emergency situation .
.The structures will be located on the right abutment, as shown on Plate 4~, and
will consist of an i~take structure~ pressure tunnel, and an energy dissi9~
tion and control structure housing six fixed-cone valves v1hich \"fill dischrarrge
into the river 150 feet below.
{a) Approach Channel and Intake
..
The approach channel to the outlet facilities wi1·1 be shared with the power
intake, The channe 1 will be 400 feet wide and excavated to a depth r;;.f
approx~nately 150 feet in the bedrock with an invert elevatio~ of 2U~7. The
intake structure will be founded deep in the rock at the end of the than-
nel. The single inlet passage will have an invert elevation of 2003 .. It
will be divided upstream by. a central concrete pier which wi 11 suppct·t
~teel trashr•:~.cks located on the face of the structur~, spanning the ~pcn
ings to the water passage. The racks \vi11 screen out suomerged debris
which could damage the downstream valves. Th"e racks will remain in place
with no provision for their withdrawal,. which would be e!(per'sive and is
considered unnecessary because of the infrequent use of the outlet. If
12-37
/)
(b)
(c)
there.ever should be .a blockage) this 'tlOU1d have to be cleared by divers.
Dovmstream of the racks, located between the pier and ·eacn of the side-· ·
walls~ 'VIill be two fixed wheel gates operated by a m~chanica1 hoist mounted
(at the surface) above the deck of the structure. r~e purpose of the fixed
\·Lheel gates will not be to control flows thrcugh the out'iet, but to close .
off the downstream tunnel to allow devfater·ing for maintenance of the tunnel
or gates within the discharge structu1"a~ Stoplog gate:s wjll ba pr·ovided
just upstream of the gates to allow insertion of stoplogs for dewate~ .. ing of
the structure and access to the gate guides for maintenancB.
Intake Gates and Trashracks
The gates will be of the fixed wheel vertical 1 ift type with downstream
skinplate and seals. The norninal gate size \~ill be 16 feet wide b_y· 32 feet
high4 Each gate will be operated by a single drum wire rope hoist mounted
·;n an enclosed tower structure at the top of the intake. The height of the
to,tJer structure \·:li 11 permit raising the gates clear of the intake concrete
·for inspection and maintenance.
The gates wi11 be capable of being lowered either from a remote control
room or locally from the hoist area. \:late: raising will be from the hnist
area only.
The trashracks will have a bar·spacing of about 7 inches, and tttill oe de-
signed for a maximum differential head of about40 feet. The maximurn net
velocity through the racks will be about 6, ft/s. Provision \vi11 be made
for monitoring the head loss across the trashriicks.
Shaft c.:ld Tunue 1
Discharges ~vill be conveyed from the upstream gate structure by a contrete-
1 ined shaft and tunnel terminating in a steel 1 iner find manifold. The man--
ifold wil·l branch into six 78-inch diameter steel-lined tunne1s \<Jhicil \'!ill
run through the main spillway flip bucket structure to the va:ves mo~nted
on the dovmstream face.
The water passages v.;ill be 28 feet in diameter up to the steel manif:gld.
The upstream concrete-1 ined port ion wi 11 run a short distance hori Z)ntally
from the back. of the intake structure befor·e dipping at an angle of s:/' to
a 1ovJer level tunnel of similar cr--oss section. This angle of 55° is con-
sidered the fl~test slope at which the tunnel can be "self-muckingn during
construct ion and is cost effective in the r~esu1tant ratio of shaft to tun-
nel length. The lower tunnel will run at a gradien~ of l~lOtoth.e point
where the overly1ng rock is insufficient to v.fithstand the large hydrostatic
pressure which vlill occur within the tunnel. Downstream of this point the
pressure will be transferred throughout the mass concrete and resisted
jointly by the surrounding rock and the stee1 liner. The steel liner wi11
be 26 feet in d i arneter and surrounded by mass concrete filling the space
between the 1 iner and the surrounding rock, The area of the outs ide face
of the 1 i ner and the <'Oncrete \-Ji 1l be grouted to fin all voids and ~"educe
non-uniform external ground water pressure build up from the ground water.
12-38
•:'.·.·:
. '
I . ._
I
I
-. -:-~
I.
I
I
I
I
, •.
I
••
I
·~
••
I
I
I
I
I
;
•• ,,
I
I
I
I
I
I
I
I
I
I
I
,I
I
I
I
.,
I
Upstream from the discharge structure the 1iner will terminate tn a. steel
manifold vtith six parallel 8 foot diameter steel-lined branches. These
will continue through the back face of the discharge structure. and termi-
nate in discharge valves mounted at the downstream end of the structure.
(d) Discharge Structure
The concrete discharge structure is shown on Plate 52. It will form the
flip bucket for the main spi 11\-vay and wi 11 house the fixed cone valves and
individual upstream ring follov1er guard gates. It wlll be founded on sound
rock high above the river The valves will be set with a centerline eleva-
tjon of 13560 feet and will discharge. into the river approximately 85 feet
below. Openings for the valv~s will be formed in the concrete and the
valves will be recessed within these openings sufficiently to allow enclos-
ure for ease of maintenance and heating of the moveable valve sleeves. An
access gallery upstream from the valves wi 11 r·un the length of the dis ...
charge structure, traversing the steel liners upstream of the valves~ and
wi 11 terminate in the access tunne 1 and access road on e1 ther side of the
structure. Housing ring follovJer guard gat-es will be located upstream from
the gallel"'Y gate chambers. These gates wi 11 operate on the steel 1iner·s
and wi 11 serve to i so 1 ate the discharge valves. A common monorai 1 hoist
wi 11 be located above each valve and gate assembly to provide for their
removal and transportation to the access gallery .. From the access gallery
they can be maneuvered on a low trolley to the service area adjacent tc the
end of the gallery.
(e)-Fixed Cone Discharge Valves
Eight 78-inch diameter fixed cone discharge valves will be installed :at the
downstream end of the outlet ,nanifold, generally as shown on Plate 52 .. The
valves were selected to be within current experience, considering the valve
s 1 ze and operating head (see Figure 12 .20) ~ The fixed cone valves ar~ si m-
i lar to Hm•1ell Bunger valves except that the cone support valves ext~nd
further upstream and are more streamlined. The va 1 ves have a slightly
higher discharge coefficient than Howell Bunger valves and are less prone
to vibration. ·
Electric jacket heaters will be installed around the cyllndrical sleeves -of
the valves which extend outside of the valve room, and since the valves
will be located within a heated valve room, they will be capable of year-
round operation. Normally, when the valves are closed, the upstream ring·
follower gate vdll also be closed, so that freezing of leakage at the valve
seat should not be a serious concern.
The va 1 ves will be operated either by two hydraulic cy11 nder operators or
by a sere\~ stem hoist. For preliminary design purposes, hydraulic opera-
tors have been assumed. The valves may be operated either locally or re-
motely.
At the time of valve design 3 careful consideration must be given to prevent
vibration. Considerable research will be carried out concerning experience
12-39
r ~. ~
and design of existing installattons, and model tests will be necessat'Y to
help ensure satisfactory VrilVe operatton. Adequate design of the fix~d
vanes wi11 be of prl,me importance.
In si.z1ng the valves for the· pre1 imtnary design, it has been assumed that
·the valve gate_opening will be restrh:ted to 80 percent full stroke because
of v1 brat ion conslderati mrs.
Careful considerat1on must also be given to machining and surface finish of
the valves, in order to prevent ~avitation and erosion of the valve seals
resulting from th.e extremely high water velocities.
· (f) ·{Ring Fo 1lovJer Gates
A ring follower gate will be installed upstream of each valve a~d will be
used:
-To permit inspection and maintenance of the fixed cone valves;
To re 1 i eve the hydrostatic pressu·re from the valves ~tJhen they are in the
closed position; and~ , ~
-To c 1 ose against flowing water in the event of ma 1 funct 1 on or fa 11tn'e of
the yalves.
The ring follower gates will have a nominal diameter of 90 inches and \~i1l ..
be of welded or cast steel construction. The gates will be designed to
withstand a total static head of about 630 feet. Existing large diameter
high head ring follower gdtes are summarized in Table 12.:6.
~The ring follower gates wi 11 be designed to be lowered under flowing Nater
conditions with maximum head, but they will on1y be raised under balanced
head conditions. Valved bypass piping will be used to equalize the pres"
sure on both sides o.f the ring follower gate before raising. The gates
will be operated by hydr·aulic cylinders with a nom·inal operating presstrre
of 2000 psi . Either loca 1 or remote operation of the ring fo 11 ow2r gates
will be possible.
A grease system will be installed in each gate for injection of grease be·
tw€en the gate 1 eat .and the gate. body seats to reduce fr1 ct 1 on a 1 forces
when the gotes are operated.
(g) 01 sc;h arg e P..rea
Immed1 at ely downstream of the discharge struc;ture, the rock wi 11 be cut at
a slope of 2H:3V to a lower elevation of 1,510 feet. This face will be
heavily reinforced by rock bolts and protected by a concrete slab anchored
to the face. The lower level will consist of unlined rock extending to the
river. Much of the lower trajectory of the discharge jets \·li 11 impinge on
this area in the form of a very heavy spray, while much of the upper part
of the jets will carry as far as the river. Because of the high degree of
dispersion of the discharges and the infrequency of operation of the
valves, it 1s not anticipated that erosion will be a problem requiring
other than occasional minor maintenance.
12-40
I
I
I
I
I
I
I
I
I
I
I .-.
. \
I
I
I
I
I
·I
I,.
It
[I
I \'•
J(
~~.{···!· -.· ~ ---\;
I
I
I
I
I
I
I
I
I
I
I
I
I
l
I
12.9 --Main-Sei11wax
{a) General
The main spillway 'vill provide discharge capabi1 ity far flows exceeding the
capacity of the out.1et facilities, passing the excess of floods with a
corresponding return :period of less than 1~50 years. The comtdned total
capacity -Of the main spilh1ay anJ outlet facilities ,,;111 be sufficient to
pass routed floods with recurrence frequencies of up to 1 in 10,000 years.
The spillway, shown on Plate 49, is located on the right abutment and con-
sists of an approach channel, a gated agee control structure, a concrete~
lined chute, and a flip bucket.
(b) Approach Channel and Control Structure
(c)
The approach channel is excavated to a depth of ctoproximately 100 feet ·into
rock. It is adjacent to the p0\1er facilities approach channel, and in
order to minimize its length, it partly offtakes from this channel, inter--
secting it at elevation 2125 feet,
The control structure is a massive concrete structure set at the end of the
approach ch anne 1 . Flows are contro 11 ed by three 42 feet· high by 36 feet
wide vertical lift gates. As shown on Plate 50, each gate is contai:n~d
~tlithin a separate unjt consisting of an ogee overflow weir, piers~ a~d an
integral roadway deck~ The units are of amonol ithic box type structul'"e
with split pier construction. The box configuration will give rigidity
during seismic shaking, and the spl1t piers will allow for. some rala~ive
motion and stress relief during earthquake motion in order to minimi;;:B the
possibility of the gates jamming in the closed position.
Model tests will be 1ecessary during the f,inal design stage to determine
final geometry and dimensions such as pier noses, crest shape, and p3er
length.
The structure wi 11· be located adjacent to the right dam abutment in ~ ine
w.ith the dam crest. The main access route from the Denali Highway 'n'i1l
pass across the spillway deck and along the crest.
The ma.in dam grout curtain and drainage system \'/ill pass beneath the struc-
ture~ Access to the grouting tunnels will be via a shaft within the struc-
ture and a g a 11 er y running through the ogee \\'e i r.
Spillway Gates and ·stoploas ...:.:...;_
The three spillway gates will be of the fixed wheel vertical 1 ift type op-
erated by double drum wire rope hoists located in an enclosed bridge struc-
ture.. The gate size has been se 1 ected as 36 feet wide by 40 feet high, \n·
eluding a 3 foot freeboard allowance above maximum normal reservoir level~
The gates will have upstream skinplates and seals ~vi11 be totally encJosed
to permit gate heating in the event that winter operation i.s necessat-y~
Provision \vi11 also be w;.Je for heating the gate guides~
12-41
,_
,_.
(d)
(e)
The height of the to\'ler and br·idge structure will be selected to ·permit
raisi'ng of the gates above the top of the s.pillwny pier fm~ gate. inspection
.and maintenance.
An emergency ga?o 1 i ne engine 'lf.Ti '11 be provided to enab 1 e the gates to be
r·aised in the ~vent of loss· of power-to the spi1hJay gate hoist motors.
A set of stoplog guid_e;:; will be installed upstream of each of the three
spillway gates. One set. of stoplogs wi11 be provided to permit raising the
gates with maximum reservoir levels vlithout discharging water over the
spi 11way.
The stoplogs \'iil1 have dov-mstream skinpl ate and seals and will be sr'f'anged
in sections suitable for handling by a mobi1e crane with a follower.
Spillway Chute
The control structure will discharge dovm an inclined chute that tapt:!rs
slightly to a width of 80 feet, which is maintained over the remainder of
its length ..
The chute section will be rectangular in cross section~ ex~avated in rock.~
and lined with concrete which will be .anchored to the rock. An under-
drainage system will be constructed beneath the slabs to relieve uplift
pressures and will be formed by drilling holes from a centrally located
gallery, in the rock, to intersect box drains located oeneath the chute
floor slab. Provision will be made at two locations dovm the chute to
aerate flmvs and prevent cavitation of the concrete floor. Aeration vti1l
be attained by means cf anincline.d step into the flow. Air wi11 be drawn
of from a transverse 1o\ver gallery via ducts \'1hich exit at the downstream
vert ica1 face of the step .. Convergence of the chute r1alls at the upper end
of the spillway will be gradual to minimize any shock ·aave.deve1opme!lt.
Nadel tests will determine the maximum allowab1e convergence _of the 1~a11 s
to assure both economy atld sat i.sfactor·y hydrau1 ic performance~ ivloue1 tests
will also determine aeration requirements (number and size of aeration
ducts) a~d ~he shape of the chute floor.
Flip Bucket ·,
The function of the flip bucket will be to direct spillway flows clear of
the spillway and \''e1l downstream into the river belov1. The jet issuing
from the bucket will be partly dispersed during its passage through the air
with a corresponding loss of energy. The remainder of the energy \v111 be
dissipated on impact with the plunge pool~
The mass concrete block, as described in Section 12.8, will house the gates
and discharge valves for the out1et facilities, and a1so will form the f1 ip
bucket for the main spillway .. The elevation of the structure will be con-
trolled by the elevation and head limitations of the dischat·ge valves.
Final geometry of the bucket will be determined by model studies, as well
as dynamic pressures on the floor and v~all s of the structure. Although the
structure shown on Plate 52 shows a simple) cylindrical type of bucket, it
0
12-42
I"
I
'I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
·I
I
I
I
I
I
I
I
I
I
I
I
-·,
I
I
I
I
II
I
I
I ;..-'
~' 'f; J
it is foreseen that; a more effective, dispersive type bucket ~tlill be deve1-
oped.during model tests.
_Such. a bucket was Clev-e1or:~:d for_ the Portage f'~ountain Project in or it ish
Columbia, 'tlhen.:· flows u-p to cfs under a .str1tic head of feet af'e
ct'~scharged at a 90~ angle into the river below., In .order to prevent ero-
s ii:.tn of the oppos 1te river bank, the jet is spread by means of a disk-·
shaped bucket so that the area of impingement in the river is spread in a
narrow 1ine parallel to the bank and normal to the chute centerl ire.
12.10 -Emergency Spill way
The emergency spillway wi11 be located on the right side of the river beyond tr1~
main spillway and the pm'ler intake structure (see Platg 53}. The spilh-1ay \·d11
consist of a long straight chute c·ut in the rock and leading in the direction of
Tsusena Creek. An erodible fuse plug, consisting of irnper·tious and fintl gravel
materials, wi11 be constructed at the upstream end; it will be designed to wash
away \•then overtopped by the reservoir" thus releasing floods of up to 160,0UO ·
cfs in excess of the combined main spillway and outlet capacities and preventing
overtopping of the main dam.
-
(a) fuse P1uq and t\pproach Channel
The appr·oach channel to the fuse plug will be excavated in rock and wil1
have a width of 310 feet and invert elevation of 2170.; The main accas-s·--
road to the dam and power-house wi11 cross the channe1 by means of a b}~iage~
The fuse plug will c 1 ose the approach channel) and \'li 11 have a maximum
height of 31.5 feet with a crest elevation of 2201.5 feet. The plug \xi11
have an impervious core up to l.O feet \vide, steeply inclined in an upstream
direction" with fine filter zones upstream and downstream. It \vi1l be sup-
ported on a downstream erodib1e shell of crushed stone or gravel up to 1,5
inches thick~ The crest of the plug will be 10 feet JtJide and ~vill be tt"a-
versed by a 1.5 foot deep pilot channel. The zoning of the plug will te
similar to that designed for the New Melones Lake project in California~
For this project a half scale version of the plug 'tlas constructed and
tested to destruction. The.p1ug failed completely in 1essthan an h.:our and
the water level rose no more than 6 inches above the invert of the p11ot
channel~ The principle of the plug is based on erosion progressing r-apidly
dmvnward and 1 aterally from the pi1 ot channel as socn as'' it is overtopred.
The channel sect 1on at the fuse plug is cons 1dered as a broad crested \•;eir
with a coefficient of dischar'ges of 2.7. A gated control structure \'ias
considered as an a1tetnative to the fuse plug, but this would give higher
construction and maintenance costs and 't~ou1d not provide the automatic
breakirl;g and discharge of the plug.
(b) Discharae Channel
The rock chanoel dm·mstr--earn of the fuse plug will narrow to 200 feet and
continue in straight line over a distance of 5,000 feet at gradients of 1.5
percent and 5 percent in· the direction of Tsusena Creek. The channe1 !.~ill
12-43
' .· --~·. ·~ ,.
': •• "~' ,, •• -J,• -,' '.: • --·,_
7
,:<>' ''-:::~,
·,:__.
"\ :..·,
d isc.harge in a buried Valley 110tl the ,pownstream Side of the main re1 ict
channel from \'lhich flmvs will r·un down into the 'creek. It is estimated
that flows down the channel wi 11 cent inue. for a period of days unaer
probable m~imum flood conditiohs" Some erosion in the channel wfll be
'acceptab1_e,fbut the integrity of the main dam will be maintained. The
reservoir-will be dra\··m down to Elevation 2170) and reconstruction of the
fuse plug wi1.1 be required prior to refilling of the reservoir.
12~11 -Intake
(a) Genera 1
,I)
The intake structure at Watana will be required to fulfill the following
functions:
To pt~ovide independent power fl0\•1 to each of six Francis turbines, up to
a maximum flow of 3,800 cfs per unit, for any reservoir level from EL
2000 (maximum extreme flood level J to the minimum dr·awdown 1eve1 {Selec-
tion of drawdown level is reviewed below); ...
To provide an upstream control gate on each penstock to perrnit de'natet'ing
of the pen&tock and turbine water passages for routine inspection and
maintenance; and
To contro 1 the temperature of water d i sch ar·ged ftom the reserve ir tt•d thin
acceptable 1 imits to mitigate the environmental impacts of the Susitna
development on dmvnstrearn fisheries and vegetation.
(b)· Environmental Constraints
The seasonal operation of the Watana reservoir will have two major tapacts
on downstream flows:
-In summer, the temperature of dm•mstream releases wi 11 be cooler than the
normal river regime; and
In winter, the temperatures \-J11 1 be warmer than the normal r·iver r·e:gime.
Cooler water in the summer months could have a significant impact on down-
stream fisheries, particularly in July and August when sa1mon are mnvit'lg
into the sloughs downstrea11 from Devil Canyon to spa't'm~ ~~armer water in
winter will-affect the formation of ice, resulting in extensive opcln 1.111ter
downstream from the._reservoirs.
Temperature simulation using a Corps of Engineers Hydraulic Engineering
Center (HEC) progra.'TI was used to model the dO!tmstream effects of reser·voir
operation using a variety of differ·ent pov1er intake designs at watana ano
Devil Canyon .. These studies have indicated that dovmstr·ea1l temper·atures
wi 11 remain constant in winter at 39°-F reg~rd1ess of the type of po'.';er in-,
take design used. Hm·1ever,. the do'lmstream river temperatures in the summer
months can be significantly improv€d by power intake design at ',"/atana which
12-44
l ···::t
l ~ ' ..
I
I
I'
I
I
I
I
,~
I
I
I
••
I
I
I
I
I
;I
~~· r::,
e
I
I
I
I
I
I
I
I
I
••
I
I
'I
I
I
I
I
I
(c)
{d)
months can be significantly improved by power intake design at Watana which·
wou 1d ·permit water to be dr a\·tn ·off to the reservoir surface at an times .
. The power intake design at Oevi1 Canyon ·will be 1ess·significant because
·the maximum drawdm·m is only 50 feet.
The se 1 ected power intake design at Wat an-a \·li 11 perm1 t water to be dt"awn
from the reservoir at four distinct levels thr·ough the antfcipa:ted range of
dra\'Jdown to mitigate the environmenta1 impacts on do;.mstream river t~mpera
tures {see Volume 2}.
Details of the reservoir temperature modeling are presentad in Appendix 84.
Orawdown
The maximum drawdown .at the power intake contra 1 s the 1 ive storage of the
reservoir (the volume available for seasonal and over year regulation of
the r'iver flows). With no drawdown capability, trH~ dependable (firm}
energy from Hatana would be controlled by the oependable flow of recor·d to
the required reliability criterion; this has been defined as the lowest
fl mv in the second driest year' of record, \vi th a recurrence of about. 1 in
70 yeat·s.
As the drawdmvn is increased> the firm flow from the reservoir \'li 11 in-
crease. The firm energy \vi11 also increase fOt"' dra~t!do~·m up to about 140
feet. If drav<~down is increased beyond 140 feet, hO':tever, the firm f]cw
will increase but the firm ener~gy will start to reduce, since depend.able
energy is governed by a combination of flow and head.
Average energy avail able from the reservoir shows a very slight decrease
\·lith increasing drawdown because of the imposed constraint in the computer
simulation that the reservoir should be fu1T at the end of the 32 years of
recorded flows. Hm~Jever, the usab ~e energy \'li 11 increase with dratid~wn be-
cause the storage available l'lil1 permit excess energy in summer to be
stored for use in the winter.
Costs of the intake structure and approach channel will increase sig~ifi
cant1y with the depth of-dra\~down, while the cost of penstocks wi11 te re-
duced. A detailed study has shown that the extra benefit of increasad firm
energy from ~~latana will be in excess of the required incremental cost of
the intake, approach channel, and penstocks for any drav;~owo up to l..:i,J
feet .. Therefore, the maximGm dravldo\·m selected fof' preliminary design was
140 feet.
Desi~n
The power intake wi 11 be a free-standing concrete structure located in rock
excavation at the upstream end of the approach channel. Access to the ·
structure ~<~ill be the same as access to the intake for the outlet \IJQ~"'ks,
since the two structures have a common approach channel. ·
In order to draw ~'/ate:r close to the reservoir surface over a ctrawdmvn range
of 140 feet, four openings wj11 be provided in the upstrearn concrete wall
12-45
(e)
i.
I
of the structure for each of the six independent power intakes.. The uppet·
opening wi11 always be open~ but the lower three openings can be closed off
by s1iding stee1 shutters operated in a ccommon guide. ~All openings wi-11 be
protBcted by upstream trashracks~ ·A heated ice bulkhead will be provided
. for winter operation ..
An intake contra 1 gate will be pr·ovided on each penstock.' A s iog1e up..-
stream bulkhead gate will be provided for· routine maintenance on the six
intake cont~--ol gates~ I~ an emergency., stoplogs can be insta11ed in the
upstY'eam wall of the power .intake for work on the trashracks or shuttet
guides.
The vJidtb of the intake wilLoe controlled by the minimum spacing of pen-
stock tunne1 excavations, taken as 2.5 times the excavation diameter.
The upper level of the concrete structure will be set at EL 2200, corres-
ponding to the maximum anticipated flood level. The mi11fimum structure
.. level will be governed by the physical dimension of the penstock tunnel ex-
cavation and the vortex criterion for flo~l 1-nto the p-enstock from the mini-
mum reservoir 1 eve 1 EL 2045. The found at ion of the str-ucture will be an out
150 feet below existing ground level and will be expected to be generally
in sound rock.
Mehanica1 equipment will be housed·in a steel-frame building on the upper
1 eve1 of the concrete structure~ The general arrangement of the pC\w~:r in-
take is sho\vn on Plate ---
Approach Channel
The \'lid th of the approach ch anne 1 wi11 be governed by th-e combined \'tiuth of
. the power intake and the intdke to the outlet fac·n ities, which in t~trn
will be governed by the minimum spacing of the penstock and outlet turme1s,
The overall width of the channel wi11 be about 350 feet. Excavated s1.:lpes
in sound rock will be generally 1H:4V.
The maximum flow in the intake approach channel \'li11 occur-when six
machines are operating and the uutlet facilities are discharging at reaximum
design capacity, with ~he reservoir· drawdo\·ffl to EL 2045. Under these con:-
ditions the maximum ve1ocit_l/ of flow in the approach channel wil1 be
3.5 ft/s, which wil1 not cause any erosion problems~
Higher flm•Js will be possible where the intake approach channel intersects
the ·approach channel to the main spi 1lway and the approach channel. to the
emergency spilhvay. The maxi_mum velocity of f1ow 't;ill be about 25 ft/s'!
and excayated slopes in this vicinity may need increased support at local-
ized are-as of sheared or fractured rock.
Excavations in overburden wi11 generally be trimmed at 2H;1V; rip-rap pro-
tection v'li11 be required in the areas \~there high-flO\'/ velocities ar·e anti-
cipated_
12--46
I
I
;I'_· __ ·_
~. . --· I
j
" .' ~ ' 1··_.
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I.
·I
I
I
I
. ·I
I
I
I
I
I
I
(f)
(g)
Geotechnical ConsidEwations
The excavationw.f11 be over 200 feet deep in ruck in the northwest corner-,
with a total~ excavation d~pth of 240 feet. The southern end of the struc-
. ture will be located in tne ao feet vJi th shear and fracture zone. The ex~
cavat ion depth at the north end of the str"uctura wil1 be 120 feet.
\Hth sufficient rock support, main1y fr·om rock bolting, tne rock -:lopes can
be cut nearly vertical, \1/ith the pussible except.ion of the southern end.,
v.lhere the excavation wil1 "intersect the fractur~ and snear zor1e. If it
proves impracticable to support this face nearly vertically~ it will have
to be trimmed back to a stable slope. The int-ake structure wuulcl therl bd
pa\"tially free-standing .. The spillway tunnel portal will also be 1ocat-ed
in this zone of fractured rock and will r·equire substantial rock support
installed in the portal face. Since the intake structure will~ when ccm-
plete, support this rock face, the required support will be temporary.
The found at ion wi 11 be in sound rock~ but the shear and fracture zones. at
the southern end may require consolidation grouting. t11inor shears and .
fractures exposed in the remainder of the found at ion area may require "!oca 1
grouting and/or dental concrete.
Mechanical Arrangement
( i) Ice Bulkhead
A heated ice bulkhead will be installed in guides immediately up-
stream of the trashracks for each of the six pO\•ter intakes. Tb: ic~
bulkhead will be operated by a movable hoist with a double point
1 ift and will be automatically raised and lm1ered so that it w111
always be at r·eservoir level. The ice bulkhead wi11 serve to~
-Ninimize ice accumulation in the trashrack and intake shutte·~
area; and
Prevent thermal ice-loading on the trashracks ~
The bulkhead will incotporate fixed \vheels which will serve t,::
transmit thermal ice loads to the embedded guides. The bu1knead
\'lill be totally enclosed and heated~ The power supply cable t:: the
bulkhead will be located on an electric, motor-operated, tak~-~~P
reel \vhich will operate in conjunction with the bulkhead hoist, The
bulkhead ~vill be handled by a movable hoist to facilitate re~r:\.~',ial of
the bulkhead for trashrack cleaning.
In the unlikely event that it is desired to dewater the compl~t2
intake for inspection of the shutter guides, stoplogs can be pJ~~
chase;d and installed in the ice bulkhead guides~ The height uf the
stop 1 ogs will depend on the reserve ir 1 eve1 at the time of inspec-
tion.
12-47
·I
{if) Trashracks
Each of the six pmver intakes 'Nil1 hav:e four sets of trashracks, one
set ·;n front of four intake openings.. Each set of trashracks wi11
be in two sections to f ac n i tate hand 1 i ng by the intake service
crane with a follower. Each set of trashtacks wil1 cover an opening
30 feet wide by 24 feet high.. The trashr acks wi 11 have a oar spac-
,..ing of about ti inches and \'fill be designed for a maximum differe-n ....
tial head of about 25 feet.
(iii) Intake Shutters
Each of the six pm,ter intakes will have three intake shutters h'hich
will serve to prevent flow through the intake openings behind which
the shutters will be installed. As the reservoir level drops) the
sliding shutters will be removed as necessary using the intake ser--
vice crane,
Each of the shutters will be designed for a differential head of
about 15 feet~ Suitable alarms and generating unit shutdown devices
will be installed to activate when the differential head is atl01Jt
1/2 th·<: value. In addition, at least one of the shutters at each
power intake will incorporat~ a flap gate which, with 15 feet dif-
ferential head across the shutter, Nill a11ow maximum turbine flmv
through the flap. This will prevent fdilure of the shutters in the
event of accidental blocking of a11 intake openings be1ow wat~Y"
leveL
The shutter guides will be heated to facilitate removal in s~b
freezing weather. In addition, a nubb1er system will be pr·ovided in
the intake behind the shutters to keep its water surface free of ice
in this area. The bubb1er nozzles will be located at several 1eve1s
to per1nit bubbler system operation as the reservoir is drawnd~wn.
tiv) Intake Service Crane
A single, overhead, travel ing-br·idge type intake service crane_ t~i11
be pr·ovided in the intake ser-vice buildings. The crane will be
used:
For serv1c1ng the ice bulkhead and ice bulknead hoist;
-For handling and cleaning the trashracks;
-For handling the 'Hater intake shutter-s;
-For handling the intake bulkhead gates; and
-For servicing the intake gate and hoist.
The overhead crane will have a double point lift and will have fol-
lowers for handling the trashrack shutters ana bulkhead gates. The
crane will be radio-contr-olled with a pendant or cab control for
backup. A grappling hoist rlill be provided for cleaning debris from
the rack area*
12-48
I
I
I
.;
I
I
I
' .•. . .
I
I
I
I
I
I
I
I
:1
I
I
I
I
••
I
I
I
I
I
I
I .,.
I
I
I
I
I
••
I
I
I
. {v) Intake Hulkhead ti~tes
One set of intake bulkheads, consisting of two gate SciCtions, will
be provided for closing any one of th{~ six inta~<e openin9s upstream
from the intake gates~ Tt11~ gat~s ,,,; 11 be used to permit inspect ion
·and maintenance of the intake gate and intake gate guides. The
gates will be raised and lowered under balanced ¥·Jatet conditions
only. To balance water pressute ·in order to raise the intake bulk-
head gate, the space between the gate and the dovmstream contro 1
gate will be flooded by a follower-operated bypass valve oti the top
gate section; an air valve will be provided in the top of the ~ate.
The gates wi 11 have a downstream sk inp 1 ate and wi11 seal on the
dovmstream sid-e. The gate \vil1 be oes igned to withstand full dif-
ferential pressure.
(vi) Intake Gates
The six intake gates!! one for each intake, will e3ch be provided to
close a clear opening 17.3 feet ~tlide by 16.7 feet high. It is pro-
posed that the gates \vi ll be of the vet·t ical fixed wheel 1 ift type
with an upstream skinp1ate, and they will seal on the upstrea~:1 side.
Each gate \~ill be operated by a hydraulic cylinder type hoist~
mounted below the 2,200-foot deck level. The length of a cyli~d~r
will allow withdrawal of the gate from the water f1m·t. The c): illder
and gate will be connected by a series of hooked links. The i::take
ser'vice vli11 be used to raise the gate, cmnplete ;·lith links aP~ cyl-
inder, by dogging and removing each link in turn until the gate is
raised above deck lev~l for maintenance. The gates wil 1 normally be
closed under balanced flow conditions to permit dewatering of the
penstock and turbine water passages for inspection and maintenance
of the turbines~ The gates will also be designed to close in an
emergerfcy with full turbine flov1 conditions in the event of 1oss of
control of the turbine.
The hoist pumping unit will have an operating pressure of l,OCD to
2,000 psi.. The hoist will be designed to allow gate closure ",;·;ithout
any ac power supply.
A heated air vent will be provided at the intake deck to satisfy· air
demand requirements when the gate is closed with f1owing water-..
12.12 -Penstocks
The general ar·rangement of the penstocks is shown on Plates 54 and 55.
The maximum design static head on each penstock is from normal full reservoir"
elevation (EL 2185) ·to centerline distributor level {El 1422). An allowance of
35 percent has been tnaae for pressure rise in the penstock besause of hydrau) ic
transients. The maximum design head is therefore 1,030 feet. Maximum ext~sme
head (corresponding to maximum reservoir flood level) is 1,050 feet.
12-49
:;.::.· .. ·-.
(a) Steel liner
It has been assumed that the r·ock. adjacent to the powerhouse cav.ern wi 11 b~!
incapab\~J:~J. long""'term restr·aint against the forces transmitted from pen-
stock t;:Ydt"'ati'tic pressures, Consequently, a steei liner will be requ·ired
which ·wi11 wholly r'esist the maxi-mum design head, without suppor"t frc~:n the
surround rock.. This section of steel liner wf11 extend 50 f~et"frcm t~1e
powerhouse. Beyond this section the steel liner will be extended a further
lbG feet, and allowance in the design wHl be madf.! for partia1 rock StiPi)Or't
to 'mitigate the maximum design stress. For pre11minary design purposes it
is assumed that not more than 50 percent of the maximum. design _head w111 be
taken by the rock support over this transition length,.
The steel 1 iner \<till be surrounded by a concrete infi11, \viti1 a minimum
thickness of 24 inches. The optimum internal diameter of the steel lining
will be 15 feet based on the minimum total cost of construction and the
capitalized value of anticipated annual energy losses~ A tapering transi ...
tion will be provided to increase the internal diameter of the steel liner
to 17 feet at the.junction bet\•Ieen the stee1 linsr and the concrete 1inet".
(b) Concrete Lining
The penstocks wn 1 be fully 1 ined with concrete from the intake to the
steel lined section:1 the thickness of lining varying with the design head.
The optimum internal diameter of the concrete l_ined penstock will be 17
feet, based on the minimum total cost of constr-uction and the capitalized
value of anticipated annual energy losses, and assuming an av0rage coocrt.:te
lining thickness of 18 inches. The minimum linin_g thickness \1i11 be 12
,. f . -1ncnes.
(c) Grouting and Pressure Relief
A comprehensive pressure relief system wi11 be required to protect t~~
undergt~ound caverns against seepage from the high pressure pens tcck ~ The
system will comprise sma11 diameter bm~eholes set out in patterns anc t:ur-
tains to intercept the jointing in the rock.
Grouting around the penstocks will be provided to:
-Seal and fi11 any voids bet\11een tbe concrete lining and the steel
~-~h-ich may be left after the concre:-J:e placing and curing; and-
-Fi 11 joints or f~~actures in the rock surrounding the penstocks to reauce
flow into the pressur·e relief system and to consolidate the rock ..
12.13 -Powerhouse
( a), Gen·er al
The underground_ pm·Jerhouse complex ~tli 11 be .constructed underground in the
-right 7 abutment.. This will require the excavation in rock of three :najor
12-50 . . ·::-
I
---. --
l
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
·~
I
I
I
I
caverns, the powerhouse, transformer gallery~ and surge chambers 1.v~th Hl-
terconnecting rock tunnels for the draft tubes and iiolated phase bus
ducts.
Unlined rock tunnels will be reauired for vehicular access to the three
main rock caverns and the penst~ck constructi~n adit. Vertical shafts wil1
be required for personne 1 access to the undergrounti po·:;erhou';~) fer cable
ducts from the transformer gallery, for surge chamber 1enting and f0~ the
heating and vent i 1 at ion sys tern.
The gener-al 1 ayuut of the povverhouse comp 1 ex is shr;wn in p 1 an and se•:t ~on
in Plates 54 and 55~ and in isometric projection in Plate 5b. The t?s~s
former gallery will oe located on the upstream s~d~ of the po·dm·huus-J
cavern; the surge chamber 1Nill be located on the do'tmstream side. S~~a~~
dimensions bt=twaen major roc~ excavations have been :set at 1.5 t hm:.:;;:. :~~:.~
main span of the larger excavation. This criterion contrc,ls not O~i:} tnt:
minimum distance bet\veen caverns, but also the spacing between tr·an~f"~"\Ht:J'
gallery and penstoc~, beb·teen Lus shaft and penstoc<) and the min~mt,::~ s~;dc
ing of penstock and tailrace tunnels.
The draft tube gate gallery and crane will oe located in the surga .. ~ .. ~.;;:~;~e~~
cavern, above the maximum anticipated surge level. Provision wil1 a~s: o~
made in the surge chamber for ta i 1 r cce: tunne 1 intake stop 1 ogs, \vh h::h ,..;~ 11
be handled by the draft tube crane.
(b) Layout Consideration
The location of the powerhouse \'/as selected from consideration of th~~
following data:
Plots of the known major faults and shear zones on the·right abut~~~:;
-Estimated cost of approach ch.:mne1 excavation) intaKe str·ucture, ~'e'i
stocks, .and tailrace; and
An ass!lmed argle of 55Q to the horizontal for the inc1 ined section ,)f'
penstock~
Prel iminal~y cost estimates indica~e U1at the intake ::tructun? and .a~-~~ .... -ac~:
channel excavation are the most sionificant items in the overall ay'T~nce-
~ ~ ~ent of the power facilities; the underground powerhause costs are a~;~n-
dent only on installed capacity. The optimum atTa'hgement has the~'t.:f,.,,~~t:
been determineo by adjusting the position of the i~take to giv~ the :~ast
cost for intake, penstocks, and tailrace~ Since the costs of tunne: ~-,~ ar·e-
small compared to the intake costs, the intake will ne sited as far ~J
stream as possible) consistent \'Jith the required minhnum draN:.im·m 1tE;•,21 ..
and a reasonable 1 ength of access tunne 1 s ~
The underground transformer gallery will be located on the upstremn s~~e of
the powerhouse.. This arrar1gement gives the minimum possible d1stanc-::
bet\<Jeen the turn ines and-the surge ch ar.1ber, for mt1X imum protect icn cf the
12-51
(c)
(d)
draft tubes under transient load conditions .. The trans· .. ormer gallery and-
the powerhouse cavern vii 11 be pr·otected against high pr ;ssure seBpage from
the penstocks by a .200 foot long stee1-1ined section arid an extensive pres-
sure relief system (see Sect ion 12~12) 4
Access Tunnel~ and Shafts
Vehicular access to the underground facif)~·+:ies at Hatana will be provided
by a single unlined rock tunnel f}*Om tilt: right bank area adjacent to tht?
diversion tunnel portal. The access tunnel will cross over the diversion
tunnels and then aescend at a uniform gr·adient to the south end of the
powerhouse cavern at generator floor level~ at El 1463. Separate branch
tunnels from the main tunnel will give access to the trilnsfot?mer gallery at
EL 1507, the penstock construct 1 on ad it at EL 1420, and the draft tube gate
gallery at EL 1500.. The maximum gradients will be o.l percent on the con ..
struction access tunnel, and 6.9 percent on the permam.mt access tunne1s,
The common access tunnel will be size.d to provide passing c1earanc2 fer the
construction-plant used during penstock construction. Tne size of articu~
1ated trailer required to de1 iver heavy items of machinery such as the tur'-
bine runner, turbine spiral case, and generatot rotor, will be less crit1-
cal with respect to tunnel size, but will dictate the minimum radius of
vertical and horizonta1 curves. Fat~ preliminary design, tr1·e cross section
of the access tunnel has a modified horseshoe shape, 35 feet wide by 28
feet high. The access tunn.e 1 branch to the surge chamber and draft tttbB
gallery will have a reduced section, consistent with the anticipated size
of vehicle and lo-ading tequir·ed.
The main access shaft ~vi 11 be at the north end of the poworhouse c ~ivern ~
providing personnel access from the surface control builtJing by elevator.
Access tunnels will be provided from this shaft for pedestrian access to
the transformer gallery and the draft tube gate gallery. Elevator access
wi11 also be provided to the fir8 protection head tank, located about 250
feet above powerhouse 1 eve l.
Powerhouse Cavern
The main pO't'.fer·house cavetn 15 designed to accormn0date si:< vertical shaft
Francis turbines~ in line, with direct coupling to coverhung generators.
Each unit is designed to generate 170NW at a rated head of 680 feet~
The vertical dimension of the pow·erhouse cavern is determined by the physi-
cal size of turbine and generator, the crane height reqairea for routine
maintenance, ano the design dimensions of the turbine draft tube. The
length of the cavern \vill allow for a unit spacing of tiD feet, tttith a 110-
foot long service oay at the south end for routine ~aaint.:::nance and for con-
struction erection. The width of thecavern allovts for the physical size
of the generator plus galleries for piping and air-conditioning~ electrical
cables, isolated phase bus ducts) and generator circuit oreakers. Contin-
uous d~ainage galleries will be provided to a low level SJcp.
12-52
I
I
1
..
'--.:::
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
·~
I
·-
I
I
I
••
I
I,
I
I
I .,
I
I
1-
Vehicular access will be by tunnel to the generator floor at the south end
of the cavern; pedestrian access win be by e1 eva tor frone the s_urf ace con-
trol building to the north end of the cavern. Multip1~ ~tairway access
pqints '.vi11 be availaole from the mairr·f1oor to e:tch gai18ry level. Access
to the transformer gallery from the pmverhouse \·+ill be ·by tunnel frcm -the
main access shaft; or by stairw.ay through each of the h:~1ated phas~ ~)'JS
shafts. A service el~vator will be provided from the maintenmtce area on
the main floor level to the ma~.:hine shop and stores i'lrr2'l t;n th~ turb~ne
floor level~
Hatches \'lill be provided ttrcough all main floors fnr in5tii1latic..m anJ main ...
tenance of heavy equipment using tha over~head traveling cranes.
(e) Transformer Galler_x
The transformers will be locat~d underground in a separate gallery, 120.
feet upstream from the main pow::=rnouse cavern, with three connect in;1 t"tjn-
nels for the isolated phase bus: Tl1ere will be nine sing1e-phase tt"2!!1Sfm~
mers rated at 15/345 kV, 122 MVA, installed in groups of three one gr0up
for each pair of turbines. Generator circuit breakers wi11 be required,
and wi11 be installed in the oo'.-Jerhouse on the lm..;er Clenerator floo~~ 1eve1.
. ' -
High voltage cables wi11 be ta~~en to the surface by tltiO cable shafts, each
with an integr·al diameter· of 7.5 feet. r'rovision has been mad~ fo;--~"1sta1-
1ation of an inspection hoist in each shaft. A spare transformer w~~~ be
located in the transformer gal18ry, and a spare HV circuit will also~~
provided for improved re1 iab11 ity~ The stat hm service 1 11Xil iary b"J.t~sfor
mers {2 MVA) and the camp services aGxiliary transformer (7.5/10 MV~~ will
be located in the bus tunnels. Generator· exc-it.:iticn tran:;,formers .... d]1 ne
located in the powerhouse on the main floor.
Vehicle access to the transformer gallery \'li '11 be the main power'hous::: 3C-
cess tunnel at the south-end. Pedestrian access wi11 be from the ';1::t~f'J ac-
cess shaft or through each of th2 three isolatud phase bus tunnels.
(f) Surge Chamber
A surge chamber will be provided 120 feet downstr~~l frcrn the power~:~se
cavern to control pressure rise in the turbine oraft tubes and tai~r3:B
tunnels under transient load conditions, and tJ prlviae s:~rage of water
for the machine start-up sequence. The chamber vii ll oe ~Jzmon _to aJ: six
draft tubes~ and under normal operation will discharge equally to th~ two
tailrace tunnels.
The draft t~be gates gallery and crane will be located in the same Cai.'8rn.)
above the maximum anticipated surge 1evel. The draft tti:Jc gate crar.t? ha;,
also been designed to allow installation of tailrace tunnel intake st~plogs
for emergency closure of either tailrace tunn~l.
The chamber will generally be an unlined r·ock excavation, ,~· .. il'!:h loca1i.:~ed
rock support as neces s ar~y for st abl1 ity of the roof arch and ~tla 11 s. The .,
12-53
gate guioes for the dr~aft tube gates and tailrace sh;p1og;; ;,till be of :~t'in ...
for•ced concrete) anchored to the rock by rockbo 1 ts.
Access to the draft tube gate ga 11 ery wi 11 bt; by an aai t ft'Ohl the 111a in iV.>
cess tunnel; the tunnel will be widened locally for storage of draft tube
bulkhead gates and stoplogs.
12.14 ... Reservoir --
The Watana reservoir, at rormal operating level of 2,18~ feet~ w~ll be appr~xi
mat~ly 48 miles l0ng with a maximum wioth in tha ord~r of 5 miles. The total
watel~ surf ace area at normal operating leve 1 is 38,000 ::1cr·8s. .Just up-s t~"e,}.;li
from the dam, the maximum water depth ;vill be appro;dmatt~1y 6:W feet. The i~~~ni-
mum reservoir level \vi11 be 2,045 feet during normal fJperat1on, y·esulting ;11 a
maximum dra.wdm~n of 140 feet. The reser·voir will have a total capacity uf
9,515,000 acre-feet of which 4,210,000 acre-feet will be live storage.
Prior to reservoir filling, the area below Elevation 2190, five feet above ma~i
mum operating level, ·wnl be cleared of all trees and brush. A field r€C\1nnais--
sance of the proposed reservoir area was undertaken as part of these studies.
This work included examination of aerial pnotographs and maps, an aer'ia1 uV-2¥"'-
flight of the reservoir and collection or recent (1980 field season) for2st in-
ventory data fr\un the U.S. Forest Service. Most of the vegetatal material :.dth-
in the reservoir consists of trees, with very little in brush. The trees are
quite small, and the stands are not very dense. In the Watana reservoi~ ar~a,
an estimated U;;;UUU,OUO cut>ic fe~~t of \~·ood exists averagi11g appr:>ximatt;ly 5JO
cubic feet of lo•tJ comlller·c.i3.1 quality, and some ver-y significant 1ogginq prJn1ems
would be poseo oy the steep s1opes arld incised tt:rrain encot.H1tm·ed in trh: aJ-ea.
Approximately 87 perc2r1t of the avail ab1t2 timt;~.;r-an~ soft ltiJO~"l~~. TtH} r·~,is:.;~ts of
the timoer reconnaissanct: studies ar-t: aescriDi;;d in more detail in Auot?.nuh. C3. . . .
, .
The comDination of steep terrain, moderate-1 ight tr~e stocking levels} Siila11
trees, :2rosive potential 'Jf the rese~~voh· slopes, rt:!mutt:.:ness, and very re-
stricted access to the reservoirs ~·~ major factors ~ff~cting the choice of har-
vesting systems to be utilized for this project. Sucr1 syste:ns include h1;;n
lead, skyline) tractor, whole tree logging with or without chippers, ba11oD~ and
helicopter. fach system has its O\'m advantages and disa.:wantages and set :rf
conditions under which its used is optimized.
Present mar·ket demand for the timber at Sus i tna is l 0'11, ho.·:ever the wor 1 o~·~ ~de
demand v;ood fluctuates consicerably. It is anticipated that use of the hs.r-
vested material would be limited to either sale as wood-waste products ~~d as
fue1.
Slash material including brush and small trees, v1hich will be unsuitab1e ::-0-r~
either of the above uses, will be either burned in a carefully control~ed ~anner
consistent with applicable laws and rt:gul at ions, or-rl3u1ea t~ a disposal sit.e in
and adjacent to the reservoir. Nate:ia1 placed in dispos:11 areas will be cov··
ered ~'lith a earthfill cover suffic ier"!t t•J pr·event erosion and subsequent expcs-
ure.
12-54
I
I
I
I
'I
I
I
I
I
I
I
I
I
I
I
I
I
I
1.
f
l
I
t·
I
I
I
I
I
'I
I .. ,
. •
••
I
I
I
I
I
I
I
12 .15 -Ta i 1 race
Two tailrace pressure tunnels will be provided at Watana to c~rry water from the
surge chamber to the river. The tunnels wi 11 have a modified dL:HilE!L;r cf horse-
shoe cross-sect 1 on \vi th a major i ntt:~rna 1 d:imens1 on of 35 fe.et. For prel imi nar·y
design tiley are assumed to be fully concrete-lined throughot:r., with a miflirm,;_m
c-oncrete th1ckness of 12 inches and a lennth of 1,800 feet ..
The tailrace tunnels \•Jill be arrang~:~d to discharge into the~ r~'JE!r between the
ma1n dam and the main spillway. In viev-t of the SGver-e limitations on space i·1
thjs area, one.tailrace tunnel will be designed to discharge t~rough one of t~e
diversion tunnel portals. The cross section of the tailrace !:unnei will be ~~t--:d-
1f1ed over the common length of 300 feet to the shape of tre·::nversion tunne1 in
order not to impair the hydr"aul ic per~fm'r.Mnce of the tailrace tunnel. P\fte!~
commissioning, the diversion tunnel upstream section will be plugged w'th t~n
cn:te.
The size of the t\10 tailrace tunnels ;·1as selected after an e::onomic study of the
cost of construction and the capitalized value of average annual energy losses
cause-d by friction, bends, and changes of section. In an emergency, however:.
the station can be operated using one tailrace tunnel, with increased head
losses. For such an emergency condition, tailrace intake stoplog guides will be
provided in the surge chamber. The surge chamber ~vi 11 be designed for ~u 1 ~ 1 oad
rejection with either one or b'iO tailrace tunnels in operation.
The tailrace portals will be reinforced concrete structures d~~igned to r~~~ce
the out 1 et f 1 ovt ve 1 oc i ty ~ and hence the vt: loci ty h~:ad 1 oss a: the ;;;x 1 t t a t ·;~
river. The.mtnimum rock cover required above the tunnels ~il1 be 1.5 times the
major excavated dimension (about 5,~ feet}, and the pcrtais ·.dil also provi-::~2 the
necessary transition length to the river where the rock ccv~r wo~ld be 1ess than
54 feet .
12.16 -Turbines and Generators
(a) Unit Capacity
The Liatana powerhouse will have six generating unlts ,n:n a nominal c5na-
city of 170 NW. This is the available capacity with r~nhnum Decembe~ .. ~~es
ervoir level (El. 2112) and J cor~esponding gross hea~ of 562 feet on ~he
station.
The head on the plant wil1 vary f~~om 735 feet !TJaxiillu;n "724 feet net h~s.d)
to 595 feet min 1 mum (584 feet net 1 eve l) . Because na;dnum turbine o:r:put
varies approximately with the 3/2 power of·head, the ~aii~um unit out;ut
will change with head, as shown on Figure 12.21.
The rated head for the turbine hds been established a~ 580 feet, which is
'the weighted average operating head on the station. · ~~lowing for generator
losses, the rated turbine output is 250,000 hp (186.5 ~,:~;).
12-55
.J
I
The generator rating has been selected ·as 190 MVA with a 90 percent power
factor, whi-ch corresponds to a power output of 170 MW. fhe 9€nerat;:,rs ·.·ri11
be capable of a continuous 15 percer.t overload; ·this w1 1! anoN a unit OIJ~
put of 196 t1H. At max 1 mum resetvo 1 r water" 1 eve 1, the turb i n~s !fli 11 be ope~~
ated below maximum (full gate) output to avoid ov~rloading of th~ g~nera~
tors.
(b) Turbines
The turbines will be of the vertical shaft Franci~ tvoe with steel spiral
casing and a concrete elbow-type drnft tube. The dr·'dft t.Jbt; will c~Hr:pr'U s:?
a single water passage without a no renter pier.
The rated output of the turbines wi 11 be 250,000 hp at 6;;D feet rat..;;d :iet
head. Maximum and minimum heads on the units will be 724 fe~!t and 584 f2~~·~
respectively. The full gate output of the turbines w111 he about 275 .,GOO
hp at 724 feet net head and 200,000 hp at 584 feet net he~d. Overgatin9 df
the turbines may be possible, providing approximately 5 percent additional
pm'!er; hm1ever, at high l}eads the turb1ne output ~vill be rr::stl"icted to
avoid overloading the generators. The best eff~ciency poi~t of the tur-
bines will be established at the tjme Gr preparation of bid documents for
the generating equipment and will be based on a detailed ana1ysis of the
anticipated operating range of the turbines. For preli~inary design cur-
poses, the best efficiency (best gate) output of the unlts has been a~s:.n~r:d
as 85 percent of the full gate tu~b1ne output. This percentage may v~r
from about SO percent to 90 percent; in general, a lower percentage rec~(~S
turbine cost.
The full gate and best gate efficiencies of the turbines will be about 91
percent and 94 percent r2spectively at rJted he~d. Tho effi~iency w~l1 be
about 0.5 percent lm.;er at maximum head and 1 p1::1"cent 1t.Mr;r· at minitt'!~;::~
head. The preliminary performance curve for tho turbine is shown on ~i~ure
12.22.
A speed of 225 t~pm has been s~1ected for the urdt for preliminar·y des.~;n
purposes. The resulting tarbine specific speed (Ns) is 32.4. As sh0~~
on Figure 12.23, this is within present day practice fDr turtdnes op~~·?:tinq
under a head of 670 feet. In general, a lower speed ~ac~ine will inc~eas~
the cost of the turbines and generators as well as the ~ow~~house civ~:
cost because of the increased physical size of the ge~er~ti~g equi?ne~~. A
hjgher speed unit, on the other hand, requires a deeper ~nit setting ~rd is
generally considered to be a less conservative design with increased ~is~
of vibration and rouah oueration. The differencE~ in efficiencv het\-.ee·: t,he .. , ' .....
higher and lower speed machines at this head ra~ge is v~ry s~all, w~tn the
increase in efficiency which is associated with a phys~:31ly larger (~~w~r
speed) runner offset by higher disc friction and sea1 !ea%ag~ losses. For
an undergn:)und powerhouse, the incremental cost of I r:cre~s i ng the ~mit
setting is usually relatively inexpensive; therefor~, as;~rning no ch3~9~ in
efficiency, the trend in unit selection is to choose as hi9h a speed as
possible consistent with .satisfactory precedent and g0od operating e:x;,?er·i-
ence with similar specific speed turbines. Draft tube vo:tex and surge
12-56
. , I,
t
I
I .• ,
I
I
I
I
I
I
I
I
I
I
I
I
I
I
. ' .. ,. ··,.
I
1·~.
.
I
'1:
I
I
I
I
I
I
I
I
I
I
I
I
phenomena may also have an influenc~ on the selection .of u~lt speed as
discussed below. The ttwbine data is summarized in Tat71e 12.7~
On the basis of lnformat1cn from turbine manufacturers and the studi~s on
the po\'ter plant layout> the centerline of the turbine distributor· has bt1en
set at 30 feet below min1mum taih;atet level. The final s~:tt~ing of the
unit 'v'lt11 be established in conjunction w1th the turbin:; !i1anufacture~· ·~·;hDn
the contract for the supply of the t:.n~bine e.quipment h,j-; b~?en a~·i~trl\2·~~.
Tht turbines will be of conventional design, f1enera11y >if welded Or' ·c15t
steel construction with forged steel shafts and pins. 9ecause o~ t~e
remote 1ocat1on of the project and the desir·ed high rG·~~ability/av~i,.,atli 1-
; ty of the eqtd pment, spec i a 1 .cans i deration shou 1 d b~ q Lten t n red:.;c h~~
cavitation pitting on the turbines. This will include:
-Provision of weldable stainless steel runners;
Careful profiling and finishing of the water passage
runner ~·ticket gates and stay r·ing;
- A conservative unit setting; and
-Extensive cavitation tests on the turbine model.
surf .1ces of
Bulkhead dm?tes wi 11 be: provided with two of the turbines (Un1ts 3 ar:.d .:!) to
be installed at the bottom of the draft tube liner at the time of t~~~~ne
installation. The domes perm1t work to continue on turbine insta11a::Jn
after thd tailrace, surge chambe:~, and draft tubes are f1ood•:d (pi"L:r !~.:>
startup of Unit 1), without installing <.h~aft tube gates.
Because of the relative1y short lengtn of tho intake penstock and a s~rge
tank location immediately downstream of the powerhous~?, the hydr·au1 L:
transient characteristics of the turbines are favorable. Assuming ~c~~al
generator iner·tia (H == 3.5 f/1~4-Sec/MVA), a preliminary ~:1alysis has i~~:;i~
cated the following:
-t·Jater star·t ing t tme (T'.'I) ............................ 1.6 seconds
-r.·lechanical starting tim~ (Tm) ....................... 6.6 seconds
-Regulating ratio (fm/Tw) ............................ ll.l
-Governor time .......................................... 6.0 seconds
-Speed rise on full load rejection ................... 42 percent
-Penstock pressure use on full lead rejection ........ 30 percent
The regulating r·ati0 is above the minimum recor:1~ended b~: the USBK fc•"' ~-~od
regulating capacity. Also) unit speed rise and penstock pressJre r~s~ are
all well within ·normal accepted values. Because of the deep unit set~1~y
and the relatively short distance bet'.veen the tw~bin::: and the ta11r.~,..'=~
surge tank, there will be no problems with draft tube water coJunn s~para-
-~-. ... lon.
The Hata11a project will form a 1 a.rge port ion of· the over a 11 system g2~'erat
ing capacity in Alaska; therefore) satisfactory operatic~ of the units over
a very wide range of loads wi 11 be 1mportant. Althou~~h there are Frc.r~.~i s
turbine installations which operate for considerable periods at very s~all
gate openings, operation belo~1 about 50 per·cent load gEr.era11y beco1: .. Js
increasingly rough because of reduced efficiency of the t~rbines.
12-57
0
The ab i 1 i ty to op~rate .at par't 1oad vii 11 a 1 so dt=pend 0n the draft tube
surge ph2nomena JJid ·associated generator· PO\'ler s·.-Jings ... The su~ging occurs
on many turbines, partit:u1.ar1y thr. Francis type, and narma1ly has a fre-
quency of about 1/3 to 1/5 of the turbine rotational s;1~ed. Thesa S:.H"'ges,
,,,h ich may occur from abtHlt 30 percent to som~t imes as h i1h as oO pet .. Ce!1t
wicket gate open;ng., result in puh~ating tor·q~Je on the turbine runfl~r· and
corresponding g~ner a tor po.wc:r f1. uctuat ion$. Thr; cond 1t ~on tH;.comes rnore
severe when the sur·ge fn1quency c1os~ly cncr"'!sponds t'1 the rnt.;ral f~-e-
quency of the gener JtiJr. To rt:thh>~ the pos~:; in 1 'I i t,y of •.~n rlCt>:-pt 1ti 1 ~:: pO'•'l.::O~"
swings and unit vibration, it is lif~sir·able to hj.Vt~ V·~ ~wr~;.:; ftd~~WL!ncy
d,'f,t:;~rent r,-.ro1~ t'h·e o· (.'JJ"'.-+~rJK pr·li'C'Jy.--• •li;,\I.M. r"~,.._.l.l r···~~~v "t•..-4 trj,•· /''"r't'<J""'"'">.>' I ..,. . h , ..:..11 .:> t.IJ ... 1\ t! ..') '::' '-.0 t: •n:<, ':::: • l";..l.'l})~.!lt\..., ~.,>I.< ,., F,i ,p.;; ,r;. C ""'·''
natural frequency.
Ti1e estimated no.tura1 fr·equency or tne gener~otors Ni ~ 1 be at;:Jut 1. 3
cycles/s~ \'lhich is unoesirdb1e •·,.hen cons ider"ir1g a poss io1a s.ycge frequency
of about·o.7S to 1.25 cycles/s.
The selection of a lO'tter' unit S·peed 'tiould separate the draft tube sw~ge
frequency from the gener,atcr natural frequency; hm·Je':er > this ~·li 11 1ncrease
the generating equipment as we 11 as pm"·erholise cost. lt may be poss it 10 to
increase·th? natural frequency of the generator by 'feducing the generator'
inertia (wH2) as much as possible; however, the lowe~ ~R2 has an
adverse effect on the transient ch-aractel-istics of t-::e Jnit and may aff2ct
electrical syst~m stability.
Careful model studies of the turbine to accurately ~r~dic~ draft t~~~
Vo rtex/s·u·rgi=l Pt1r-lt'•o·",",·.!lln?<-01 t'"""~ng ~.<L1-s''1"'··-.,,; .:.r:c; .. ,., .. ; .. t.•")~'l t·a·l ~"':::s·'·!·r-"' , -~ -.. I"'-"~ o c ~~"...-, .. ,, .t!:~ t. t~t l::! , ~.~_r.:t: u-. \.,..-! .... · ~· :.~~o~ ,,.... ~ ·).~" ·'~ ~-..., • ~ -U~--t'::ll\:'i
surge problems; howeVef, a r·~d~JC t ion of pt.a~t-1 0 ad ;;;~f ic h~~Ky mawv f~St~} t.
-Pr~ovision for air admission to the draft t.~lbe. Th~5 is done on rncP~~ or'
less a trial and error basis ana mav includt! inie~:":: . cf 5ir frc1:! ':h2
station 100 psi COt:tpressea o.ir-syst~m, spc!C iGJ.1 1t.•·i iJress~.r,.·e compn:?:SS:Jrs
•d • C • 1 1 f ~ b • l ' • I • • ~ prov1 ed specn 1ca. 1Y or ufart tu e a11~ ao!ll1$S1Gl1~ ac;aicw prov1s1 ot
an 11 air-head 11 ~·mich allo~·Is atr:~Dsph2ric a.ir to b·.~ c:~·::M: dc.m the ge:-;-,;;r·att)f
and turbine shaft aod through the runner o.::nt:.:.
-Additions of fins t8 the i'lt'aft tube cone h'm~~dia.te~J bela~·; the rLm::~t·.
\4hi1e this has ne~~n us,~d on 1:~any insta11aticn:;~ ::~ere have teen i~s:ances
of structural failure of the fins.
"not· n' aY. ::~nnro;lrl., r'tflr"C.l1+-1y ••nn~., .. s·blr1Y, ~)v rn·· nq~:.,: ~r. '-n . .-1;::~ ... ~·r'1o t')J" .• ·.,;_;,.. .n ~t ~tJfl \.A'l.#tJ '-"'t._~ · "'-'~ t_. ~~ u._..; ._ y\.. "\J -J ;..,;;\Jt': \o-.J..J-~i\ "woP w'u ...t·~ .. ;.p 'r..-._ ,t'""V4i-'t-..,..,;;
oscillations resulting frcm the draft tube sJrges by va~yi~g the generst0r
excitation~ Computer simulation irwicatas that is f~asioie tJ damp~n th~
large oscillations; however> this has yet to be t~sten in a prototype
unit.
Employing one or' fiiure of the above approaches, a d>2s;gn may be achievt:d
that operates satisfactorily \vithOllt serious gt:nerat:=..:r power· svJings. The
potential problem, however, must ae given serious consid~ration in th~
design stages~
12-58
t
I
1
...
.
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
l
I ... . .
~·
I
I
I'
I
I
I
I
I
I"
I
I
I
I
I
I
:..,_,. _
_ , .. · .
I
. {c)
.::,
Gener a.tors
(i) Tvpe and Ratino
The six generators ;,'l the Watana powerhouse; w:n l be of the vert :cal
shaft, overhung type .l·\rectly connected to the ver-tical Francis tur-
·Lines~ The arrangement of the unit-s is shown in p],ates ana the
""''t~J!~-
s i ng le 1 i ne diagram i ~. sho\·JO in P 1 ate . -
The optimum arrangement at \~,~tana will consist 'If tv,;o gsnerator~s p~r
transfnrmer bank:~ with each transformer bank ctJmpris ing three
singl-e-phase transformets. (Development of this scheme is desct·ioeu
in Section 12.18)4 The generators will be connected to the trans-
formers by iso1 ated ph as~ bus through generator circuit or~ake~s
directly connected to the isolated pnase bus ducts.
Each generator wi11 be proviclt-~d \vith a high in'itia1 respo11se st:=tic
excitation system, The units wi11 be contro11ed from the viatJn~
surface control room~ with local control facilfty also provide-a at
the powerhouse floor. The units will be designed for black start
operation.
The generators ar·e ~"'.:\t~~d as fo11m•ls:
Rated Capacity!
Rated Pov1er ~
Rated Vo 1 tage:
Synchronous Speed:
Inertia Constant:
Transient Reactance:
Short Circuit Ratio:
Efficiency at Full Load:
190 MVA, 0.9 power factor
170 Wtl
15 kV, 3 phase, 60 Hertz
225 rpm
3.5 kW-sec/kVA
2B percent (max imur.t)
1.1 (minimum)
9B percent (minimum)
The generators \>Ji11 be of the air-~coo1ed type, ~lith water-ca;..,:::;:J
heat exchangers located on the stator periphery. h'1e ratings ;:iven
above are for a temperature rise of the stator and rotor 'tlinc~·1;s
not exceeding 60°C \vith coo1 ing air at 40°C-
The generators wi11 be capable of delivery 115 percent of rate·: :..VA
continuously (195.5 N~l} at a voltage of +5 percent ~·rithout ex=-;;~ding
BOOC tempei~ature rise in accordance ;·lith t-\NSI Standard CSO.lC.
The generators will be capable of continuous operation ftS synchron-
ous condensel"S Whtln the turbine is Um'latered, ltdth an underex~ited
reactive power rating of 140 /ltiVAR and an overexcited rating c-t lliJ
t~1VAfL Each geoerntor ,,,lill be capable of ener•gizir.g the trans;-:13ssion
system without risk of self-excitation.
·The design data of the generators stated above should be revie.-~ed
during the detailed design stage for overall economic ana tec~nical
design and performance r·equ 1rements of the pm·;er p1 ant and the powe;
system.
0
( i i)
The generator will be of a mad ified umbrel1 a type ov'~~rhung const""tlc-
tioni) Nith a combined thrust and guide bearing be1o•;t th~ roto~--and a
guide bearing abovt;~t_he rotor._ The lower oearit1g brack>llt wi 11 s~p
port the rotor and turbin1-1 tum1er weights and the unba1t1nr:ed hy,jrau·
lie thrust of the runner. All removable parts, inchtding tur·bin~:
p_arts~ wi11 be designe·j for removal througi1 the ':]erwr1t>:w st~tur.
Approximate dimensions and weights uf the pr·incipal P·:trts ~f
generator are given below:
Stator pit diameter
Rotor' diameter
Rotor length (without shaft)
Rotor weight
Total 't'le .;ght
36 feet
22 feet
7 feet
3B5 tons
660 tons
It should be noted that these are approximate figures and th~y \dl1
vary between manufacturers.} sometimes considt:rab1y~ However·) at
this stage of design feasibility and planniYlg) the dimensio~is and
weights are considered appropriate and repr2sentative.
The generator stator windings will be insulated with G1ass b ins~la
tion as clefint:~d by ANSI Standard CSO.lO) of eJ;oxy resin tron;::~·J :.:.J~e.
The stator windings will be wye-connact~d fur gr0unded op~~~~=oo
through a neutra 1 ground 1 ng trans formet' 1 uc i.ite1 in t.h~.: ~t.~:h:~ .... ~t~r
neutral cubicle. The sta!:nr windings and lari!ir:ated-c~we v:n~ be
shop-assemb l i?.d in thre~ or four sections f;Jr f ac i 1 i t:y of tr· -~~·1s;:cw~
and erection in the powarhous~.
The rotor wi 11 be designed to s :1fe 1 y '.4i th:Jt : .. H1:1 tne maximum ~· :~·~ .:t'day
speed of the turbine. The rotor hub, yok~J, ~·-;:i laminated ~"'~n :'l1i11
be designeo for assemb:y at the powerhou::.t~. 7•H:: assen:b1~d i~~~t.;r·
will be erected in the genc:rator pit without '!:he shaft, :ne;"~:>y f'e-
quiring min iri1um crane 1 ift ano a cons idl;!t~at 1~ t>2duct ion of t.:~~
powerhouse cavern height.
The rotating parts of the generator and turtine will be desi~nej sJ
that the critical speed exceeds the run a·t~a.t Spee:d of the un i ~ Jy at
least 20 percent. The design of the plant ard po'Jier system ~..v~! 1· not
require ad~itional inertia in the rotating psrts; the inertia con-
stant specified thus will correspond to the unatura1 11 inert·f3 of the
machine. Damper windings of low resistance anj rugged constr~:tion
wi11 be provided on the po 1 e shoes and des ig;;~.:; to pr~ov ide a~~~quate
damping currents for stabilized operation.
The thrust bearing '."''ill be of the adjustable srH::e (Kingsbur}) or
precompressed spring (General Electric) type, oil-cooled, with high
pressure oil inject ion our ing :>tart ing and st-:>pping.
12-60 :· -~-
I
' I .
;: ..
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
"
I
"':
I
I
···"·. . .
I
•••
•
'
I
I
I
I
I
I
I
I
I
I
I
I
I
(i li) tienaratOt~ Excitation S.1:2.~ei!~
The generator will De provided with a high initial r~spons~ type
't .t. ::. . it:)""·~. ,...,_,....;.. ·r. .• t ··1 i ::~4 ., • .J·H 't' .... C:f'~ ~"" -.. v-·~ ::..1..: ·n n,...h!e"" sa 1c exc .... ~.-H.m ::.>J:.,:.t: •• l Sd~"'IP .t-~~ H•·~·1 r~c~..~,,~H t.ht_l~.-'.A;.,o. ~-'""'l1.
fr-om tJ~ansformers connected ah·ectly to the genetiJtor· t~:H·minals,
The exc:itat ion syst~m wi11 be capabh;! of suppJyir19 200 petcent of
rated excitation field (ceiling voltage) with a generator terminal
voltage or 70 pet·cent. The power rectifi~rs t-Iili hava a one ... third
• .. • ~· t ... ..t..o"" .1 ... spare capacity to ;~t:nnt211n :F:rwr·a~.1on t3'1en (Jl.H'H1g nu ure tJr a com-
plete rectifier modul~.
The excitation system wi11 be equipped \·d!.£1 a f:;l ty static vo1taye
regulating system maintaining output fro~ 30 percent to 115 p~rcent,
~·lith in +0.5 percent accuracy of tne· voltJge s,2t~ ing. ~ianua1 :ootr•o:
will.be possible at the excitation board locat~d on the powerhJ~s~
floor, a1 though the unit wn 1 ntH'ma·l_ly b~! undt:!r remote control 'I as
described in Section 1~.18 coverin9 tht~ control syst;:-}mS of the
plant.
The static excitation system ;·1n l also includ~ :n~:dmurn and mHn:nun;
1 imi ters and react ivt! current compensator, and ~·~i! l be su itao 1e fO!''
paral1e1 joint cont~~ol of the units. Fi~ld fl::shing during st:lrtup
will oe from the 125 volt de station battery.
(iv) Erection and Tests
A . , -1 h ~ t . t ,, l. • .. , ' s 1 s norma. ror i :ir·g~ .r:!n)e f ec r 1 c generd .ors, -cne macn mes :,-;1 1 ~
not oe assembled Ci.1[;~p1ete1y and tested in the factory~ The e"""eCtion
and tc;sts of the generatot~s at the powerhouse, therefore, wil ~
as s.ume greatBr impottance in the successful ccm;r: is~ ion i ng of tht:
station and should ~e carefully coordinateo witn that of the tur-
bines and civ1l works.
The assembly of the stator sections will be aone in the pit. The
rotor 11'1i 11 be assemo 1 ed in the er·ect ion bay. -:-he powerhouse c~~a:ll~
will be capable of 1 ifting the comp!eted rot1~r ass~mb1y and l:t.lier·ing
it into the stator~ m1d onto the thrust be?.ring and shaft ass~r;b1y
on the bracket supports. Alignment and tests of the rotor) t~r~i~2
runner, and shaft will be done to tol·~ranc::;s specif~ed in PtB\;h:ANSi
-c-~ ' . ' d .)J.. an-l1 ar s.
The generators will ne fu11y tested aftar assembly and mecha~L:a1
run tests, including die~ectric tests, satura~ion tests, heat ·un,
efficiency, and fu11-1oad rejBction tests. (ei1ing voltage ,:.,;" . .1 ~,;e
sponse of the excitation sys:em will be tested. ~perati~n cf the
unit within specified vibration limits will be chec%ed.
(d) Governor System
The governor system which control the generating unit will include a gover-
nor actuator~ and a govarnor p..tmping unit. A single system will be p~"':Jvioed
12-61
fur each unit. The governor syste~ operdting pressure w~11
psi, as t·ecorr.m-anded by tha gov~rnor· system r.:anuf ao:t ~u-r~r.
The gove:"nor actuJtor wi 11 be the electr'ic h,yd~·1n~ i:; t.y!Jt: .1~Ht ·\-;; ., 1 ne con-
... t•r:.t~n +.:J ttJ~:~ comput£_,rl·z~~d r.::t;1t:1.rJn. coqtr·ol -:J-..· .. ":'hp ;;H,~v.::.,...n.,r· n.f't'i'"'~.;,"' It---..... ~..J. _.._ -'-~-w-¥ -....-.... u ..;J' ...,; 1t ~at,.._ .,:j·:...l "J' "lV :t~~'" .Jl:-l(:o::¥¥-c;~ _
unit wiil include governor pumps, an accumulatol t·)n~., .an(i a s~,~Np t::~n~.
Each unit wn 1 have. three gavern,;r pumps: two ir~ ~.H.ll;'p;, ;·;h ich \}pr,;t• ~t;;::
intermit t;,:'!nt 1 y, and one jock~;y lJWP.p which ope~· ~Jt>£s c;)n t i w-t>Jus 1 y ~~hi h:~ tb;;;
turbine 'tjicket gates are op• . .::n tmd intermittently litl~J~~ t'E= ~ptes Jr'.:! f";11y
c1o~ed.
1-
12.1?·-Niscen~r:~~ous ;-.1-::char~ic:al Ecg,dpment
I
I
I
I
(a)
~ ,_ .. ____ --
Powerhouse Cranes
Two overhead traveling bridge type powerhousE~ c:~an';r': ·wi l! .o~ inst.J.1 h: . .:i 1n
the powernouse. The cranes will be used for:
Installation of turbines, generators, '"'r·d o"-"'· .,, I..J. l L:lt!t
Subsaq~ent dismantling and reassembly of
overhat~i s.
cranes.
I , ..
I
Each c ran;, will h;}, e a ;na in an,J aux i1 i ilt' y h•J L;:: . Th ~ c·;·.:.!:d :·rc:d c ·1:1 • • :} of I
tne main hoist for both CtJJL:~s will be suff1ch:nt f•.~r ~:h17• !i·.~~Vh!-;t ~·: ... : .. ·~-. .
~n.snt lift, which will De the y2neratnr 'fOt.:lr', pl j!t., •?•,J•~.i!tzing tH::,l:~. A
tentative crane capacity of 205 tons has bet:'l c:·,t,.,;,:isl-:f!•l. Th·~ auxi:i~fy J
noist capacity \•.'ill be about ;~:'? tons. ·
The powef·hou se o~ arH~s \1/ i 11 bt3 cab con tn) l12rl~ Cor~: L:~2r ::t ~orl may
given to proviaing radio control for the crdna~.
a '!~"'-""" ll _:)· ..•• ;.
Jraft t~b2 gates will be provided to pennit d~~d~erirg cf the turbin~ ~ater
...,~S"'a'"'0 ~ ,c,Y' ;,..S.,e-""l'on and "~'a1·ntan:':lnre·· Of-.J-\.-12} ..... t·~•~r.:.~ Tne -lraf.~. '"'·~·.::.'> ~'\.4 ~ ~~..J !v, lll .,._.,. ~~ ... t )t ~J:>i v· .... !J. 1,.; ....... ro;:: \..1..1 LttJ~-.:). ~~f U' ~ :.:..-...i'""'~';:
gate openings tone opening p~?r· uni~) will oe 1ocat.~d ir the sur·ge ct::·~.::~.~r·.
Th:::~ nar"';;:;. ''111 hr.) of th;:l bul!/h~~ri t··;·f)O ,·,,..,,sta11c,,.l :·(·,··~;:..·!'" h:;:-';;~nrpr! ·n· >:)"":'"' '~n t.Jtt".. ~~.~c_. -.,f, 1 u--.. ""'"'-"",.-t;,..l ..... t......, -, 1 11-J -.A•---:..r..-.tM v-u ._.:._:..Ji \.,"• . ._
ditions ~sing the surge chwnber crane descrioeti o~~o~. Fo~r gates n~~~
been ass~ffif2d for the six units, ·.-Jith each gate a 3 iP;;,; 1e :~~f, 2U fee: ny 20
feet.
t~hen Un ;: 1 is ready fm" startup,
5, and b~ with one gate avail ab1e
h.~ 1·n~+~1 ~~a· ~~ 'u•1~~-~ ~nd· 4 '"'-H -'H,• ,.\.# i ~· !. , l I t,. ::::> v v.. • •
12-62
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
'
•••••• .~.:.
I ..
I
I
I
I
I
I
I
I
',' "-
(c) Surge Chamber Gate ~rane
A crane will be installed in the surge chamber for installation an<!, removal
of the draft tube gates as well as the tailrace tunnel intake stoplogs ..
The crane will either be a monorail. (or twin "monorail) crane, a top running
crane~ or 'a gantry crane. For the· .pre1 iminary design, a twin mOnorail
crane has been asstJned. The crane will be about 45 .tons in capacity,
pendant operated, and will have a t\'MJ point lift.. A follower wi11 be used
with the crane for handling the gates and stoplogs. The crane will nor-
mally travel along the upstreMl side of the surge chamber; however, the
crane runway will have a transfer mechanism for mov;.ng the crane to the
downstre~ side of the surge chamber for install at ion or removal of the
tailrace tunnel intake stop logs.. The crane runway will extend over the
tailrace tunnel stoplog storage area at one end of the surge chamber.
(d) Miscellaneous Cranes and Hoists
In addition to the powerhouse cranes and surge chamber gate crane, the
following cranes and hoists will be provided in the power plant:
-A 5-ton monorail hoist in the transformer gallery for transformer main-
tenance;
-A 4-ton monorail hoist in the circuit breaker gallery for handling the
main circuit breakers;
-Small overhead jib or A-frame type hoists in the machine shop for hand-
1 ing material; and
-A-frame or monorail hoists for handling miscellaneous small equipnent in
the powerhouse.
(e) Elevators
Access and service elevators will be provided for the power plant as
follows:
-An access e 1 ev ator from contra 1 bui 1 dings to powerhouse;
- A service elevator in the powerhouse service bay; and
-Inspection hoists in the cable shafts.
For preliminary design purposes, a 12,000-lb, double-deck elevator has been
assumed for access to the powerhouse from the control building.. The ele-
vator will be locat-ed in the access shaft and will trav.el at a speed of
about 500 ft/min; it will be operated by a friction type hoist located
above the elevator shaft.. The elevator will have a single landing at the
control building plus four underground landings.
·The service elevator in the powerhouse service bay w·ill have a capacity of
2,000 to 4,000 pounds and will provide access to the various powerhouse
floors. The elevate,-will travel at about 100 to 150 ft/min and will be
operated either by a friction hoist or a hydraulic cylinder.
I
"Alimaku type rack and p1n1on man hoists have been assumed for the cable 1· sh~fts, to use in inspection and/or maintenance of the oil-filled cables ; . '·
and control ca.bJes. The hoist would also provide emergency access from the
power plant.. Each hoist will have a capacity of about 900 pounds and 1,,·~·
t rave 1 at a speed of approx in!itte 1 y 130 ft/mi 11 • .
(f) Power Plant Mechanical Service Systems.
The mechanical service systems for the pOwer plant can be grouped cinto six
major categories:
-Station water systems;
-Fi~e protection;
-Ca~npressed air;
-Oil storage and handling;
-Drainage and dewatering; and
-Heat;ing, ventilation and cooling.
( i) Stat ion Water Systems
The station water systems will include the water intake, cooling
water systems, turbine seal water systems, and domestic water sys-
tems ..
The water intakes will supply water for the various station water
systems in addition to fire protection water. The water can be
taken frooi the penstock; however, pressure-reducing valves will be·
necessary because of the high pressure of the water {about 330 psig.
maximum).. Alternatively, water can be supplied from the draft tube
using pumps to provide suitable pressure. For preliminary design
purposes, the 1 atter approach has been adopted with a water intake
at each draft tube. The water will pass through an automatic back-
wash strainer which will limit the maximum particle size in the
water to about 1/16 inch. An interconnecting header will permit a
strainer to be taken out of service without affecting operation of a
generating unit.. Each strainer will be .sized to handle the water
requirements for two units.
On a unit basis, cooling water will be required for generator air
coolers, turbine and generator bearing cooler~, transformers~ and
powerhouse unit air coolers. The total cooling water requirements
for each unit will be about 4,000 gpm. In addition, the compressed
air systems in the service bay will require approximately 100 gpn of
cooling water .. One cooling water pump will be provided per unit
which will take water fram downstre.am from the water intake
strainer. To ensure suit.able reliability, the cooling water pumps
for two units will be interconnected, with each pump capable of
handling the flow for both unitso Two cooling water pumps in the
service bay will handle compressor cooling water requirements. The
cooling water for each unit will discharge into the turbine draft
tube, while the compressor cooling water will flow into the station
drainage system~
12-64
I
I
;I
r
I. .,
I
I
I
I
I
I
I
I
I
••
I
I
I
I
I
I
I'
I
I
I
I
I
I
I '~'·
I
I
I
I
I
Turbine seal water wi 11 be supplied to the seal on the main shaft
and to the runner seals when the-unit is spinning in air(i .. e •. , in
spinning reserve mode)~ Filtered water may or may not be required,
depending on the type of shaft sea.l. If no filtration is needed,
the seal water will be _taken directly from the high-pressure side-of
the cooling water pumps. If filtration is necessary, a single ·
system will be provided for the powerhouse .. -The system will have
two filters and two pumps which will take water from downstreClll from
the water intake strainer and distribute the water to each unit via
a looped header ..
Domestic water wfll be req4ired for the washrooms, lunch rooms,
drinking fountains, and a service sink and emergency eyewash in the
battery room. Peak domestic water requirements are expected to be
about 30 gpm. The system will-have two pumps and a hydropnellflatic
tank Q Water wi 11 be taken from the water intake system and will be
treated by chlorination or-other means as necessary.
(ii) Fire Protection System
The power plant fire protection system will consist of a fire pro-
tection water system with fire hose stations located throughout the
powerhouse and transformer gallery; sprinkler systems for the gener-
ators, transformers, and the oi1 rooms; and portable fire extin-
guishers located in strategic areas of the powerhouse and transfer-
. mer gallery.. Carbon dioxide could be used in the generator r-ather
than a sprinkler ~Jstem; however, the water system is re,commended
because of the $a~ety hazard of C02·
Fire protection water will be taken fran the station water intakes.
Pressurized water will be provided by a pumped system with two main
fire pumps as well as a jockey pump, or alternatively by a head tank
with two supply pumps which keep the head tank full. For prelimin-
ary design purposes, a system with a head tank has been selected
because of the increased reliability of the system. With an under-
ground powerhouse, a head tank can be provided quite easily at a
suitable elevation as an ad it to the access shaft.
The capacity of the head tank will be about 100,000 gallons; the
tanks will have two compartments to permit draining of half the
water for inspect ion and maintenance. For reliability, the water
supply ptJnps will have two electrical power sources.
Fire hose stations will be provided on all floors of the powerhouse,
in the transformer gallery, and in the bus tunnels. Service water ·
outlets will be in.stalled at the various fire hose stations to
supply water for washing downs floors or equipment. The sprinkler
systems for generators, transformers, and oil rooms will be the dry
deluge type, operated by a solenoid valve which in turn will be
activated by detectors in the respective area.
\)
ll-65
The portable fire extinquishers will generall.Y be carbon dioxide or
,a dry chemical type. ·
(iii) Compressed Air Systems
" Compressed, air will be required in the powerhous·e for the following:
-Service air·;
-Instrt~nent air;
-Generator brakes; .
-Draft tube water leve-1 depression;
-Air blast circuit bra.:.:.~a~rs; and
-Governor accumulator tanks.
For the preliminary design, two compressed air systems have been
assuned: a 100-psig air system for service air~ brake air, and air
for draft tube water level depression; and a 1,000-psig high-
pressure air system for governor air and circuit breaker air. For
detailed plant design, a separate governor air system and circuit-
breaker air system may be provided.
The service air systems will have three aiv-compressors of the ro-
tary screw or reci procafing type, each with a capacity of about 200
cfm. The sY5~em will have four air receivers, two with approxi-
mately 800 ft capacity used for the draft tube water level de-
press ion system, and two with approximately 150 ft3 used for ser-
vice and brake air. The system will be designed to give priority to
the brake air system!! Service air piping with air hose stations
will be located on all floors of the powerhouse and in the trans-
former gallery.
The high-pressure governor/circuit-breaker air system will have
three reciprocating air compressors with approximately 30-cfm capa-
city each, and three small air receivers •. The governor air system
wi 11 supply air for initial filling of the governor system accumul a-
tor tanks and for makeup air to replace air lost through leakage and
air dissolved in the governor system oil.
The circuit breaker air system will provide compressed air for oper-
ation of the main breakers. To insure dry air for the breakers, the
air will be stored at 1,000 psig and then reduced to about 350 psig
for operation of the breakers.
Instrtment air will also be taken from the high-pressure air system.
( iv) Oil Storage and Handlin.9
I
I
·1,,;· ..
' ;:' , .. f
'I
I
I
I
I
I
I
I
I
I
I ... . ..~~-
' -----·-----...-;10'·-
Facilities will be provided for replacing oil in the transformers
and for topping-up or replacing oil in the turbine and generator
bearings and the governor pumping system.. For preliminary design
purposes, two oil rooms have been assumed, one in the transformer
gallery and one in the powerhouse service bay~
12-66
1
I
I
••
I
I
I
I
I
I
I
,f
I
I
I
I '"·
I
t ,,
I "·
I
The transformer gallery will have two oil storage tanks, one for
filtered oil and the other for unfiltered oil.. Each tank will have
a capacity at least equal~""to the volume of oil in one transformer
(about 8,000 gallons). · A header with valve stat ions at each trans-
former will be used for tr·ansferring oil to and from the transfer-
. mers. Oil ~·fill be transferred by a portable pomp and filter unit.
A similar system will be provided in the powerhouse with a filtered
and unfiltered oil tank and distribution header with valve stations
at each unit. The oil tank capacity will be equal to the total oil
vo'il.l!le for one unit (about 3,000 gallons).
During the detailed design stages, consideration should be given to
the use of mobile oi.l tanks located in a parking area near the
po~erhouse and transformer gallery, near the access iunnel.
(v) Drainage and Dewatering Systems
The drainage and dewatering systems will consist of:
-A unit dewatering and filling system;
- A clear water discharge system; and
- A sanitary drainage system.
The dewatering and filling systems will consist of two sumps each
with two dewatering pumps and associated piping and valves fran each
of the units. To prevent st-ation flooding, the sump will be de-
signed to withstand maximun taiiwater pressure. For preliminary de-
sign purposes, submersible dewatering pumps have been assumed.. Ver-
tical turbine type pumps can also be considered; however, since the
dewatering system acts as an emergency drainage system, the pllllp
columns would have to be extended so that the motors are above maxi-
mLm tailwater level. Another optinn is turbine-driven pumps, but
these are generally very costly. A valved draft tube drain 1 ine
will connect to a dewatering header running along the dewatering
gallery. The spiral case ·will be drained by a valved 1 ine connect-
ing the spiral case to the draft tube. Suitable provisions will be
necessary to insure that the spiral case drain valve is not open
when the spiral case is pressurized to headwater level. The de-
watering pllllp discharge 1 ine will discharge· water into the surge
chamber. The general praced ure for dewatering the unit wi 11 be .. '
close the intake gate, drain the penstock to tail water 1 evel thrc. ~~h
the unit, then open the draft' tube and spiral case drains to dewater
the unit. Unless the drainage gallery is below the b.Jttan of the
draft tube elbow, it will not be possible to completely unwater the
draft tube through the unwatering header. If necessary, the remain-
der of the draft tube can be· unwatered using a submersible pump low-
ered through the draft tube access door. Unit f i 11 i ng to t a i lwater
1 eve 1 wi 11 be accomp 1 ished from the surge chamber through the
dewatering pump discharge 1 ine (with a bypass around the pumps) and
~then through the draft tube and spiral case drain 1 ines. Alterna-
tively, the unit can be filled to tailwater level through the draft
tube drain line from an adjacent unit. Filling the unit to head-
water pressure will be-accomplished by 11 Cracking" the intake gate
and raising it about 2 to 4 inches.
The clearwater drainage _system wi)l handle normal drainagec-into~-the
po~er plant .. · Drainage wi11 be collected by a network of floor
drains, trench drains, pressure relief drains, and equipment drains
which discharge into gravity drainage sumps where items are pumped
to the surge chamber. The station will have three main sumps, two
in the powerhouse adjacent to the dewatering sumps and one in the
transfonner gallery. Smaller sumps will be located in appropriate
areas such as the elevator pits and the upstream drainage gallery ..
The sumps in the powerhouse will have submersible pumps for the same
reasons as discussed above for the dewatering system. The transfor•
mer gallery will have vertical turbine type p1.111ps.. The drainage
sumps in the powerhouse will have an overflow line which will dis-
charge water· into the adjacent dewatering sump should inflow into
the drainage sumps exceed the capacity of the drainage pumps. The
overflow 1 ine wi 11 have a flap valve to prevent reverse flow from
the de~atering sump.
Particular care will be taken to prevent accidental oil spills from
being discharged into the powerhouse. The following provisions will
be made:
-All three mair~ sumps will have oil contamination detectors to ob-
tain the pressure of oil in the sumps;
-Drainage into the sumps will_ first pass through an oil separator;
-Controls for the drainage pumps into the transformer gallery will
be interlocked with the tr.ansformer fire protection sprinkler sys•
tern. Activation of the sprinklers, which signifies a transformer
fire. and the possibility of a major ail spill, will prevent the
drainage pumps from starting until the drainage sump is almost
full. It wi 11 be pass ib le to retain about 40,000 gallons of oil I
water in the sump before the pump start (each transformer holds
about 8,000 gallons of oil). In this manner~ it will be possible
to retain a large anount of oil ~~:~ the sllflp where it may be
skimmed off; and
-Suitable oil retention curbs will be provided in the oil rooms ..
Sanitary drainage fran the washroomslt lunch room, and drinking foun-
tains wi 11 drain to a packaged sewage treatmen-t; plant and then will
be. discharged into the surge chamber via sewage lift pumps.
12-68
I
I
1~7:"
,, .-I
I
I
I .......
I
I
I "'···
I
I
I
I -~
I
I
I
. ""
I '' '
••
I
I
I'
'•
I
'>t<"""''
I
I
~....,
I ·~ ....•. :
' I
I
l
I
l
I
I.
' I
I
I
(vi) Heating, Ventilation and Cooling
The heating and ventilation_ system far the underground· power plant
will be designed primarily to maintain suitable temperatures for
equipment operation and to provide a safe and -comfortable atmosphere
for operating and .maintenance personne.l.. Air wf11 be drawn into the
power facilities through one or mote shafts or tunnels, circulated
throughout the power plant, and discharged from the power plant
through other shafts and tunnels. For preliminary design purposes
-it has been asstJned that air wi11 be drawn down the access and the
cable shafts,. and discharged out through the access tunnel; however,
-the actual arrangement will depend upon the final design.
The power plant will be located in mass rock which has a constant
year around temperature of about 40•F.. Considering heat given off
from the generators and other equipment, the primary requirement
will be for air cooling.. Initially, some heating will be required
to offset the heat loss to the rock, but after the first few years
of operation an e{luil ibrium will be reached with a powerhouse rock
surface temperature of about 60 to 7o·F.
Air cooling will be accomplished by providing suitable air changes
incorporating cooling coi 1 s in the air circulation system-Cooling
water from the station service water supply will be circulated
through the cooling. coils.. In winter, some heating may be required
to moderate the temperature of the incoming air into the power
plant. Allowance must be made in the design for the possibility ·
that large quantities of air (up to about 6,000 cfm per unit) may be
required for turbine aeration.
Other factors which· must be considered or incorporated in the design
are:
-To prevent or minimize the circulation of combustion products in
the event of a fire, powerhouse ventilation should be sepm-ate
from transformer gallery ventilation and provision should be made
for i so 1 at i ng the two areas; and
-Suitable air locks will be necessary to preclude adverse chimney
effects in the shafts.
(g) Surface Facilities Mechanical Servi~e Systems
The mechanical. services at the control building on the surface will
include:
- A heating, ventilation, and air conditioning syst1.!m for the control room;
-Domestic water and washroom facilities; and
- A halon type fire protection system for the control rc,!Jm.
12-69
Domestic water will be supp 1 i ed from the powerhouse domestic water system,
with pumps located in the .powerhouse and piping up through the access ·
shaft. St1nitary drainage from the cont.rol building will drain to the
sewage treatment plaat in the powerhouse through piping in the access
tunne].
The standby generator building will have the following services:
- A heating and vent i 1 at ion system;
-A fuel oil system with buried fuel oil storage tanks outside the
bu.ilding, and transfer pumps and a day tank within the building; and
- A fire protection system of the carbon dioxide or halon type.
{h) _[achine Shop Facilities
A machine shop and tool room will be located in the powerhouse service bay
area with sufficient equipment to take care of all normal maintenance work
at the plant, as well as machine shop work for the 1 arger components at
De~il Canyon. For preliminary design purposes, an area of about 1,500
ft (las been allocated for the machine shop and tool room. The actual ·
equipment to be installed in the machine shop will be decided during the
design stages of the project.; however, it will generally include drill
presses·, 1 athes, a hydraulic press, power hacksaw, shaper, and grinders ..
12.18 -Accessory Electrical Equipment
The accessory electrical equipment described in this section includes the
following:
.. Main generator step-up ~5/345 kV transformers;
. Isolated phase bus connecting the generator and transformers;
.. Generator-circuit breakers; ·
• 345 kV oil .. fi lled cables from the transformer terminals to the
.. Control systems of the entire hydro plant complex; and
.. Station service auxiliary AC and DC systems.
switchyard;
Other equipment and systems described include grounding; 1 ight ing system, and
communications.
The main equipment and connections in the power plant are shown in the single
1 tne diagram, Plate 60A. The arrangement of equipment in the powerhouse,
transformer gallery, and cable shafts is shown on Plates 57 through 59.
(a) Selection of Transformers and H.V. Connections
I
I
I . ¥,;;l .
.,_ .. _
~ ..
-'"'
I
I .
.
--..,;,:~.-
1 ·-.. ,~
I
I
~-.. ~·-"··
" •. ·-·
I
I t.,.-
1 .., ....
' .
I ...
,.. .... ~ ... ~~-..._. ... --..,..._\ -
(i) General
Nine single-phase transformers and one spare transformer will be lo-
cated in the transformer gallery. Each bank of three~single-phase
12-70
t
I
I
I
' I
I
I
I ....,_
' I .._,, r
I
t
I-I
·' ,, ; ,.
I ~
I· ......,._.
I ..... _
I
I -
'<
I
I ...,
-·
tran.sformers will be connected to two generators through·generator
circuit breakers by 1solated ph.ase bus located in--individual bus
tunnels~ The HV termtnals of the transformer will be connected to
the 345 kV switchyard by 345 kV single-phase oil-filled cable
installed in 700-foot.-long vertical shafts. There will be two seats
of three single-phase 345 kV oil-filled cables installed in each
cable shaft.. One set wi 11 be maint:ained as a spare three phase
cable circuit in the second cable shaft. These cab1e shafts will
-also contain the control and power cables between the powerhouse and
the surface control roorn, as well as emergency power cables from the
diesel generators at the surface to the underground facirities ..
A nllllber of considerations led to the choice of the above optimum
system of transformation and connections. Different alternative
methods and equipnent designs were also considered. In summary,
these are:
-One transformer per generator vs one transformer for two gener-
ators;
-Underground transformers vs surfa~e transfomcrs;
Direct transform·ation from generator voltage to 345 kV vs inter-
mediate .step transformation to 230 kV or 161 kV~ and then to 345
kV;
-Single-phase vs three-phase transformers for each alternative
method considered; and
-Oil-filled cable vs solid dielectric cable for SF6 gas-insulated
bus.
( i i) Re 1 i abi 1 ity Consider at i_OJ'!!
Reliability considerations will be ~ased on the general reliability
requirements for generation and transmission described in Section 15
regarding the forced outage of a single generator, transformer, bus
or cable in addition to planned or scheduled outages in a single
contingency situation, or a subsequent outage of equipment in the
double contingency si·tuation.. The system should be capable of re-
adjustment after the outage for loading within normal ratings and
for loading within emergency ratings.
The generators will be rated with a 115 percent continuous overload
capability. All main connections and equipment including the trans-
formers, circuit br.eakers, isolated phase bus, and 345 kV cables
will be rated for continuous operation at the 115 percent overload
rat tng of the generators ..
Emergency ratings are different for different items of ·equipment and
emergency periods. It generally varies between 110 to 130 percent
12-71
in s~.m~~er to 120 to 140 percent in winter for a 4 to 12 hour ,period-,
with somewhat higher values for very short (1 hour) emerg~!l~JI .
periods.
(iii) Technical and Economic Considerations
The use of surface transformers connected directly to· the under-
ground generators by i so 1 ated phase bus was ru1 ed out at the outset
due to significantly higher costs and higher losses associated with
generator isolated phase buses. The incremental cost could be de ..
creased if three units were· connected to one-transformer, but such a
compromise is not acceptable due to reliability considerations.
In general; 3-phase transformers are preferred to single-phase
transformers because of their lower overall costs, smaller overall
dimensions ·and smaller underground gallery dimensions.. However,
transport limitations seriously affect the use of the larger size
3-phase transformers, both in dimensions and weight. The following
are the road and rail data avail able:
-Parks and Denali Highways
Maximum load -150,000 lb
Overweight's require special permit~~.
-Railway
Maximum Weight -263,000 lb
Dimension Limits -16 feet high, 10 feet wide
A further check of these design limitations for the selected sizes
of transformers is recofl111ended during the detailed design stage.. A
careful route reconnaissance study is also required.
Single-phase transformers are therefore recoomended for the 6-unit
power plant. The grouped unit arrangement with two generators per
transformer will allow a smaller gallery length, with center-to-
center spacing comparable to the generator spacing. The grouped
unit arrang.ement is the recomnended arrangement. The alternative
with one transformer per generator will requie a gallery about 300
feet longero
One distinct advantage of single-phase transformers is that a spare
transformer can be provided at a fairly low incremental cost ..
The double-step transformation scheme (15/161 KV generator-
transformer, 161 I<V cable and 161/345. KV auto-transformer at the
switchyard) is economica111y competitive with the direct transforma-
tion scheme· (15/345 KV), resulting from a number of tradeoffs:
cost/MVA per transformer is lower; also dimensions, weights and cav-
ern dimensions are lower; but the intermediate-voltage transformer
costs are additional.
12-72
~~-.. _,
,. ·.
I
f,
' I
I
I
I
I
I ""~r
I
'I ~
I :,..../
I
I ,....._
.t
I
'i ·8'
I . ' '' '' ~ .;
·"-·~·· .".k._ ... ,.-..~~,. ....... ·:~~~,.-.._.:v--~~.i&H'"-""""h;·~-b~~~-~a'~''$tl~
' .'-"! , ..
I '.;'
I
I '-r·
I
I;
"-.··}
t
' ,.
I
'-'"
I
I '\.<-'
' I
' I
I
I
(c)
Direct transformation {15/345 KV) is. better from system transient
stability viewpoint since the overall Jmpedance of the generator
unit to the 345 KV bus is lower. Furthermore, it has a better over~·
all rel iallility since there is: no one less voltage level and, there-
fore, less equfpment in the generating nchain" of equipment. This
scheme costs about $2 mill ion less in over.a11 costs compared to the
double .. step transformation scheme ..
The comp.arison between 345 KV oil-filled cab1es and other .345 KV
cable and bus system is made in Section 12.18. The SF6 bus is about
5, to 6 times the cast of the oil-filled cables. It also requires a
larger diameter cable shaft. The oil-filled cable is well proven at
a nliTiber of underground power installations and was therefore sel-
ected for both technical and economic considerations.
Main Transformers
(i) Rating and Characteristics
The nine single-phase transformers (three transformers per group of
two generators) and one spare transformer, will be of the two wind-
ing, oil-immersed, forced-oil water-cooled-{FOWT-type, with rating
and electric characteristics as follows:
Rated capacity:
High voltage winding:
Basin insulation level (BIL)
of H .. V .. winding:
Low voltage winding:
Transformer impedance:
145 MVA
345 I 3 kV~ Grounded Y
1300 kV
15 kV, Delta
15 percent
The temperature rise above air ambient ·temperature of 40~C is 55°C
for the windings for continuous operation at the rated kVA ..
(ii) Construction
The transformers will be of the FOW type with water-cooled heat ex-
changers which remove the heat from the oil circulating through the
windings. A one-third spare cooler capacity will be provided.. The
transformer will be of the forced oil directed type with a design
aimed to achieve minimum d-imensions and weight for shipping pur-
poses. The low voltage terminals will be connected to the isolated
phase bus,. and the high voltage terminals to the 345 kV oil-filled
cable box termination at the transformer.
Lightning arresters will be connected directly to the high voltage
terminals. The transfonner installation in the gallery will be de-
signed to provide the necessary ground and safety clearances from
the· live 345 kV terminals to all nearby equipment and structures.
12-73
(d)
The tank underbase wi 11 be provided with. flanged wheels for trans-
port o.n rails •. The spare single-phase transformer wfll be exactly
identical to the remaining ·nine single-phase transformers .. · It will
.be maintained tn a state of .maximllll readiness, for connect ion in the
shortest practical time to replace any of the main transformers ..
Tha ttansformers wi 11 be ful'ly tested· andi}inspected. in the factory
according to ANSI/NEMA StandardS~ -They will be shipped without. oil
and filled with inert gas for protection. At the site, erection
would be mainly for external fittings such as bushings, 1 ightn ing
ar~esters, heat. exchangers, piping, and electrical connections.
{iii) Fire Protection
Fire walls will separate each single-phase transformer.. Each trans-
former will be provided with fog-spray water fire protection equip-
ment, automatically operated from heat detector.s located on the
transformer.
Generator Isolated Phase Bus.
( i) Ratings and Characteristics
The iso 1 ated phase bus main connections wi 11 be 1 ocated between the
generator, generator circuit breaker, and the transformer.
Tap-off connections will be made to the surge protection and poten-
tial transformer cubi.cle, excitation transformers~. and station ser-
vice transformers. Bus duct ratings are as follows:
Rated current, amps
Short circuit current mcn11entary, Clllps
Short circuit current, S)«'dlletrical, amps
Basic insulation level, kV (BIL)
Generator
Connection
9,000
240,000
150,000
150
Transformer
Connection
18~000
240,000
150~000
150
The bus conductors will be designed for a temperature rise of 65°C
above 40°C ambient temperature.
(ii) Construction
The bus will be of standard self-cooled design with conductor and
tubular enclosure of aluminum. The c~rrent r?ating is such that
either a self-cooled or forced cooled design will be possible. With
a forced cooled design, the size and costs will be lower; however,
· if the forced-cooling plant fails, the bus would be severely derated
to a rating 1 ess than 50 percent of the forced cooling rating .. · The
self-cooled designs are used up· t.o 30,000 amps rated current and are
therefore recommended for this installation where the ratings will
not exceed 18,000 amps.
•.
,, 12-74
' I
I .......
I
I ----~c
I
1 -·., ·'
I "'"'
' I .... ~~
' I .....
' I .... ..
I
1'· ...
...
I
''-"
I
' __ ,-'.
--
I
I
1-
.
I~ ._., .,
' ,,
' t
' I ;...Jf
' I .. ,..
l
'
_ The enclosure will be of welded construction and each bus will -be
'grounded. The constructiQl1 Js_highly reliable; will eliminate
phase-to-ph-ase faults,-neutralize -the magnetic field outside the
enclosure; and provide protection against contamination and moi s-
ture, with consequent minimum maintenance requirements ..
(e) Generator Circuit Breakers
(f)
The generator circuit breakers will be of the enclosed air circuit breaker
design suitable far mounting in line with the generator isolated phase bus
ducts. They are rated as follows:
Rated Current:
Voltage:
Breaking capacity,
symmetrical, amps
9,ooo Amps
23 kV class, 3-phase, 60 Hertz
150,000
The short circuit rating is tentative and will depend on detailed analysis
in the design stage. -
The breakers will be designed and constructed with a high degree of rel ia-
bil ity. The phase spacing of the breakers will be generally the same as
the isolated phase bus duct. The breakers will be mounted on strong foun-
dations on the generator fl oar designed to absorb the reaction forces when
the breaker operates. A separate compressed air plant will be provided for
the high rel i ~bi 1 ity compressed air system requirements of the air circuit
breakers.
345 kV Oil-Filled Cable
( i) General
The recommended 345 kV connection is a 345 kV oil-filled cable. sys-
tem bet·ween the high volta.ge terminals of the transformer and the
surface switchyard .. The cable will be installed in a vertical cable
shaft. Cables fran two transformers will be installed in a single
cable shaft.. ·
This system of 345 kV connection was chosen after a technical and
economic analysis of alternative methods of connection~ including:
-SF6 isolated bus system;
-High pressure oil pipe cable system; and
-Solid dielectric cable system.
The SF6 bus system is considered to be the best alternative to the
oil-filled cable system. Its advantages are a generally better
overall reliability, including a low fire nazard. However, it costs
approximately 5 to 6 times that of th.e oil-filled cable install a-
t ion, and ··requires almost twice. the dianeter cable shaft of the
cable installation. The overall cost di·fference is approximately-
$7,000,000 in direct costs.
(g.)
The oil pipe cable will consist of three conductors contained within
an ci,l-fi'lled steel pipe~ · This system has the highest potential
fire hazard of all the cable systems and is not recommended fo.r high
head vertical cable instal 1 at ions. The solid dielectr·ic (pol}flleric}
c~bles are still under development at the 345 kV to 500 kV voltage
class. ~
. .
It is reconmended that further detailed. study of the oil-filled
cable in comparison with the SF6 bus and other more recent SF6 cable
designs under developnent be undertaken at the design stage ..
By far the greatest number of high voltage, higtT capacity install a,-.
tions utilize oil-filled cables. A formidable experience recor·d·ts
evident for the oil-filled cable installations associated with large
power plants all over· the world. Typical installations include the
525 kV/650 MVA units at Grand Coulee III, the 345 kV/550 MVA units
at Churchill Falls in Canada, the 400 kV/2640 MVA cables at Severn
River crossing in Great Britain, and the 400 kV /2340 MVA cables at
Di"norwic pumped storage plant in Great Britain ..
(i i) Rating and Characteristics
The cable will be rated for a continuou~ maximum current of 800 amps
at 345 kV +5 percent. The -max imiJTl conductor temperature at the max-
imum rating will be 1o•c over a maximum ambient of 35°C. This rat-
ing will correspond to 115 percent of the generator overload rating ..
The normal operating rating of the cable will be 87 percent, with r.
corresponding lower conductor temperature which will improve the
overall performance and lower cable aging over its project operating
life. Depending on the ambient air temperature, a further overload
emergency rating ·of about 10 to 20 percent will be available during
winter conditions~
The cables will be of single-core construction with oit flow through
a central oil duct within the copper conductor. Cables will have an
al uminun sheath_ and PVC over sheath. No cable jointing wi 11 be re-
quired for the 700 to 800 feet length cable installation.
Control Systems
( i) Genera 1
A Susitna Area Control Center will be located at Watana to control
both the Watana and the Devil Canyon power plants as shown in Plate
• The control center will be linked through the supervisory
-sy-s~t-em to the Central Dispatch Control Center at Willow'as described
i~ Section 14.
The supervisory control of the entire Alaska Railbelt system will be
done at the Central Dispatch Center at Willow~ A high level of con-
trol automat·lon with the aid of digital computers will be sought,
12-76
I
I
' ,I
I ......
I
I
-·1 .,-..,...-.
t
"s~.: 0 •
I
' t
' I
' I
' • . w
I
' I
I
I
I
'
'
' I
I
,,
I
I ..
' I
a
' a ,,
but not a complete computerized direct digital control of the ,watana
and Devil Canyon power plants.. Independent operator controlled
loca:l-manual and local-auto operations will still be possible at
Watana and Devil Canym1 power plants for tes-ting/commissioning or
during emergencies. The control system will be designed to perform
the fo llo,wing functions at both power plants:
-Start/stop and loading of units by operator;
-Load-frequency control of units;
-Re~ervoir/water flow control;
-Continuous monitoring and -data logging;
-Alarm ann unci at ion; and
-Man-machine communication through visual display units (VOU) and
console.
In addition, the computer system will be capable of retrieval of
technical data, design criteria, equipment characteristics and oper-
ating limitations, schematic diagrams, and operating/maintenance
records of the unit.
The Susitna Area Control Center will be capable of completely inde-
pendent control of the Central Dispatch Center in case of system
emergencies. Similarly it will be possible to operate the Sus_itna
units in an emergency situation from the Centra 1 Dispatch Center,.
although this should be an unlikely operation considering the size,
complexity, and impact of the Susitna generating plants on the sys-
tem.
The Watana and Devil Canyon plants will be capable of 11 black start"
operation in the event of a complete b 1 ack out or co 11 apse of the
power system. The control systems of the two plants and the Susitna
Area Control Center complex wirJ be supplied by a non-interruptible
power supply.
(ii) Unit Control System
The unit control system will permit the operator to initiate an en-
tire sequence of· actions by pushing one button at the control con-
sole, provided all preliminary plant conditions have been first
checked by the operator, and system security and unit commitment
have been cleared through the central dispatch control supervisor.
Unit control will be designed to:
-Start a unit and synchronize it with the system;
-Load the unit;
-Stop a unit;
-Operate a unit as running spare (runner in air with water blown·
down in turbine and draft tube); and
-Operate as a synchronous condenser (runner in air as above) ..
Unit control will be essentially possible at four different levels
in a hierarchical organization cf the control system:
(iii)
....
-Local control at the machine floor at individual turbine-generator
control boards: (primarily designed for coomissianing and recommis-
sioning of units}. It will be the responsibility bf the operator
for performing individual control operations in~ the correct se-
quence, and monitoring instrumentation during Tocal control opera-
tions. · ·
-Automatic or semi-automatic system for start-up and shut ... down of
generating unit at the local board at the machine floor.
-.Fully automatic system at Susitna Area Control (at Watana) for
Watana and Devil Canyon power plants.. (This will be the normal
Susitna operation.)
-Fully automatic system through supervisory control from Central
Dispatch tenter at Willow. (Abnormal or emergency situations
only).
Computer-Aided Control System
Traditionally, control systems for power plants in general, and
hydro plants in particular, have utilized hard-wired switchboard
type equipment (such as electro-mechanical relays, instruments,
alarm annunciators~ signal lamps, mimic diagrcm and control swit-
ches) for the operation, indication, alarm and control of the power
plant. Such equ·i(lllent was installed both at the plant local control
area on the machine floor as well as in the control room, with a
limited degree of miniaturization of equipment at the control desks
in the control room.
Whfle traditional switchboard type equi1J11ent is still utilized at
the local control level, supplemented with progranmable control sys-
tems at many plants, the design of control and display equipment at
modern central control rooms has been rapidly moving towards
computer-aided or fully computer.acontrolled systems, especially
where remote control operations are contemplated. One of the prob-
lems encountered by utilities is the necessity for operating person-
nel familiar with the conventional control systems to adapt to the
new computer-aided control systems·. In this contect, establishing a
mojern computer-aided control systan in the Alaska Power Authority
electrical system for the Susitna Project complex should not pose
any special problems for the adapt ion and training of -operators.
The computer-aided control system at the Susitna Area Control Center
at Watana will provide for the following:
·-Data acquisition and monitoring of unit (MW, MVAR, speed, gate·
position, temperaturess etc .. );
-Data acquisition and monitoring of reservoir headwater and tail-
water levels;
12-78
I .
'· ','
' I
I
I
I
.I
'
' I .
,,
' t
' I
I
t ,..,..
I
I
.
I
. ' '
···.
,f.
' I
f
I
I
e
a:
·I· ,. '
' t ; .
,
I
I·
I
t
I ., ...
. .
..
.. Data acquisition and monitoring of electrical system voltage and
fre_quency;
-Load-frequency control;
-unit start/stop corytrol;
-Unit 1 o ad in g;
-Plant operation alarm and trip conditions (audible and vi sua 1
alarm on control board, full alarm details on VDU on demand);
-General visual plant operation status on VDU and on giant wall
mimic diagram;
-Data logging, plant operation records;:
-Plant abnormal operation or disturbance automatic recording; and
.. Water management (reservoir control).
The block diagrCJn of the computer-aided control system is shown in
Plate . The supervisory control and telemetering system and
central dispatch center system details are described in Section 14 ..
(iv) Local Control and Relay Boards
Local boards will be provided at the powerhouse floor equipped with
local controls, alarms, and indications for all unit control func-
tions •. These boards will be located near each unit and will be
utilized mainly during testing,. conmiss ion ing, and maintenance of
the turbines and generators. It wi 11 also be uti 1 ized as needed
during emergencies if there is a total failure of the remote or
computer-aided control systems.
The unit electrical protective relays will be mounted on relay
boards, with one board for each generator located near the unit.
Differential protection will be provided for each generator and
transformer. The differential .zones of protection overlap will in-
clude all electrical equipment and connections.. The 345 kV oil-
filled cable. to the surface switchyard will be protected by a p·ilot-
wire differential protection·relay. The overall differential relay
protects the generators. transformers, and 345 kV cable. Sensitive
ground fault stator, protection will be provided for the genetator.
Protection will also be provided for negative phase sequence opera-
tion, loss of excitation, ov.ervoltage, and under frequency. A phase
impedance relay will provide backup protection for the generator.
Other protective relays are shown in Plate ..
(h)
{v) Load .. 'Freguency Control (Automatic ~enet·ation Control)
The load frequency control system will provide remote control of the
output of the generator at Watana and Devil Canyon from the -central
d~ispatch control center through the supervisory and computer-aided
control system at Wat~ana: The basic method .of automatic generation
control (AGC) will use the plant error (differential) signals from
the load dispatch center and will allocate these errors to tne power
plant generators automatically through speed-level motors. Provis-
ion will be made in the control system for the more advanced scheme
of a closed-loop control :;ystem with digital control to control gen-
erator power.
The control system will be designed to take into account the digital
nature of the controller-timed pulses as well as the inherent time
delays caused by the speed-leve1 motor run-up and turbine-generator
time-constants.
The load set-point for the Susitna area generation wi 11 be set at
the Central Dispatch Center.. The sumnated power will be telemetered
from the Susitna Area Control center to the Central Dispatch Center,
from which the required differential plant generation (15 error11
) will
be determined and transmitted by the supervisory system to Susitna
Area Control Center. From this point, the remaining functions for
the automatic ~eneration t.ontrol will be carried out by the plant
supervisory control systems to load the individual generating_ units
at Watana and Devil Canyon ..
The unit will be automatically removed from load-frequency control
for various conditions including failure of supervisory system, unit
controller or computer system, abnormally high plant frequency, unit
shut-down, and de power failure. When the unit is taken off au~o
matic lo.ad-frequency control, it will be returned to manual load and
frequency control by the operator at Watana Control room ..
Station Service Auxiliary AC and DC Systems
( i) Auxi 1 i ary AC System
The station service system will be designed to achieve a reliable
and economic distr-ibution system for the power plant and switchyard,
in order to satisfy the following requirments:
-Station service power at 480 volts will be obtained fran two 2,000
kVA auxiliary transformers connected directly to the generator
circuit breaker outgoing leads of Units 1 and 3;
-Surface auxiliary power at 34.5 kV will be supplied by two sep-
arate 7 ~5/10 MVA transformers connected to the generator leads of
Units l and 3;
12-80
' t
·I
' 11 ,,
'
' a
I ~
I.
' ,-
t
t
8
I
' I
' t
~·'
I
, ' ,,
';
' ,,
'\
J
I
' t
' t
I
I
. ,
I·
... Station service power will be maintained even when all the units
are shut down and the generator circuit breakers are open;
-100 percent standby transformer capacity will be avail able;
- A spare auxiliary transformer wi1l be maintained, co·nnected to
Unit 5; and
... 11 81 ack start" capability will be provided for the power plant in
the event of total failure of the auxiliary supply system, 500 kW
emergency diesel generators will be automatically started up to
supply the power plant and switchyard with auxiliary power to the
essential services to enable startup of the generators ..
The main ac auxiliary switchboard will be provided with two bus sec-
tions separ·ated by bus-tie circuit breakers. Under normal operating
conditions, the station-service load is divided and connected to
each of the two end incoming transformers. In the event of failure
of one end supply, the tie breakers will close automatically.. If
both end supplies fai 1, the energency diesel generator will be auto-
matically connected to the station service bus.
Each unit will be provided with a unit ~uxil iary board supplied by
separate feeders from the two bus sections of the main switchboard
interlocked to prevent pa.ra.llel operation. Separate ac switchboards
will furnish the auxiliary power to essential and general services
in the power plant.
The unit auxiliary board will supply the auxi1 iaries necessary for
starting, running, and stopping the generator...:turbine unitJ These
supplies will include those to the governor and oil pressute system,
bearing oil pumps, cooling pumps and fans, generator circuit break ...
er, excitation system, and miscellaneous pumps and devices cannected
with unit operation ..
The station essential service supplies will include powerhouse sump
pumps, drainage pumps, compressors for circuit breakers~ station air
and generator brakes, de battery chargers, control and metering de-
vices, communications, fire protection pumps, and other miscellan-
eous. essential power requirements.
The station general supplies w'll1 include powerhouse. lighting, heat-
ing~ ventil xting and air-conditioninp, elevators, cranes, machine
shop and trJols, and othar miscellaneous pumps and general requi$'e-
men~s ..
The 34.5 kv supply to the surface fac'ilities will be distributed
from a 34'e 5 kV ~:~itchboard located in the surface contt~,l and admi n-
istration bu11d 1 r.~. Power supplies to the switchyard power intake:t
and spi 11way as well as the· 1 ight ing systems for the access roads
and tunnels will be obtai.ned from the 34.5 kV switchboard •
12-81
The unit auxiliary board wfll supply the auxiliaries necessary for
,. starting, running, and stopping the generator-turbine unit. These
sup·p 1 ies wi 11 include those to·. the governor a.nd ail pressure system,
bearing oil_ pumps, cooling water pumps -and fans, generator circuit
breaker~ excitation ·system, and miscellaneous pumps and devices con•
nected with unit operation.
The station essential service supplies will include powerhouse sump
pumps, drainage pumps, compressors for circuit· breaker, air and gen-
erator brakes, de batte,ry chargers, control and metering devices,
conununications, fire protection pumps, and other miscellaneous
essential power requirements.
The station general supplies will include powerhouse lighting, heat-
ing, ventilating and air-conditioning, elevators, cranes, machine
shop and tools, and other miscellaneous pumps and general require-
msnts.
The 34.5 kV supply to the surface facilities will be distributed
from a 34.5 kV switchboard located in the surfa.ce ~ontrol and admin-
istration building.. Power supplies to the swi~~J~,;r.ard power intake~
· ar..d spillway as well as the lighting systems for the access roads
and tunnels '#ill be obtained from the 34.5 kV switcnboard~
The two 2000 kVA, 15000/480 volt. stations service transformers and
thle spare transformer will be of the 3-phases dry-type, sealed gas-
filled design. The two 7.5/10 MVA, 15/34.5 kV transformers will be
of the 3-phase oil-immersed OA/FA type. -
Emergency diesel. generators, each rated 500 kW, will separately sup-
ply the 480 volt and 34.5 kV auxil ia.:--y switchboards during emergen-
cies. Both diesel generators will be located ·;n the surface control
building.
An uninteruptible high security.power supply will be provided for
the computer control system.
(ii) DC Auxiliary st·ation Service System
The de auxiliary system will supply the protective relaying, super-
visory, alarm, contr.ol, tripping and indication circuit in the power
plant. The generator"' static excitation system will be started with
. "flashing•i power from the de battery. It will also supply the
emergency 1 ighting system at critical plant 'locations.
Separate duplicate lead-acid batteries for 125 volt de will be pro-
vided in the powerhouse. The 48 volt battery supply for the super~
visory and .computer aided control system and microwave communica-
tions will be located in the surface control building.
12-82
I
I
I
t
,I
I
I
I
I -' 'I
' I
'
' I
I
'
'
' .I
' I
I
I
t
I· ,.
' t
t
' I
I
t
I
I
The main battery system will be supplied by double charging equip-
ment consisting of a full wave rectifier system with regula:ted out-
put va:ltage which normally will supply the continuous de lo~Q io the
system. The battery capacity will be suitable for an emergency
loading based on a fai 1 ure of ac station service 1 ast i ng 5 hours ..
(iii) .:J31ack Start 11 Capabilit,y
The Watana power plant will have a built-in capability of starting
up a completely blacked-out power system in a very short time. Only
a, few basic requirements will have to be satisfied:
-Sufficient water will ·be available in the reservoir for the mini-
mum generation required for "black start" operation;
-The governor oil system will have sufficient stored energy capable
of operating the turbine wicket gates to full open position;
-The generators wille be equipped with static exciters capable of
being flash-started from the stat ion battery system
-De control power will be available for the startup circuits.
-The above described emergency power requirements will not exceed
about 200 kW for one unit and wi 11 be easily supplied from the emer-
gency dies~l generator. With the startup of a single unit~ the com-
plete power plant and switchyard auxiliary power will be immediately
avail able, enabling a11 the units in the power plant to be started
up sequentially within the hour.
(i) Grounding System
The power plant grounding system will cons it of one mat under the power
plant, one mat under the transformer gallery, risers~ and connection ground
wires. Grounding grids will also be included in each powerhouse floor ..
The power plant grounding system will be co~nected to the swi tchyard
grounding system by three 500 MCM copper ground conductors to minimize the
overall resistance to ground.. The grounding system will be designed to
provide. a ground resistance of 1 ohm or lower.: All exposed metal part and
neutral connections of generators and transformers will be ~onnected to the
grounding system for the purpose of protecting personnel and equipment. from
injury or damage. ·
(j) Lighting System
The lighting system in the powerhouse will be supplied from 480/208-120
volts lighting transformers connected to the general ac auxiliary station
service system. The lighting system will be all fluorescent-and incandes-
cent fixtures operating on 120 volts and all outdoor type high pressure
sodium fixtures operating on 208 volts. The lighting level varies gen-
erally from 20 to 50 foot candles depending upon the powerhouse area; the
\ ...... ·--· . \)' .
I
, I
I
'
(k)
--.(1}-
·nigher levels will be.at control areas. Adequate illumination will be
provided on vertical switchboards with local 1 ighting canopies.
An emergency lighting system will be provided at. the power plant and at the
control room at all critical operating ·locations with an illumination level
of 2 f\lOt candles. The emergency 1 ighting system will operate from a sep-
arate 120 volt ac circuit whic.h, by means of automatic transfer switches,
will be automatically connected to the 125 volt de system upon failure of
the a~ system. ·
Co111nun i cat ions
The power plant will be furnished with an internal communications system,
including an automatic telephone switchboard system. A communication sys-
tem will be provided at a11· powerhouse floors and galleries, transformer
gallery, access tunnels and cable shafts, and structures at the intake,
draft tube chamber, spillway, and darn.
The convnunicat ions system for the central dispatch control system, tel e-
rnetering, supervisory and protective relaying system is described in
Section 15 ..
-lnsul at ion Coordination and Lightning and Switching Surge Prot<iction
I
' I .
t
I
I
I
I
t
The electrical insulation and protective devices will be sele~ted and co-·--J
ordinated to provide a safe margin of insulation strength above the maximum ·
abnormal. voltages permitted during 1 ightning, switching, and shOl .. t-circuit
surges. The 1300 kV basic insulation level (BIL) specified for the trans-
former and other BIL values stated for the electrical equipment and connec-
tions are tentative and are subject to detailed study in the design stage
of the project.
In principle, lightning arresters will be mounted on or adjacent to all
major electrical equipnent having wound-type internal construction, and
will be provided at the generator 15 kV terminals and the main transformer
345 kV terminals. ·
12.19 -Switchyard Structures and Equipment
TO FOLLOW
12.20 -Project Lands
Project 1 ands acquired for the project wi 11 be the minimum necessary to con-
struct access and site facilities, construct permanent facilities, to clear the
reservoir, and to operate the project.
,,
' I
' I
I
' t
I
I
I
t
I
I
I
I
t
t
I
'
' I
I
I
t
I
I
I
Appendix C contains 1and status backgrou~d information relative to the susitna
Project, together with an inventory or private and public lands required for the
project. A 1 arge amount of public ·1 and in the Watana· aret\ is managed by the
Bureau of L.and Management. There are large blocks of private Native Village
Cor_poration Lands along the riv-er .. · Othe'!' private holdings consist of ·widely
scattered remote parcels. The state har$ sel'ected much of the federal 1 and in
this area and is expected to receive a patent.
1
;::·· ... ,.... .
. . . ·.
(. . ' . . . . . . . . . . . '· . 12-85 ' ~· !f<t:!t~~.,,g,...,..:;~>~•,;,,....,..,".~<'""'-.. •'h''""''"'"""""'-•""r~~-,,~,. •o0 "o•·'"'•'•';.~ .... ~~<.,_,;:<"~~·><•A•~';;;>;....d';_~'•.,7...., ........ "•'=''"~~· ....... "':',....._,,~....,..,,..,=o.• ..... >•••...-.·•• ...... ~'~.::...· '"""""'"'' .... ~-~•·•.,.,...'"'""' ...... ' .. '"""'""""""''' ..... '"'•• ..... "'"' ....... ., ..... '-,..........,, ..... ._ ....... _.,, .... , ........ -····;;i;;,i•~~-· ...........,...........,_....,_...._ .... ·.-.··-.,;,.<··~--· ··-·----~-·-·••··iiilli·"-··iiiiiiiioii•••·•~•'
Calertdar Year
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
TABLE 1Z.1: WATANA P£AK WORK fORCE AND CAMP/VILLAGE
DESIGN POPULATitlN .
Yearlx ·.Pea)(. Force _£_amp/Village Design
910 1000
1360 1500
4005 4400
5635 6200
5635 6200
5635 6200
5635 6200
4000 4000
2000 2ZOO
1090 a 1200
270 300
I
I
I
:I
I
I
I
' I
I
t
I
f
I
e
I
I
I
I
I
' I
I
I·
I
I
I
I
' I
I
t
I
I
I
I
I·
I
TABLE 12.2: ROCKF'Ilt. AND EARTH DAMS IN EXCESS OF 500 FEET
Dam -
Ragun
.Nurek
Watana
Tehri ·
Kishaw
Sulak
Mica
Patia.
Chicoasen
Chivor
Oroville
Esmeradla
Sayansk
keban
Altinkaya
New Melones
Don Pedro
Swift
Portage fobuntairc
New Bullarda Bar
Dartmouth
Okoy
Ayvacik
Takase
Hasan Ugurlu
Nader Shah
Gura Apelor Retezat
Hagar in
Charvak
Boruca
Kremcu~ta
Trinity
Thomson
Talbingo
Tokujama
LaG£-ande No9 2
Palo Quemado
Grand Maison
Sao reli:K
F'ierze
Cougar
Yacambu
Emborcacao
f''instertal
Cumberland
Canales
Narmata
Goesi::heneralp
Salrajina
Gepatach
Foz do Are1.a
Tedorigawe
Carter
Country
·USSR
USSR
USA
India
Ir.dia
-
I.ESR
Canada
Colombia
Mexico
Colombia
USA
Colombia
USSR
Turkey
Turkey
USA
USA
USA
Canada
USA
Australia
Turkey
Turkey
Japan
Turkey
Iran
Romania
Jordan
USSR
Costa Rica
Gi:eece
USA
Australia
Austr.alif!
Japan
Canada
S. America
France
Brazil
Albania
USA
Venezuela
Brazil
Austria
Australia
Spain
Japan
Switzerland
Colombia
Austria
Brazil
Jaoan
USA
reet -
1,066
1,040
885
856
830
802
794.
787
787
778
771
754
738
679
640
626
614
610
600
590
590
590
587
577
574
574
568
561
5':·1
548
541
538
53{}
530
528
525
525
525
525
519
5'19
519
519
519
510
510
508
508
505
503
503
503
500
TABI,.E 12.3: SUMMARY OF' ~SlGN OATA FOR LARGE EHBANIO£NT DAMS
IN SEISMICALLY ACTIVE AREAS
·--~---""""'
.. ·-' •-"'+----... "' . .......,., treat Ratio of .
Height Width Corel· Width
-::·~-~. "
.. Dae Feet FreebOard Feet . to J)am Height
Watana (U) 885 25* 35 a .. so
Mica (C) 794 26 111 0.45
Chicoosen {M) 787 33 az 0 .. 42
Oroville (U) 771 22 30 0.34
·Don Pedro (U) 614 -
Ayvacik (T) 587 17 50 0.34
lakase (J) 577 17 46 0.40
Tedorigawa (J) 503 13 40 0.31
Netzahualcoyotl (M) 453 18 50 0.43
Iweya (J) 413 62 33 0.33
Kazurya (J) 413 39
Narakura (J) 4'10 16 y:} 0.56
Pyramid (J) 400 36
T amahara ( J) 380 13 39 0 .. 43
Seta (J) 364 20 36 0.29 .. -.. -~-"" ----..
* Watana freeboard -normal maximum operat~ons level to nominal crest
(additional heicjlt allowed for seismic slumping)
Up at reM
.. Slope
2.4
2.25
2.2
2~6
2.4
2.5
2.6
2 .. 6
2.0
2.5
2.6
2.7
2.5
2.7
2.5
I
t
DownstreBI! I .. Slope
' ..
2.0 I 2.0
z.o
2.0 I
2.1
1.8 I. 2.1
1.85 I 2.0
2.0 I 1.8
2.7 I 2.0
2.2
2 .. 0 __ ,
I
I
I
I
I
I
t
·f
~
;I
I
I
I
·I
I
I
I
I
'I
I
I
I
I
I
I
t
I
I
,..,
Narr.e
Watana
Mica
Ciiicoasen
Oroville
Ayvacik
Tokase
falo Quemado
Tedorigawa
El Infiernilla
larbela
Netzahualcoyotl
Mangla
Oerbendi Khan
Tseng~en
Pueblo Viejo
Be as
Alicura
Ramganga
Iwaya
Narakura
Shimokotori
Bao
Tamahara
Seta
Guri
Legend
Earthquake:
H -High
M -Medil.lll
L-Low
lUi-:1 UJ.Mt:.N::J
Crest
Seismic Height Length
Countrr Activity ( ft) (ft)
USA H 885
Canada l 794 2,600
Mexico H 787 1,640
USA L-M 771 5,600
Turkey M-H 587 1,400
Japan H-H 577 1,200
S. Americ~ H 525 1,215
Japan M-H 503 1,380
~texico H 486 1,100
Pakistan H-H 469 9,000
Mexico H 453 1,570
Pakistan H 453 3,400
Iraq H 443 1,460
Tah.an H 436 1,440 c. Arne ric~ H 436 820 India M-H 435 6,400
Argentina H 426 2,620
India M-H 413 -Japan H 413 1,200
Japan -I H-H 410 820
Japan M-H 390 915
C., America H 388 1,312
Japan H 380 2,000
Japan H 364 1,120
Venezuela l 361 1,970
"
Impervious Core:
CV -Central vertical
C -Central very slightly sloped
5 -Sloping
-~
Free,..
board
(ft}
25
26
33
22
17
17
26
13
25
1.8
18
32
33
33
49
30
16
22
62
16
13
24
13
13
20
TABLE' 12.4: DAMS IN SEISMIC AREAS
Crest
Width Core
(ft) Type
35 cv
111 s
82 CY
80 5
50 c
46 cv
40 cv
40 cv
40 cv
40 s
50 cv
41 s
56 cv
:n cv
43 cv
45 cv
39
I
cv
39 cv
33 c
<· cv 39
36 c
26 cv
39 cv
36 c
36 cv
foundation:
R -Rock
A -Alluvium
D -Downstream
U -Upstream
-· .J
t:ol:'e
Width Ratio Core
at Base Width to
(ft) Dam Height
440 0.50
360 0.45
330 0.42
263 0.34
197 0.34
230 0.40
295 0.56
157 0.31
164 0.34
262 0.56
197 Oe43
230 0.51
330 0.74
410 0.94
157 0.36
'!31 0.30
275 0.65
197 0.48
138 0.33
230 0.56
151 0.39
184 0.47
164 0.43
105 0.29
180 0.50
CG . .,. Consolidation grouting
~B -Concrete block over rock
flt't:.I'!!!~IJ~ ~~Ht.
Slopes Unified
of Core Classi-
Zone fication
0.25:1 SM --0.15;1 Cl -GC -CL
0.15:1 ~·
0.25:1 Ml
0.15:1 -0.15:1 CL -GW/SH
0.15:1 ML/HH -CL o. 3:1 CH/CL
0.4:1 SM/GM
0.15:1 CL
0.1:1 CL
0.3:1 CL
0.2:=1 CL --
0.2:1 ---0.2:1 CL ----0.2:1 Hl
fJ.Ll ~I'(:;) t I,UNU/Jal .\ L}N
liquid Plastic U/S Thick-0/S Thick-treat-Under
Lim!t. Limit ness {ft) ness (ft) Type ment Shells
2) 8 60 60 R :i:G R/UO ----R CG A/UD
40 20 25 25 CB CG A/UO ----. --CB ---50 50 R CB R/UD ---50 R -A/UD
33 7.5 1J 13 R CG R/UD --26 26 R -A/UD
49 25 8 a R CG A IUD --_, -A -A/UD
50 20 -13 R CG A/U ---13 R CG A/UD
50 26 20 30 R -R/UD
22 8 --R CG R/UD
41 19 23 23 R CG R/UO
30 12 20 20 R -A/UO
35 15 10 10 R -A/UD --98 79 R -A/UD --33 65 R -R/UD --20 49 R -A/UO --39 39 --R/UO
40 20 20 20 R -CB & CG --79 79 R CG R/UD --39 39 R -R/UD
50 15 5 6.5 R CG R/UO
0
G_-
J
I
I
I
I
I
I
I
I
I
·I
I
I
I
I
I
· · CColumn
c
D
E
and
G
H
I
and
J
TABLE 12.5: GENERALIZED SURF'IC!AL STRATIGRAPHIC .COLUMN AREA "D"·ANO.RELICT CHANNEL
Unit -
SUrficial
Alluvi\111 &
Fluvial
Deposit$
Outwash
Till/Waterlain
Till
Alluvi\111
Till
Alluviun
Estimated
Thickness
o-s•
0-18'
12' average
0-15'
0-35'
15' average
2-50'
12' average
0-40'
>10' to 65"
20' average
to 160'
.Description
Boulders, organic silts .and sands.
Silty sand with. sane gravel and cobbles.
occasionally. Usually brown although becomes
gray in limited areas. Thickest in northern
portions of area,.thickening southward, often
absent near Suaitna River.
Sand with some silt, occasional gravel.
Generally brown, found only along course of
limited drainage channels formed in o~twash
"E". Generally sorted.
.
Sand, silt, gravel and cobbles, pertly sorted,
with fragment~:~ sub-angular to r:ounded. Silt
and sand lenses often present. BroNn to gray
brown with a cobble/boulder zor1e often present
at the basa of lhit 11 F". Contac.~t bebeen ~£"
and "F"'' is often poorly defined.
Clayey, silty sand, usuai.J.y gray, often
plastic. Contains cobble\$ and .gravel in many
areas~ Occasionally present as a lacusttine
deposit showing l811inatior~s. &tid/or varves ..
Generally a till deposited th~~ugh or near
stand!ng water.
Sand, silt, gravel, partly to well sorted.
Often absl!nt between Units "!.. and "G". \.bit
represents period of meltirg pr,oduci~
alluvium/outwash. between these .'deposits.
Appears as narrow bands representing channel
fillings. Thickest in western portion t1f' the
area.
Poorly sorted sand, silt; gravel and cobbles,
occasionally with clay. Generally gray to ·gray
brown. Continuity uncertain due to lack of
information at depth. Silt o~ sand layer
2 inches -6 inches thi {·1< often found in
cen.ter of Unit "1". Ba~~ unit on top of
bedrock, except in buried channel. Contact
between "111 and "J" often poorly defined.
-,
Sand, gravel, cobbles, boulders, few fines,
permeable. Found only in bottom of buried
channel. Top at 292 feet extending to rocl<. at
454 feet.
Note: Letters used to define units are arbitrarY-and were used for .
correlation purposes. Two letters may define parts of the same unit.
TABLE 12~6: RING FOLLtr.fit:R GATES
".
PROJECT LOCATION SIZE (IN.)
(1) New ~lanes i California 96
{2) Ne~ Melones California 72
(3) Portage Mountain Canada 84
(4) Hungry House Montana 96
(5) Yellowtail Dam Montana 84
(6) Trinity Dam California 84
(7) ~and Coulee · Washington 102
"
(8) Glen Canyon Colorado 96
(9) Green Mountain Colorado 102
*Maximum static head; maximum oper~cing head -250 feet.
"'· .. •: 'i HE' , : . ~·~T)
~~·'
607
.591
550
495
470
450
354*
337
261
_l~H Uf -c
INITIAL OPERATION
1979
.1979
1967
1952
1967
1962
1940
1965
1943
I
I
I ,-
I
' I
I
I-
I
I
I
I
I
I·
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
TABLE 1Z. 7: PRELIMINARY UNIT DATA
1 -.GENERAL DATA
4\iumber of Units ................. ~-. • .• • • • • • • • • •. • • • • 6
Nominal Unit Output •••••• " ~> •••.• • • ., ............... .
Headwater Levels:
ncrl'nlll maximum ............................ ., ••••
__,_ mini~~~t~~. • •. ~ ........................................ D
Tailwater levels:
minirnucn ••••••••••••••••••••••• .. ••••••••••••••
normal ••• ·• •••••• ~ •••••••• o .••••.••.•• o .e ••• ·• •••.••
maxifM.Iin ·~••••••••••••••~•••-••••••••••o•••••
2 TURBINE DATA
17.0 MW
El •. 2185
El. 2045
El. 145Z
El. 1459
El. 1465
Type · ...................................... ~......... V~ttical Francis
0 Rated Net Head • • • • • • • • • • • • • • • • • • • • • • .. • • • • • • • .. • • • 680 t ·~et
Maxinalm Head ••••••••••o•••••••••••••••••o•~••••• 724 fee~·
Minimum Head •••••••••••••••••••••••••••••••••••• 584 f.:tet
Full Gate Output:
at rated head •••••••••••••••o•••••••••• .. ••••• 250~000 hp
at maximun head •••••••••••••••••••••••••••••• 275,000 hp
--at minimum head .......................... ·• • • • • 200, 000 hp
Best Gate Output •••••••••••••••••••••••••••••••• 85 percent full
-Full Gate Dischar~e at Rated Head ••••••••••••••• 3560 cfs
Speed •••••••••••.•••••• ~. :, •• ~ •••••••• o ,, •• c.,.. . • • • • ~25 rpm
Specific Speed ................................... 32.4
Runner llischarge Diameter ·~····· .. •••• .......... ., •• 132 in
Runaway Speed •••••••• f) • • • • • .. • • • • • •• • • • .. • • • • • • • • • • 385 rpm
Center line Distributor ···············~·········· El. 1422
Cavitation Coefficient (sigma} •••••••••••••••••~ 0.081
3 -GENERATOR DATA
Type ••••• fll , • ··• ••••••••• : •........................... Vertical Modified Unbrella
Rated Output •••••••••••••••o•••••••••••••••••••• 190 MVA
Power Factor •••••••••••••••••••~••o••••••••••a~• 0.90
\'oltagf.! .................. ·••••••••••••••••u••,.••u.;•4' 15 k\'
-Syncht\'mcus Speed ,., .......................... e.. . .. 225 rpm
In~rtia ~.stant (H)-~t , ••• ,. • • .. • • • • • • • • • • • • •• • • ... • 3.5 MW/sec/MVA
Flywheel Effect (WR2)'1P .. • • • • • • • ..... • • • • • • • • • • • • • •• 5Z x 106 lb-ft2
Heav.iest l.if't •••••••••••&•••••••••••o••••••••••• 770,000 lb
*Including turhitle
J
...
';·.1
I
..
TABLE 12.B: ASSUMED PROPERTIES ~~OR STATIC ANALYSES or .WATANA. DAM
-~: ... R'i£erJ.ai : R Rur RF" R5 .~. ito ,s. n m.
-·-
CORE:
--Soft(1) 140 200 300 .. a
--Stiff(2) 140 700 800 .35
TRANSIT!ON(3) 145 1300 1500 .4
SHELLS (4) 145 1800 zooo .4
where:
= Unit weight, pcf
K = ~ulua number, ksf
Kur = Elastic unloading modulus nunber, ksf
n· ~ Modulus exponent
Rf = Failure ratio
Kb :Bulk modulus number, ksf
rn ~· Bulk modulus exponent
C = Cohesian, psf = Friction angle, degrees
.6 60 .a .a 280 • .2
.72. 900 .22
.67 1300 .16
= Decrease in friction angle perlog cycle increase in 3' degrees
Ko = Esrth pressure coefficient
Q_ 35 Q. .43
0 JS 0 .43
0 3.5 6 .43
0 JS 6 .43
Note: Value~ taken from ~mean et al., 1980, "Strength, Stress-Strain and Bulk Modulus
Parameters for Finite Element Analyses of Stress and Movements in Soil Masses," Report
No. UCS/GT/SD-01, University of California, Berkeley.
( 1 ) Mica Creek Dam Core, 2 pe.rcen\; wet of opt inu.Jm
(2) Mica Creek Da11 Core, 2 percent dry of optimum
(3) Orovi!le Dam silty sandy gravel
(4) Oroville Dam Shell -Amphibolite gravel e
I
I
I
I
·I
I -
I
I
I
I
I
I
I
I
I
I
I
I
I
I
•••
I.
I
I
I ,,
I
I
I
I
I
I
I
I
I
I
I
TABLE 12 •. 9:. WATANA OAM-CR£ST EL£VATION AND.JP.EE80ARD
1 in$0 Year
River Inflow Storm -·
Normal max~ reservoir
elevation 2185 Storra surchargo 6
Still water elevation 2191
. Wave runup allowance 6)
Dry freeboard allowance 3)
Elevation top of core 2200
Roadway ovGr cere 3
Minimum crest elevation 2203
Governing elevation f'or cres~ of main dam
Highest still water level to be 2 feet above
fuse plug pilot channel
Sill of pilot channel in fuee plug
1 In 10,000 Probable
Year Storm Maximum F'lood
2185 2.185
8 . 17
2193 2202
6 NIL
2199 2202
J 3
2202 2205
2205
2200
Note: The above· elevations do not include allowances for static. settlement and
seismic slumping.
TABLE 12.10: RECENT HIGH HEAD fRANCIS TURBINES
I
I',
I
I
I
I
I
I
I
I
I
I
I
I
•• ,,
I
,_,
I~
;, --- --· --· -... ----·-~ .. -·--
.
l
.. 0
. ,
.......
~
4p -~-' -
IC i ' •
tf!
u -1&1
C!»
~ c
~ % u ., -Q
·. . -..
l470' 1480 1480 1500 tOlO tD20
HEAOWAT~R f='-t:VA"ON (FT .. )
WATANA DlVEit$10N -
' .
TOTAL. FACI~ff'Y RATlNG CURVE
--
''
·; ;?'
A ,,,
"'"':•
'.
< -
. ...
-'
-·---
~
1j
l
f
f ~
j'.
j
(
r
i ,
j
1
....
FIGUR~ 12.1
')
I
I
01
I
I
I
1:
I
I
I
I
I
-1
I
I-
I
I
-0:1
2300
2200
2100
~ 2000
l.L -
z
0
1-
<[ > LLJ
...J 1900 w
·"
" ·t -
-·
\ \
\ ·~ ~ ~ .......
--
·-
.· 2.200
"
2100
~, 2000
\
\ v·WEl YEAR -
\
\
\ 1900
"
~ .
\
.\
WET Y~
\
-
AR
2300
2JOO
2000 ,-
/
1900
.·~.·
l ,.
i
' ' --·
f
l
' ".j. .
--
' ~\
\ -· -·
" ~
i '
' r-WET ~ EAR
' ;
' ;
cr:
0 > n:
LLJ
(I)
"' Q!_ 1800
AVE I
MON
lr AGE
HLY_.~
~ ~-
\\
\.\ 1800 \ 1800
\ ' ~ 1\
\ !
' / 1\ \
\ .
\
' ' \ / 1\
' I
1700 1700 1700
\ --
\
' l
I
\
~~ \ 1600 1600 1600
' \
/ l~
N -
~
'
0
A J
1500 1500 150
J F M A M "' J A S 0 N D J F M A M J J A M J J A S 0 N D J F M A M J J A S 0. N 0 J F M M J
2 4 6 8 10 12 14 16 18 20 2 4 6 8 10 12 14 16 2. 4 6 8 10
MONTHS MONTHS MONTHS -·
JANUARY START MAY START SEPTEMBER START
WATANA -RESERVOIR
EMERGENCY . DRAWDOWN
\
.
A
12
\
--w· 1\
.I
!
"
'
I
i
£
I 4 ,,
~
i
I t
I
i
·~ 4
--
' 1
i' i
1
~
s 0 -N
14
'
'
D
16.
~. ~-
.-I .
••
1:
··I:
I'
···.··1·· . . ' ~·,
I
1:
.. · .. ·1:
I.
I
I.
I.
I.
I
I
-....1
(/)
:E·· -
z
0
t-
ct > w
..J w
a::
0 > a:
laJ en w a:
t-------UN!T
2000r-----~-----1·--~~------~----------~~----~~----+-----------~--~--~~--~--~--~--~
--. 2 NITS
-3 U ITS
1900
'">
1800
1700 0
1400 ~-----~~--------L-----------L-----. ......... ._. ___ _,__ _____ .-..~~.....-_____ ___., ______ ~-+'
0 5000 10000 15000 2.0000 2.5000 30000 35000
. CAPAC,T.Y (CFS} ' .
'
I·
~· -------- ----'"-·--l_ .
t
.
-
I t
•
i 10 .
I
... z • -•o Ill • ~-
; 10 -...
~ 40
• •
l t
\ ~-
~ ... ' u
~ ... to .. "'!';
ii
I
! I
I
100 f o.a O·QI .,
j
. WATANA
COMPARISON' b.F-GRAIN Jt~E "' ~
-CURVE$ FQR VARIOUS COqE NATER.IAt.S ~,. .. . '\. · fii .
·----~.
. . 1
·'
' ~-._.
l
I . FIGURE 12~0
~ ·. . .,
~-.
.... ..
-------------------
90
80
-
70
~
IW'\ -
tO
0
i ' • I . a IIJI • ' I' 1~ ~
.. .
. . , .
' .
!c.
.
.
:~
.
... -. • BOli.DERS OJ88LES ·coor••
M91STU8E CONTiNT
l'QUIQ 1-IMlT· ..
P'-ASTICITY U~D~X
.
.. RANGt;
,.,. ,. 9/Q
o-.11
NP-~
I ~ 1/1111 4 It) II ~ . IOIDD ~(I) I'D
~ ~'"' r..r\ ~ r\ ~ . . r• I ,. ' II . ~
~ ,...r-., ~ r\ ~ ~" ~
·l'lllo ~ ~-l'\ ~ ~'\:.~
~ ~ K_'\_~ [\~ ~
~ ~'" I\ I\ ~ "
,-~
.. .
~ [\\ ' ~ ~ ~
""' "' t\ :"\ '\ ~ ~
~ .
' 1\.. ~ ~ ~" ~
' t" ~ ~~ ~""~ -~~~ .. ----.. "'-~ ~" ~"-"'0~~
., ..
•GRWEL
I FIM
AVERAGE
a2.a 0/o
I? .I
H.P
-
p .
~ ~"'"~ " ~ f;.;.
"' ~"~ ~t\ ~
'''"" "'' ' ~-1\'\ •" I\.
1\, ' 't\.
' 0.1
~ SAND
Coone l Midi""' J fine
WATANA.·BORROW AREA Q
GRAIN SIZE QURV~S~ VNITS C,D t Clf)
~
' 0
L'\
K'
f'
.
.
~
~ ~
~ ~~
~ ~~
K' ~' '"' ~
• ~;:
~""'!' . ..
·-·.-.i:.·' .. ..
.
.
.
•.
-·
~
~~
~~ ~ !
''"" 1\1\ ~ ~
~~ " ~ ~ ~ ~
:~ ""~ ~ " ~ ~
o.ot ome
FINES
Silt Slrel ..
·.~· ...
·,., .
"71
i. -.~ -~ :,
~1-
~ Q ~
1\ . ... I)
f ~
f .
! 2Q !
i
\
$ 50 l
i
1:..-
' ~ ~
l
~ 60 ;.
!
'
t ~
~ -j
' i 10 :
:
e -. ~ .
! .. 10 ,.
~
~'\ ~-·
leiiSiztl
I
.t
l
J
i
I
f
t
I.
f
~
'l
-----------------·-----
150
-~
~
~ !Ill>•
10
I
.. I • • ' ' . .
..
. .
..
.
l
BQU..OEftS cota.£1
M01$TURE COtfTENT
LIQUID LIMIT
PLASTtC&TY tNDEX
,.
I J 1111 a ~ UJ..III 4 J ~-tl ~ . 80100 ~Cllt I 0
··' I' .'r' ~ ~ ~~""-" r.~-r--.~ ~ ~ ~
. . ~~~~"',·~~r-."11\~~~~
•· '
'COOflt
ftANGt:
.. Ci-1$0/o
O-t7
NP-1
..
'.
.
~ ~"'"" I" I\ r'\ ~ ~ ~" ~~ .
" ~" ~r-. ' ['., ~ l\: ~-~"'~ ~~
''"' 1\ !" l' ~ "'" ['\._~ ['\."' "" (\I\ r\
~~ ·" l"\ ~ ~"\, ....
0.. "" "' ~ I'.
"'Il ~ ~ ~"' [\., "'"' (\ 1\
l'f\
I~ ~'\: r0..."'" ~ 1\ 1\f\
'"' ~"""' 1\
,,
~
~" I""" ~ "" !'-.
r--~
... D· I
GRAVEL :
I Flnt .. Coarse 1 · Medium
AVERAGE
10.2 °/o
12.8
NP
' ' ' I'
-. ~ -~-~""' ;,.;:~. -·· -~·, -
~ .
[\. :·~ ~
1\ l" R ~
[\. t\ i'\."\ ~' ~
' ~"' ~ ~~ ~"' "" ~ l\ ~' ~" ~"""' ' ~ ~ l'.." ~~ i' 1'\. 1\
l' ~ ~ ~ ~"'~ !"
" 1\
~ ~ ~'\ "'"' "'""' " 1\
~ ~" ~,, I"' I' . ··~ ~"""' r-. 1\
' r-.1\
~!" ,._.
0.1
SANO
I Fine
WATA~A·BORROW AREA P
GRAIN $1ZE, CURV~S-U~ITS ~-,F +ElF
t. .. ·~ L . .I
....
~ ~
1"-[\
I' ~
~ l'
I' [\
.... ~ ~
't'
.
... Q
..
I) ..
...
•·
. ' 20
.
59
..
:.&A
. ..,_ I
n .t
eo I
~
~ . r-r
r0 l'.." .
N K ~ ~-· . '
"'
.
I
['\."\ K" ~'~ ,_
~ ['\." ~~ ~~ .-....
Ill.
~ ~ ~~ 1\1' r"'~ ~ ~
~ ~'\ 1\f\ I' ~ ~ .. ~-
~ I\ I' I"" ~ ~ ~ ~ '-~" ~
o.oa o.aoe
FINES
Slit Sfz•-~.Sll••
.. -
.... -~-. -..;::; ~
FIGURE I_ 2. 7 lil
\ '
·-
80
,•
....... .....-
to
0
- - -.. - - - - - - - - -·---
if
" .......................... .
I • •• • ~ Ill I .-JIII/I • 10 20 44 10100 . lQ:) I'D
• .. ' ~-.-&'!..." -~ I' ' ~" ~ ~ ~'" rt:~"' ~ • . ' i' ~ ' i\, -~ ~ ~
~ ' "' ~ ~'"" ~r-., ['\ "\ ~ ~ ..
.. r-.,~ ~ f'\. i'.. ~
~ ~ ~ 0 ~"' ~"'" ,I\ i'. ~ ~ ~ ~
~ ... , -..,. s ~ ~~~ "' ~ ~ ~ I'.." ~"" ... ' ~
-~ ~~'\..~ """ ' [\. :\ ~ ~" ~'~ 1\ -~
~~ "' ' ['... " 0 L'-.'\ '''" ~ ' ' ['\
' "' ' ~ ~ ~ ~ ~'"' I\ .... .... ~ .
1\,. ' ' ~, ~ " ~ r0 ~" ~"'-."0 ,, ' ~ ~
• . ~ ~ ~ ~' ~'"' 1\,. ' ' ["\
~ ~ K" [\...""~ ~ ' ' ~ ' ~~ ~"-" 1\ ' ' t\
.. '-~ ~""~ I" ~ ~ ' ' '"' .......
1\ ' " ["'\
·~ 1"\ r': ~ ' ' ~~ ' ~ ~
t'l !~ ~ ~ . '
-
.. · . ·-·--· .. D. •. • 0.1 . . .Q~
. GRAVEL ~ . 'SAND 80ll.DER3 · COBILEI. 'Coorq I Fl .. eoar.. J Mldlum I Fine
(
RANGE A~RAGE
MOISTUR' CONT~NT
LIQUID LIM.tT
PLASTICJTV INDEX
t-40°/o ~ . . . ·c.
'..,~. ~~
JW-1$
WATANA,-tiQRROW AR~A D
GRAIN SIZE C\JRV~-. UNIT$ (it F /G -
..
,, ~· -Q
.,
• IQ ' ; ...
I
rr ,. !
' t 2Q
' }
-~ '-. ' 1 !G
~ ~ 11 \.. -~ :\.."'\ ~ ~ -
~ ~ 0...'-. . ~
J '10 " ~ ~"""' ~
~ ~"" ~""" ~ l
~ ~ ~" ~"'~ "., ~ I ~
0 L'\.." '''""' r\ ' "' ~ i
~ ~ ~'\..~ 1\'\ "\ ~ ~ ? 1Q
' i'o.
f\..' ~" '~""" "' '\[',
~ ['\ ~ .
!-
~ ~ ~"' r-., '\!-\ ~ ~ ~ . . :-
' ~ ~'...~ r-..' "'' 0 ~ ~ ' ' 10 ~"' "'' "' ['.....,: ~ ......... =' --~~ .:::~ ~ I'' ,.
""' ' 0.01 00015
.. FINES
SllfSiz• ICkWSiln
-
- --· - --- - - - -·--· - - --' - -
90
80
.
0 1 -a
'i •eo
eo
40
~
~ ....
10
0
I • • ' l
.
.
BOU..DERS ~ES
1 3 Ill I ilt 1/1111
[I' 1 If ·r IT
'
-
• t ·GRAVEL !
;Coan• 1 Flnt
• 10 ~ ~ . IQIQD
' '\ ~ L'.'-' ~ It ' J '. ,.....
-:-~ ~~ Ill..
~ ·r-. ~ ~
1'1~ ~ ~
·-~ I" ~
' " ~ . ~ r--.."\
~~ ~
~ ~
' ~' ~~"'' .I~"~
'"'" ~i'>
'~"~ 1\
.
.
~
• 0.1$ .
SAND
Coaru 'I MI41U~Jt I Fine
WATANA-BORROW AREA . D
GRAIN SIZE CURV£S -UNIT H ~
'
.
tCU I fiG ..
II 0 .
'
't I) ... . ..
' . .
., ... ' 20 ..
·-. '
~ ". ' 30
l
' ,_
~
'·
' ·'
eo
'
; ~
~
1\. '
r-.~ ;
~i\ ~~ .. 10
'
1\ ['\ ~ ...
~ llillllft
~~ ~ l' -,_
.....
~ ~ ~ h.. 10 ~ f\..'~ ~ l\1' " " " o.oe oDD~
FINES
Slit Sfz• ltkWSim
'
"'· ftGURE l2.t
. -------------------
.. u. .................... .
f • ., ~ a 11 II I ~ 1/21/1 .. ·~ 21 ~ IOtQD tO ) 11'0
~ • •r-~ ~ ~ ~ ~"-"" ~"" I' li ' ' ~ II ~ ~ . 100
~ ~ ~"" " r' ["\ ~ " ~
~" I' "' 1'. ~ "" l'." ~ . . ' r--" "' ["\ ~ l'\ ~ ~"" 1'\.
~ ~ " [\.." ~~"""" ~ ~ ""' '
80
""' ~ ~ ~"" ~ ' '" r'O ~ ~"'~ "" ~ ~ ~
~" "t\ 1'\,r-.... ~ [': ~ ~ . ~ t\r-.. t\.. ~ ~ ~ ~~
~"' t'-, ~ l" ~ ~ ~ 50 "'~ ~ 0 ~" ~~
"'Il ~ ~ 10..'" ~ 40 '" !'\."-.~" ~'" 1'\ l' ~
"'Il ~ "'" I" ['\ ~ ~ 30 ~ .::s: f" ~ ~ r-.i\
-I'll~ I' ['\ [\ -. r-f'_ ~
10
. ICIO IQ D O.CJ ill Q.OI
GRAJIEL ; SAND fJOlLOERS ~s. '
'Coors• I FIM COcrM I t.tldlum I Fin•
RANGE AVERAGE
MOISTU~ COt·fTE"T
LJQUIO LIMIT
· PLASTICITY INDEX
t-13 OJq
88.2-39
0.$-1~
l0.3 °/o
29.1
7.5
WATANA-BORROW AR~A ·p
GRAI~ SIZE CURV~$-UNITS. l,J+ll.J
.
''
~
~ ~
"" ~" ~ l'\." ~ ~
........
.~
' .
..
.
.,
~
~"'" ~ .....
l\.."-~ l'f' 1"1111 ~ ~ to.
~"" ,I\. 1\'-~ , ......
~ ~~ ' " ~ .:\ ~' Q.OI
FINES
SUtSiz•
Q
10
' a)
~
., .,
' 50 l
' -~' ;
' '
i eo
~
W\ ·-
:
10
IAn ,--
~ --90 ~J.
-• I
#
J
J
!'\." i'l.."' : QOOIICX)
(Clot SiMI
FIGURE 12.10 .
·'
--~-----~-.~------------;;!
J,f
f • ~,
l
. -
----··---------:--
.u. ................. .
0
19 • '
10100 . -.
\. -""r\"~~,,~,~ ~:-== ::::: .~: ::: ::m:::::m::!~'''r--r-.r--.,i'.' '-'~'','''" ....... ~-~ ... l' I '\.. '' l "-. '' ~ • • 111 • • • • • • ; ; ~ ll.r. : :: :: ::: :::: ::!:::: t:~ "'-"\ ' ~ ~ " ~ ': ' "'-.: '-..''" " -~ f "\. l' ~ ~
I 0.1
• ~\:::::::: : : :: :: :: :. ~ r\.
· '·~ C! ~01 0.0' OJJOI QOOtQJ 10
' -GRAVEL . SAND FINES BOlL~RS (X)8BLE8 Coar11 I Fine Coone I Medium I Fine Silt Slz• .ICtclfSizu
,• 1 .,
.•
/ t
WATANA
REQUIRED GRAIN SIZE CURV~S
0 \
MAIN PA~ t r
I l
----- - - - - - - - - - -----· -•. - -
.u ....... ...,. '"'' ........
I • • • a Ilia ' !It 1/'1.1/1 .. 10 ao '44 110100 ~(I~ .. !1ft
.
T T ! I I" ' ' ' II II I ' ' 'I ~ Q ,
~ ~ .,.,
1
1'-o:o-.. ~8.50Jo L No.4 .. l 0 90
.. r-.... t--...
i
80
,
t 20 ..... ~ l
i
!
......... i I ~ t"oo
.. ~ to-..
70
. ~ ' .
'
... 40
"""
!it) ~ .
"'-, 45°/o L No.'200 eo
' ~ ""
.
40 . l 1--.
• i' :vu
~
30 ' 70 ~
...
j
.
I
l.
. """' --
20 A-
) '"'t"oo ~ -·
' r-... "-' 10 10 :~
0 100 10 I)· I 0.1 .o.oa o.oaa QOOl~
BOll..DERS 'CoesLES GRAVEL SAND FINES
·coors• I Fine Coarse I Medium I Fine Silt Siz• l~SiHI
0
· WI\TANA
COMPOSITE GRAIN §IZE CURVE-BORROW AREA 0 FIGUA~ 12.lt . Iii.
I
I
I
I
I
I
I'
I
I
I
I
I
I
I
I,
I
I
I
·I
145
140
13 = 13
!li.~
5 u
Q. -> ...
0·
12 8.7
12 5
120
2
NOTE:
.
. .
.
.Y
-/ BSOJ.
~·
v 4.20~ ~
~
l
IWI""
v ' f7.5 olc.
f\
MAX DRY -DENS ITY_\ \1
~ 10.1 .
~ .
\
6 8 10 12
WATER CONTENT (0/o)
MATERIAL PASSING 3/-4" SIEVE·
WATANA
MODIFIED PROCTOR COMPACTION
COMPOSITE SAMPLE
BORROW AREAD
'
-.,
FlGURE 12.c.13 li·l· ~~~--------------------------------------------~-------r--
I
I
I
I
I.
·I
I
I
I
I
I
••
;I
I
I
I
I
I
•.• , __
135
130
127.6
-. -:;· 125
.Q, -
-U) z
LLl
Q 121 .. 2
> a:: 120 Q
115
uo
4
NOTE~
. "
.
.
~
J .
VroJ~~
I J· l .--v . \
~ I I ~ 50fe MAX DRY DENSITY
~.0°/• -12.~0~ -
v '1!1
I/ 1\ . .
1\
.
6 10 1·2 14 .
WATER CONTENT (0/0 )
MATERIAL PASSING No. ·4 SIEVE
WATANA
STANDARD PROCTOR COMPACTION
COMPOSITE. SAMPLE
BORROW AREA D
'
-.
I
~ ~· I>
16 18
FIGURE l2.14 •
•• •••
.
. .
.
---------~---------
0
"
80 ~------~--------------~------~--------~~--~~------~------~
60 --I. ...,.,.
!II
fJ)
' Cl)
·&1.1.
0: 40 t-
Cl) .
0::
c(
UJ
X
UJ
. 20
NOTE:
EFFECTIVE STRES'-CONeOt-IDATED
UNDRAINED~ &HEAR TEST, . 4 INCJ-1
DIAMETER SAMPLE$, OPT+ I% AT
16%. MODIFIED PROCTOR G
C0¥~~9TION. . .
40 60 80 100 l20
NORMAL STRESS (psi)
WATANA
CONSOl-IOAlES'tiNDRAINED TRIAXIAL TEST RESULTS
. COMPOSITE SAMPLE-BORROW AREA D
!!-
...
140 l60
c
FIGURE 12 •. l5
;,
.•. ·-. -·~
"I
I
I . . .
I
I
I
I
I
I
I
I
I
I
I
I
~"I
0 -1-
4
a:
Q ·-0
>
.4200
'
.3800 •o =.3761
.3&00 , .
~ . . 'a__
~ l•.
.340<)
.3200
. 3000
.2800
.2600
.2400
~-2200
.2000
.1800
.1100
.1400
OJ
!
~ 'a_
"' .
~· \: ~
. " "' "' ta •t= .. 2345 -~· ~
.
-
4. . ~ -. -...
0.2 0.3 0.5 0.7' t.O 2.0 4.0. 8 10
P"ESSURE ( TS F )
WATANA
CONSOLIDATION TEST-BORROW AREA D
STANDARD PROCTOR COMPACTION-OPT. +2 °/o
20 30 so 70 100
I
1-!'I.GURE 12.16 J11l -'~----~~~----------~--------------------~--~------------.-·~~~-.··~.· \!, .
II.
---
-~·
0
KlO
90
0
160
50
40
30
~ --
10
II
I
·BOU..DERS
• • ' ~"""' ~"'~ '"""' ~'\_~.
~'~ ~'\
~"'"'!' ~~r-
. ·~" ~!' \...~~
\. "" ~~
COBBLE$
. .u.a.at...., ""'~'
I 2 U/1 I·~ 1/IJ/1 • 10 20 40 ., 100 ZOO IPD
~ !'. ."'' N l I I' I ' ~" r"\."'"" ~~ ~ ~ I ' . 1 r
t\ I' ~ ~ ~ ~ ~"""' \ 1"\ ' ~ ~ 1 UPPER ZONE \
.... ~ !' ~ r0 l' ~"' ~ 1';. '" ~ :'\ l". ~ ~ J'
I\ I\ L\ L' ~ k'\.' ~ I' f\. L'. ~ ~ ~" ~ ~
!'~ ~ ~ l'\ ~" ~~ f" ~ ~ ~ ~"' ~~ " ~ .
~I\ 1\ ['\ L\ ~ ~ ~ I "'ii ~ ~ "' ~ ~""~r-.~ I\ I' ~ ~ ~" ~'" r\ l\ K" ~'"'" I'~
" 1\ f'-I'
['\ [\ ~· ~ I'.~ ~'~ " ~' ~'"" I' r-.1\ i'.
i'l' ['\ ~ ~ !:S." ~' "~ l!l'ii ~ ~"'" "" !\I'
I'~" ,, ~ ~ ~ ~ &~ I' I' I' ~ "'"""~ i"'oi'
~ ~ " r--~' r'--[', [\ ~ ~ ~~ I' I' ~ [\ ~ ~ '-" r-1'
r'-1" l'f' ~ r0 t\..~ l~"" ~" f' [', [\ ~ ~ "i; i"
I' ~ I'
~~ I'~ L"-C'\: ['\.."-' ~~ l'r-~ ~ ~ ~"' ~"\ ~"~ I'~
r"l'o ~ """ ~" ~"" "\1' ~I' f' l'\ l"'-~ ~"" ~"" 1'1 ""'~
N ~ r0...~~1' "' r\ ~· ~" ~"~ r-.1' N ~ " I' ['\ !\ to-...
~ I'~" R' ~ K K" ~'"' 1\ I\['. r'\ ~ ~ ~
LOWER ZONE./ I"' roo
"" ~ ~ ~ ~' L':."-~ r-.f" I' ~ t'\ ""' ~· I' "' --~ ~ "' ' \. " ~ '" "V ~ " ' " ' t\.." "'" "' 1). I 0.5 QOS
. GRAVEL SAND
Coors• 1 Fine Coarse I Medium I Fine
WATANA .
GRAIN SIZE CURVES -BORROW AREA E
':;)
.
..
.
0.01
FINES
I
If ' t
i
t
·•
'~ ' ~
l' L
~
~
1 'f.
'
.. • ·.
0.
10 .
20
~
~
~
50
Ia\
l-
70
lan
·-
90
!
-·~ ·• • ...
-!
l.
QOOIK)O
SllfSiz• ~Sizn
FIGURE 12.17
-~-------, -------------------------
• /
·~~~~R~~ ~HTHH~+-4---+H~~~~~~~~~~~0~~"~~~~++~~~~+H~~r-+-~~HH~+-~--~~~~4-~~~
~~++~-+~~"---H~+·~.~~~~~~~,~~~~~~~~~~~~~~~~~----~~~~~~~~~~~~--~~~~~-4--~ ~~~"~""' ~"' ~t-~
FINES SAND GRAVEL BOU...DERS COBBLES Slit Siza Medium Fint
WATANA
GRAIN SIZE CURVES~. BORROW AREA !
. FIGURE 12.18
--,,-
j -0.30
~AsLo--~--~,0--~~--2~0 --~--~3 -0 --~---4~0--~--~5-0--~---6~0--~--~70~--~--s~o--~--~9~0--~--~100)
Time (sec)
I
I ; ~
I I '
WATANA MAIN DAM
EARTHQUAKE TIME HISTORY
FiGURE 12.19 ••
I ._.,.
I ...
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I ,,
·r---~--------------~--------------~----------~~--~~---~~·-----~----~
1200'
1000
800
400
200
'"' 0
0
l..EG£Nl
e-FiXED CONE
0 • HC1LOW JET . .
3.
at
.
4.
v;
·9-.
.
'
22 • .
--·~ ---· ·----
'
' .• . ' . . '
-
.
~WATANA
12.-.,5
le• -
14. 6. ....
17. ..--100
z·e· 7. uO . .
21·-
.
so 75
VALVE 01At.£TER -INCHES
FREE DISCHARGE VALVE
EXPERIENCE PLOT
18 •
.SiTES .. t. 1<£BAN
2. ROUND BUTTE
3. TOCT RIVER
4. MAMMOTH ~-
5. 'LA AMISTAD
6. PORTAGE MTN.
1. COPETAN DAM a CAUFORNIA
9. BIG CRE~K
I 0. GLEN CN..m:N
t .... Kf.ERV....:e£ ow
12. VJU.ARINO 13.MANGLA
14,. llARTMOUnf DAM
I e~ NEW MELONES
t6.NEW~ES
11. JUNCTAN
I a r.ul MOliN'miN
19. NEW EXCHEQUER
20. CAVADO
21. 'PORTAL •
22. LAKE MATTHEWS
[,::IL ~ANYON
•• 3 . ,.
zoe!
5-. -
lOO 125
FIGURE! 12'.20
..
I ......
._
' I 740
-
:-
I ..
720
I
-· ...
I ..
700
I
I ......... ·
680 .
.
I 1-
LU
I
LU 660 u.
I
0
~
LJ~
%
I 1-
&IJ z
"-.
640
.
I
I 620
"'
I •'
-.
l
. /, ,.-~~
BEST EFfH IEitt-1
I
I 600
1\
l
I .
RESERVOif
;
EL.. 2045 -... .,...._
580
I
I 100 120
I
I
I
it~~ G!NERA" "OR
RA D POWER
RESERV<~R EL. 2!85 .·/ .
c
....-.-.:. --.~ .
l
" ,~ I
I
I
. v WEIGHTED fl. 'ERAGE HEAD
I ~ INIMUM DECEMB ER HEAD
I .
f+-170 MW .
I
L FUU. GATE
I
.
160 180 200
UNIT OUTPUT-MW
FIGURE 12.21
.
220
•
••••
.
.
.. ·
· .. , .. , . . :;:· . ' .
. . "-ll'
I
rei ~·_/'"-
' ·"'·~ · ....
I
I
I
.~· •, "'
I -·
I
I
I
t
I
I
I
I
I
I
I
I
. ,-'
#:. -,.
i
&lj.
a ~.~ ii: 80 t--------~~----+------+------+-'-~-~~------. ~ .&&..
&II
-ell
Ll.. To~------~------~------~--------~-7~--~--~~s
. IJ.I
C)
a::
~
~
~ cs
t-· .-----+-----+-~E:.---+----+-----+---12000!
120,000 100,000 200,000
TURBINE OUTPUT { HP)
WATANA -TURBINE PERFORMANCE
(AT RATED HEAD)
. \'
. \\
2.4q:K)O
FtGuA~ 1a;a2 .• Iil· .. ·
~· .· . .
1\,.
I
I
I
I
I
I
I
I
I
I
I
I
I
a,
I
a
I
·I
.'~
2 • .,
• 0 ~~-------+--·------4-~----~~------~--------~~----~
.40 I
27 ... ~3 • •
a ~~--------+--~----4-------~~------~--------~--·----~ c w
%
.... . w. z
.. 3 . J ~· WATANA~0
• 15
~~------_.·--~~~~·a~------~--------~------~------~ DEVIL CANYON-.........,;..0 ' • 22 Jl 3~ !a . •
24 • .29 13
,.31
6
,
.32
~~----~-+------+---~"~~----4-------4------~
~N .a
J7
•21 ~·~~~ ~ 200~~-----+--------~------~~------~--~~~-'28~~~~~------~
20 S) 60
SPECIF'IC SPEED ( N. S ) "
. FRANCIS TURBINES
SPECIFIC SPEED EXPERIENCE CURVE
FOR RECENT UNITS
ao
FIGURE 12.23 [iJ
.I
'
' I
. a·· .. . '
I
I -"' ,,
I _,
'
' t
I
I
I
I
~
I
13 ··DEVIL CANYON UEVELO~MENT
This section describes th¥a various comp.onents of ·the Devil Canyon development,
including diversion facilities, emergencyrev~ase facilities, main dam, primary
outlet facilities, res~voir, main and emergency spillway; saddle darn, the power
intake, penstocks, and the powerhouse complex, including turbines, generators,
mechanical and.electrical equipment, S\'litchyr d structures, and equipment and
project 1 ands •
A description of permanent and temporary access and support facilities is also
included.
13.1 -General Arranaement
The evolution of the Devil Canyon general arrangement is described in Section .
10. The De vi 1 Canyon reservoir and surrounding area is shown on Plate 64. The
site layout in re.1ation to main access facilities and camp facilities is shown
on Plate 66A. A more deta i 1 ed arrangement of the various site structures ; s
presented in Plate 64.
The Devil Canyon dam will form a reservoir approximately 31 miles long with a
surface area of 7,800 acres and a total volume of 1,092,000 acre feet at a nor-
mal maximum operating elevation of 1455. The operating level of the Devil Can-
yon reservoir is controlled by the tailwater level of the upstream Watana devel-
opment. During operation, the reservoir will be capable of being drawn down to
a minimum elevation of 1405. ·
The dam will be a thin arch concrete structura with a crest elev.ation of 1465
and maximum height of 645 feet. The darn wi 11 be supported by mass concrete
thrust blocks on each abutment. On the left bank, the generally lower ground
surface ievei will requir-e a substantial thrusta Adjacent to this thrust block,
an earth-and rockfi 11 saddle dam wi 1 1 pro vi de closure to the left bank. The·
saddle dam will be a central core type generally simtlar in cross section to the
Watana dam. The dam will have a maximum height above foundation level of
approximately 260 feet •.
During construction, the reservoir wi 11 be diverted around the main. construction
area by means of a single concrete-lined diversion tunnel 32 feet in diameter on
the left bank of the river.
A power intake located on the right bank will comprise an approach channel in
rock 1 eadi ng to a reinforced concrete gate· structure. From the intake structure
four penstockss consisting of concrete-lined tunnels each 20 feet in diameter
will lead to an underground-powerhouse complex housing four Francis turbines
each with a rated capacity of 150 MW and four semi-umbrella type generations
each rated at 180 MVA. Access to the powerhouse complex wi i 1 be by means· of an
unlined access tunnel approximately 3,200 feet long, as well as a vertical
access shaft about 950 feet deep.. Turbine discharge will be conducted to the
river by means of a single 39-foot-diameter tailrace tunnel leading from a surge
chamber downstream from the powerhouse cavern. Compensation flow pumps at the
power plant will ensure suitable flow in the river between the dam and tailrace
tunnel outlet portal. A-.separate transformer gallery just upstream from the
powerhouse cavern will house six single-phase 15/345 KV transformers. The
transformers will be connected by 345-KV, single-phase$ oil-filled cable through
a cable shaft to the switchyard at the surface.
J3-1
-·
The primary outlet facility will consist of .seven individual outlet conduits
located in the lower part of the main dam; it will-be de:stgned ,to discharge all
floods With a frequency of 1:50 years or less. Each outlet conduit wi11 have a
fixed-cone valve similar to those provided at Watana to tni'nimize undesirable
nitrogen supersaturation-in the flows· downstream. Flows ·resulting from floods
with a frequency areater -than 1:50 years but 'less than 1;10,000 yea~s will be
discharged by a ehute spillway on the-right bank. alsQ similar in design to that
provided for Watana• An emergency spillway on the left bank wi11 provide suf··
fici~nt additional capacity to permit discharge of the PMF without overtopping
the dam;; An emergency-release. low-level outlet facility will allow lowering of
the dam to permit emergency inspection or repair.
13.2 -Site Access
(a) Roads
At Devil Canyon the main access road will enter the site from the south. A
low level bridge crossing the Susitna River will be located just upstream
·of the dam. ln addition to the main access, several ancillary roads. will
be required to the camp. village, tank farmt borrow areas, and construction
roads to the dam and all major structures. The~e roads, with the exception
of temporary haul roads, are shown on Plate 66A~~
The construction roads will be gravel-surfaced roads 40 feet wide with
small·radius curves and grades 11mited to 10 percent. Major cut and fill
work will be avoided. A gravel pad approximately five feet thic ·will be
required for the roads. This will provide a drivable surface and also will
protect against the sporadic permafrost areas.
(b) Bridges
The existing low level bridge upstream of the-dam will be used to cr.oss the
Susitna River during construction. This bridge will be used during abut-
ment excavation. After construction of the cofferdams is complete, the
crests of these structures will be used to cross the river.
After construction of the main dam is completed, the crest of the main dam
will prOw'ide access across the Susitna River.
{c) Air-strip
A permanent airstrip will be located at the Watana site, approximately 30
miles west of the Devil Canyon site. This strip will be used far the Devil
Canyon development. Thet airstrip will be capable of accommodating the
C-130 Hercules aircraft, and will also accommodate small jet passenger air..:
craft.
{d) Access Tunnel
An access tunnel wi 11 be provided· to the underground pawerhous·e and associ-
ated works~ The main access tunnel will be concrete-lined and will be
approximately 35 feet Wide and ~8 feet high. The tunnel will a'llow perman·
ent access to the operating development and will also be utilized during
construction as the main construction tunnel. The tunnel will have con-
struction adits branching off to the various components of the development
during construction.
I
-1-, ...
.,, ·.-..
·. '
I
I
I
I -...;
J
I
'
' I ...
' I
I
I
t
I
t
I
.,---." --
..
_,-) ~
L ,
01
I
I
I ...
I _,
'
' I
' I
I
I
'
' I
(e)
A vertical 20--foot diameter access· shaft with an elevator will also be pro-
vided for_ access to the underground facilities. The powerhouse access
through this shaft ~Will be at the opposite end to the access tunnel. -..
13.3 -Site Facilities
(a) General
The. construction of'the Devil Canyon development will require various
facilities to support the construction activities throughout t~e entire
construction period. Following construction, the planned operation and
maintenance of the development will be centered at the Watana development;
therefore, minimum facilities at the site will be required to maintain the
power facility.
As described for Watana (Sect-ion 12), a camp and construction village will
be constructed and maintained at the project site. The camp/village will
provide housing and living facilities for 3,200 people during construction.
Other site facilities include contractor's work areas~ site power, ·
services, and communications. Items such as power and conmunications and
hospital services will be required for construction operations independent
of camp operations.
It is planned to dismantle and demobilize the facility_upon completion of
the project. After demobilizing the site, the area will be reclaimed. It
is planned to utilize dismantled builC:·~ngs and -other items from the Watana
development as much as possible in the camp/village.
Since the Watana develo_pment will be in service during the construction
period, electric power will be available. It is~therefore planned to meet
all heating requirements with electric heat and not with fuel oil, as is
planned for the Watana development. The salvaged building modules from the
Watana camp/village will be retrofitted for electric heat.
(b) Temporary-Camp and Village
The proposed location of the camp/village is on the south bank of the
Susitna River between the damsite and Portage Creek, approximately 2.5
(see Plate 66A) miles southwest of the Devil Canyon. The south side of the
Susitna was chosen because the w.ain access is from the south. South-facing
slopes will be used for the. camp/village location.
The camp will consist of portable 'l!.'codframe dormitories for single status
. workers with modu 1 ar mess ha 11 s, recreation a 1 bui 1 dings, bank, post office,
fire·· stat ton, warehouses, hospita 1, offices, etc. The camp wi 11 be a
single status camp for approxi111ately 2,900 workers.
The village, designed for approximately 320 families, will be grouped
around a service core containing a school, gymna.sium, stores, and recrea-
tion area.
-!
'" --.''.-·: "1
The two areas wi 11 be separated by approximately 1/2 mile to provide a
buffer zone between ar~as. The hospital will serve both the main camp and
the village ..
This ctrnp 1_ocation wi.11 separate 1i ving areas frorn the work areas by a mile
or more and wi 11 keep t\t-avel t1 me to work to less than 15 minute$ for most
personnel.
The camp/village will be constructed in stages to accommodate the peak work
force as presented in Table 13.1. Table 13.1 also presents the camp/
village facility design ntJnbers. The facilities have been designed for the
peak work force plus 10 percent for "turnover 11 • The "turnover .. includes
provisions or buffers for overlap of workers, vacations, and visitors. The
conceptua 1 1 ayouts for the camp/vi 11 age are presented in P 1 ates . and ,. ----~
(i) Site Preparation
Both the camp and the village areas will be. cleared in select areas
for topsoil, and the topsoil will be stockpiled for future use in
reclamation operations. At· the village site, selected areas will be
left with trees and natural vegetation intact.
Both the main camp and the village site have been selected to pro-
vide well-drained land with natural slopes of 2 to 3 percent. A
granular pad varying in thickness up to 8 feet will be placed in
selected areas at the main camp. This wi 11 provide a uniform work-
ing surface for erection of the high density housing and service
buildings and wi 11 serve in certain areas to protect the permafrost
where 1 t under 1 i es the camp. In the vi 11 age area, a granu 1 ar pad
will be installed only as-necessary to support the housing units and
to provide a suitable base for construction of the temporary town-
center buildings~
~'-
All roadways within the camp/village areas will be flanked by road-
side ditches, with CMP culverts carrying water across the intersec-
tions. In general, drainage wi 11 be through construction of a sur-
face network of di.tcheso Peripheral ditches will intercept overland
flows from adjacent non-cleared land and carry it around the camps.
Runoff will ultimately be directed to existing drainage channels
leading to the. Susitna River for the village and the main camp.
( i i ) F ac i 1 it i e s
Construction camp buildings will consist largely of trailer-type
factory-built modules assembled at site to provide the various
facilities required. The modules will be fabricated with heating,
lighting, and plumbing services, interior finishes, furnishings~ and
equipment. Trailer modules will be supported on timber cribbing or
blocking approximately two feet above grade. -
. l
I
' I
I
I
I
I
I
' t
I
I
t
I
' I
' I
I
I~
I
' 'I
I
I·
I
'I
I
I
t
I .,
I
' I
I
I
I
Larger structures such as the, central utilities building, ware-~.
houses, and hospital will be pre-engineered, steel-framed structures
, with metal cladding.
The larger structures will be erected on concrete-slab foundations.
The slab will be cast on a non-frost susceptible layer at least
equal to the thickness of the annual freeze/thaw layer. Heated
permawalks will connect· the majority of the buildings and dorms.
The various buildings in the camp are identified on Plate 67.
(c) Site Power and Utfl ities
(i) Power
.Electrical power will be required to maintain the camp/village and
construction activities. A 345 kV transmissioh line and substation
will be in service during the construction activities. Two trans-
formers will be installed at the substation to reduce 1;he.line vol-
tage to the distribution voltage, One of the transformers will be
the same transformer used at the Watana development during its con-
struction. · ·
Power will be sold to the contractors by APA. The peak demand
during the peak camp population year is estimated at 20 MW for the
camp/village and 4 MW for construction requirements, thus totaling
24 MW of peak demand. The distribution system in the camp/village
will be 34.5 kV.
(i i) Water
The water supply system will serve the entire camp/village and
s~lected contractor•s work ar·eas. The water supply system will pro-
vide for potable water and fire protection. The estimated peak pop-
ulation to be served will be 3,950 (2,900 in the camp and 1,050 in
the village).
The principal source of water will be the Susitna River, with a
backup system of wells drawing on ground water. The water will be
treated in accordance with the Environmental Protection Agency {.EPA)
primary and secondary requirements. ·
A system of pumps and constructed storage reservoirs will provide
the necessary system demand capacity.
The water distribution system will be a ductile iron pipe sy.stem
contained in utilidors as described in Section 12.3.
(iii) Waste Water
The waste water collection and treatment system will serve the camp/
village. One treatment plant will serve the camp/village. Gravity
flow lines with lfft stations will be used to collect the waste
·water from all of the camp and village facilities. The 11 in-camp 11
0
13-5'
and 11 in .... vi l'lage'' collection systems will be run through the perma-
walks and uti lidors so that, the collection system wi 11 always be
protected from the elements.
·At the villaget· .an aerated collection basin will fie installed to
collect the sewage. The sewage will be pumped from this collection
· ·basin through a force main to the,sewage treatment plant.
An aerated collection basin will be needed at the village to balance
out the high 1 y eye 1 i c waste water flows. -
Chemical toilets located around the site will be serviced by sewage-
trucks, which wi 11 discharge direct 1 y into the sewage treatment
plant.
The sewage treatment system will be a biological system with
lagoons. The system will be designed to meet Alaskan state water
1 aw secondary treatment standards. The 1 agoons and system wi 11 be
modular to allow for growth and contraction of the camp/village.
The location of the treatment plant is shown on Plates . and· •
The location was selected to avoid unnecessary odors in the camp as
described for Watana.
The sewage plant will discharge its treated effluent through a force
main to the Susitna River. All treated sludge will be disposed of
in a solid waste sanitary landfill.
(d) Contractor's Area
The contractors on the· site will require officet shop and general work
areas. Office space for the contractors has been provided and ~its location
is shown on Plate 66A.
Partial space required by the contractors for fabrication shops, storage or
warehouses, and work areas within the camp confines has been designated and
is shown on Plate· 66A. Additional space required for the aforementioned
i terns wi 11 be located between the main camp and the main access road.
13.4 -Diversion
(a) General ,, __
Diversion of the river flow during construction will be accomplished with a
~i ngl e 32-foot diameter horseshcs shaped section diversion tunne 1. The
concrete-lined tunnel will be located on the left bank of the river, and
wi 11 be 1,490 feet in length.. The diversion tunne 1 p 1 an and profile is
.shown on Plate 69.
The tunnel is designed to pass a flood with a return frequency of 1:25
years routed through the Watana Reservoir •. The peak inflow will be 37,800
cfs. Routing effects are negligible and the peak flow that the tunne.l will
discharge will be 37,800 cfs. The maximum water surface elevation upstream
of the cofferdam will be E1 944. A rating curve is presented in Figure
13.1.
13-6
•••
I
' I
I
·I
I
I
I
t
t
I
I
I
f.
t
t ., .. _ .. ; . ;
·I· , . . . i
,. l . ~. 41
I
I
I
I
I I ,,~,~'
I
I
I
I
I
I
I
I
I
t
I
I
I
{b}
{c)
..
' ,. "" ~,' -c-.. :-;-.---,~~.:::;"--':;:·4~-;··-~"'•< ~----!? . .,.. •.••
_-::.,.._
Cofferdams
Due to the depth of alluvium present in the Susitna riverbed fount:iation~ a
grouted zone thr-ough the alluvium material to bedrock excavation and to
sound rock in the abutment areas will be installed.
The depth of al1uvium material in the riverbed ranges up to a maximum of 70
feet. The alluvium material consists of open-worked gravels with numerous
cobbles and boulders. Due to the coarseness of the alluvium material, a
grouted zone was selected rather than a slurry wall; it will be constructed
through the closure dam and alluvium material to bedrock and will minimize
the amount of seepa4Je into the main dam excavation. The abutment areas
will be cleared and grubbed with excavation of all material to sound rock
prior to placement of any cofferdam material.
.The upstream cofferdam wi 11 be a zoned embankment found on the closure dam
(see Plate 69). The closure dam wil1 be constructed to Elevation 915 based
on a low water level of Elevation 910 and will consist of coarse material
on the upstream side grading to finer material on"the downstream side.
When the closure dam is completed, the groutjng will commence and the zone
will be constructed to minimize seepage into the main dam foundation
excavation.
The cofferdam from Elevation 915 to 947 will be a zoned embankment consist-
; ng of a centra 1 core, fine and coarse upstream and downstream filters., and
rock and/or gravel shells with riprap on the upstream face.
The downstream cofferdam .J~ill be a closure dam constructed from Elevation
860 to 895 (see Plate C:9). It will consist of coarse material on the
downstream side grading to finer material on the upstream $ide. When the
closure dam ts completed~ the grouted zone can be constrw:teci in the finer
material to minimize seepage into the main dam foundation excavation.
The upstream cofferdam crest elevation wi11 have a 3 foot freeboard allow-
ance for settlement and wave .run up.. Thermal hydraulic studies conducted
showed the discharged waters from the Wat.ana Reservoir will be 34°F when
they pass through Devil Canyon. Thus, an ice cover will not form upstream
of the cofferdam, and no freeboard allowance was made for ice.
Tunnel Portals and Gates
A reinforced concrete gate structure will be located at the upstream end of
the tunnel (see Plate 70). The portal and gate for the tunnel will be
designed for an external pressure (static) head of 250 feet.
Two 30 feet high by 15 feet wide water passages will be located in the· gate
structure with guides for the diversion closure gates separated by a center
concrete pier. Each gate will be a fixed wheel vertical lift gate. operated
by a wire rope hoist ·;nan enclosed housing. The gate will be designed to
operate with the reservoir at Elevation 950, an 80 foot operating head.
Stoplog guides will be installed in the diversion tunnel outlet portal, and
. stoplogs will be provided to, permit dewatering of the diversion tunnel for
plugging operations. · ··
13-7
The. stoplogs will have a downstream skinplate and upstream seats (relative
to river flow) and will be arranged in suitable sections to facilitate
reJatively easy handling, with a mobile crane .using a fo.llower beam.
(d) pperation During .Diver.~ion
The tunnel wi 11 pass all flows from BOO cfs to the upper design flow of
37,800 cfs. The rating cur-ve for the diversion tunnel is shown in Figure
13.1.
(e) Final Closure andReservoir Filling_
Upon completion of the concrete dam to an elevation sufficient enough to
pass the environmental flows with the discharge va 1 ues that are i ncorpor-
ated in the dam, the tunnel will be temporarily closed with the intake
gates-and crynstruction of the permanent plug will commence. It is
estimated it will take a year to completely place and cure the plug.
During this time the upstream gate and intake structure wi 11 be designed
for a reservoir elevation of 1130, which will create an external pressure
of 250 feet.
The filling of the reservoir will take approximately 11 days to full
reservoir oper-ating elevation of 1455.
13.5 -Arch Dan ·
{a) General
The arch dam at Devil Canyon has been selected in preference to a rockfill
dam for the reasons given in Appendix 05$ The s_hape of the canyon is suit-
ed to an arch dam, with a crest length-to-height ratio of .approximately 2.
The height of the dam will be approximately 650 feet, \'iell within the range
of heights of dams constructed elsewhere. A comparative list of some 'large
arch dams constructed throughout the world is given in Table 13.2.
Because Alaska is a highly seismic area., the arch dam wi 11 be designed to
withstand dynamic loadings from intense seismic shaking .. Some dams con-
structed throughout the world in high earthquake areas include the 741-
foot high El Cajon dam in Honduras, the 696-foot high .Mohamed Reza Shah
Pahlavi dam in Iran, and the 548-foot high Vidraru Arges dam in Rumania.
The Vidraru Arges dam and the 372-foot high Pacoima dam in California have
both withstood high earthquake loadings, with the latter experiencing a.
bas,e ground acceleration of between 0.6 to 0.8 g.
Green Lake dam is presently being constructed to a height of 210 feet in
Sitka,. Alaska.
{b) Location
The arch dam will be located at the upstream end of the canyon at its
narrowest point. The rock is outcropping or very close to the surface at
the abutments, and the c'ontours just downstream of the left abutment swing
in toward the ,river so that the left side of the dam wi 11 be founded
against the upstream side of a slight promontory.
13-8
I
I
I
I
I
I
I
I
I
I
I
I
I
'
' I
I
I
I
' ~."li··~ ~
I
I
I
I .,
I
II
I
I
I
I
••
I
I
I
I
I
I
I
The rock forming the right abutment rises several hundred feet above ·the
·dam crest but on the left side the rock surface rises only to Elevation
1400. It will be necessary to construct a mass concrete thrust block at
t~ds point to artifically form the bearing surface of the dam.
(c) Foundations
The arch dam wi 11 be founded on sound bedrock located 20 to 40 feet below
the bedrock surface. The foundation will be excavated and trirrmed beneath
the dam so that no abrupt irre·gul ariti es \'li 11 occur at the foundations
which could cause stress concentrations within the concrete. During exca-
vation the rock will also be trimmed as far as is practical, to increase
the synrnetry of the-centerline profile and provide a comparatively uniform
stress distribution across the dam. Areas of dikes and the local areas of
poorer quality rock will be excavated and supplemented with dental con-
crete.
'lne foundation will be consolidation grouted over its whole area, and a
double grout curtain up to 300 feet deep will run the length of the dam and
its adjacent structures as shown in Plate 75. Grouting will be done from a
system of galleries which will be run through the dam and into the rock.
Within the rock these galleries will also serve as collectors for drainage
holes which will be drilled just downstream of the grout curtain and will
collect any seepage passing the curtain.
High on the left abutment open cracks are evident in the rock; these will
be excavated t·o sound rock and the excavated material replaced with can ...
crete in the form of a deep thrust b 1 ock. On the right abutment a mass
concrete thrust block will be founded at the end of the dam to match the
left block and improve the dam symmetry.
(d) Arch Dam Geometry
The philosophy and design of the dam is more thoroughly described in Appen-
dix D, but is su11111arized herein. The dam geometry is shown in Plates 72
and 73. The crown section at the center of the river wi11 be of a double
curved cupola shape inclined downstream. The static load from the reser-
voir will be taken primarily in the arches; the three-dimensional stress
action of the structure 'Ni 11 tend to induce tension in the downstream face
of the cantilever. This will be-offset by the gravity forces of the over-
hanging section, which also will counteract any seismic loadings produced
by downstream ground motion. ·
A two-center configuration wi 11 be adopted for the arches to counteract ·. :~
sli.ght assymetry of the valley and give a more uniform stress distributio.-,
across the dam. The arches ·wi 11 b·l formed by circles with c~oters located
on the vertical axis plane running along the center of the canyon. The
radi.i of the arches on the right and wider side of the canyon will be
greater than· those on the 1 eft, and the thrust wi 11 be directed more nearly
~normal to the rock abutment rather than parallel to the face" as would
occur with a smaller radius arch. The radii of the i ntrados or downstream
face will be smaller than those of the extrados, producing a thicken·Ing of
the dam at the abutments where stresses would tend to be highest.
13-9
'•
(e) Thrust .. Blocks
The· thrust blocks are shown on Plate· 7 4. The massive concrete-b 1 ock on the
left abutment wi11 be formed to take the thrust from the upper part of the
dam above the existing sound rock level,,. The. thrust block will also serve
as a transition between the concrete dam· and the adjacent. rock fill saddle
dam. Ttle incline!!.enct~~~e.nf_.the block will abut and seal against the
impervious saddle dam co_re-, and it will be wrapped by the supporting .rock
shell.
A thrust block will also be formed high on the right.abutment at the end of
the dam and adjacent to the spi 11 way control structure. The b 1 ock wi l1
improve the symmetry of the dam profile, as previously stated, and will be
stable under load conditions similar to those incurr.ed by the thrust block
on the left bank.
(f) Construction and Schedule
Construction of the dam will be completed over a five-year period as des-
cribed in Section 17. Construction will take place throughout the year
with cooling coils built into the concrete to dissipate the heat of hydra-
tion and special heating and insulation precautions taken in the winter to
prevent excessive cooling :vf concrete surfaces.. Concrete aggregates will
be obtained from the alluvial deposits in the terraces upstream of the
dam. ·
Concrete will be placed by means of three highlines strung above tha dam
between the abutments.
13.6 -Saddle Dam
The design philosophy for the saddle dam at Devil Canyon is essentially the same
as that for the main dam at Watana described in Section 12.6. The most
significant difference is the exclusive use of rockfill in the shells instead of
. river gravels used a.t Watana. The use of gravels in the upstream shell at
Watana is to minimize settlement of the sh.e11 on saturation during filling of
the reservoir and to ensure a free draining material. These aspects of the
design are not as significant for the much smaller structure at Devil Canyon.
The amount of settlement will be less and the drainage paths for the dissipation
of any e·xcess pore pressures will be much reducedo Many dams of equal or larger
dimensions have been constructed of similar materials and the design is well
within precedent.
(a) Proposed Dam Cross Section
[,etails of th.e proposed saddle dam are shown in Plate 76. As at Watana,
the central vertical impervious core will be protected by fine and coarse
filters on both upstream and downstream slopes and supported by rockfill
she.lls. The core w111 have a crest width of 15 feet and side slopes of
1H:4V to provide a core thickness to dam height ratio slightly in excess of
0.5. .
The wi'de filter zones will provide sufficient q}aterial for self-healing of
any cracks which might occur in the core du~ tc settlement or· as the result
of seismic displacement.
13-10 _,
0
I
I
I'
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
.;.lJ
I
I
I
I
·.I
I
I
I
I
I
I
I
I
I
I
:1
I
I
I • j
..
The saturated sections .of both she.11s ~lill be constructed of compacted
clean rockfill, processed to remove fine material in order to minimi.ze pore
pressure generation and ensure rapid dissipation during and after a seismic
event. Pore pressures cannot develop in the unsaturated .section of the
downstream shell and the material in that zone will be unprocessed rockfi11
from surface or underground excavation~~.
Protection on the upstream slope will consist of a 10-foot layer of rip-
rap.
(b) Sources of Construction Material
No source of materia 1 suitable for the~ core of the saddle dam has been
identified closer than the borrow areas at Watana (Areas D and H). The
current proposals are to develop Area D for core material at Watana and,
since access roads will be established to that area, the same source will
be used for the Devil Canyon core. Investigations to date indicate that
suitable material can be obtained from areas above the Watana reservoir
1evele In the unlikely event that insufficient material is available from
Area D, then Area H would be de'Jeloped. The in-place volume of core
material is 306,000 cubic yards.
The.filter matet~ial will be obtained from the river deposits (Area G)
immediately upstream of the main arch dam at Devil Canyon. This area will
also be exploited for concrete aggregates. The total volume available in
Area G is estimated to be 6 million cubic yards, while the concrete
aggregate demand is some 2.7 million cubic yards. The estimated voltrnes
required for the dam are 228,000 and 181,000 cubic yards for the fine and
coarse filters, respectively. ·
Rockfill for the shells will be obtained primarily from the excavations for
the spi 11 ways, tunne 1 s, and powerhouse camp 1 ex. The tot a 1 rock fill
required will ·be approximately 1.2 million cubic yards. The proportion of
sound rock sui tab 1 e for use in the dam-which can be obtai ned from the exca-
vations cannot be accurately assessed at this stage, but it i.s proposed to
make good any shortfall· by deepening and extending the emergency spillway
·cut. This will be more economical and environmentally acceptable than
developing quarry Area K, some 2 miles from the damsite.
(c) Excavation and Foundation Preparation
The excavation and foundation preparation will be as for the Watana site
with all alluvium and other unconsolidated deposits under' the dam removed
to expose sound bedrock to eliminate any risk of 1 i quefacti.on of the dam
foundation under earthquake loading. Weathered and heavily jointed rock
will be removed from beneath the core~ and filters and local irregular·ities
in the rock surface either_ trimmed back or concrete added to provide a
suitable contact surfaca for placing the core.
(d) Grouting and Pressure Re 1 i tll,
'· As at Watana, the rock foundation will be improved by consolidation grout-; ng over the core contact area and by a grouted, cutoff a 1 ong the center 1 i ne
of the core. ·The cutoff at any location will extend to a depth of 0.7 the
water head at that , 1 ocati on as shown on Plate 75.
13-ll
(e)
,,
(f)
(g)
. •, .
A grout ·ing and drainage tunnel will be excavated in bedrock. b~neath the dam
along the centerline of the core and. will connect with a similar tunnel
beneath the adjacent concrete; arch dame Pressute relief and drainage holes
will be dr-illed from this tunnel and seepage from -the drainage system will
be disehar.g~d into the arch dam drainage system and ultimately downstream
be;low tailwater level. ·· ·
lmpervi ous Gore and Fi 1ters.
The requirements for impervious core and both fine and coarse filters will
be as for the Watana dam (Section 12.6).
Rockfill Shells
The processed rockfill to be placed in the saturated zones of the. dam will
have the same grading as the processed alluvium used at Watana. The maxi-
mllll size shall not exceed 12 inches and not more than 10 percent of the
material shall be f·iner than 3/8 inch size. This restriction on fine
material will not apply to rockfill in the unsaturated zone above Elevation
1375 in the downstream shell. All rockfill wi:l be placed and thoroughly
compacted in 24-inch layers.
Freeboard and Superelevation
The highest reservoir level will be Elevation 1466 under maximum probable
·flood ( PMF) flows. At this e 1 evat ion the fuse p 1 ug in the emergency spi 11-
way will be breached and the reservoir level will fall to the spillway sill
elevation of 1434. The norma 1 maxi mum poo 1 e 1 evat ion i s 145 5.
It is proposed that a minimum freeboard of three feet be provided for the
PMF flood; ·hence, the crest of the saddle dam cannot be lower than Eleva-
tion 1469. In addition, an allowance of one percent of the height of the
dam will be made for potential slumping of the rockfill shells under
seismic loading. An allowance of one foot has been made for settlement of
both abutments; hence, the final crest elevations of the saddle dam will be
1470 at the abutments, rising in proportion to the total height of the dam
to Elevation 1472 at the maximum section. u·nder normal operating cond1-
ti ons, the freeboard will range from 15 feet at the abutments to 17 feet at
the center of the dam. Further allowances must be made to compensate for
. static settlement of the darn after completion due to its own weight and the
effect of saturation of the upstream shell, which will tend to produce
additional breakdown of the rock fill at point contacts. It is proposed ·
that one percent of the dam height be allowed for such settlement, giving a
maximum cTest elevation on completion of the construction of 1475 at the
maximum height~ and 1471 at the abutments.
The a·ll owances for post-construction sett 1 ement and seismic s 1 umpi ng wi 11
be achieved by steepening both slopes of the dam above Elevation 1400.
13-12 · ..
I
I
' I
1:
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
:I
I
I
I
(h} Instrumentation .
Instrumentiltion will be jnstalled within all partso of· the dam to provide
morlitoring.during cons:truction as \vell as ·durfng operation. Instruments
for measuring_-internal vertical and hortzontal displacements, stresses and
strains, and total and fluid pressures, as well as surface monuments and
markers, will be installed. The quantity and location will be decided dur-
ing final design. Instrumentation will include the following:
(i). Piezometers ... fo measure static pressure of fluid in the pore spaces
of soil and rockfill.
(ii) Internal Vertical Movement Devices
... Cross-arm settlement devices as developed by the USBR;
-Various versions of the taut-wire devices deve 1 oped to measure.
internal settlement; and
-Hydraulic settlement devices of various kinds.
(iii) Internal Horizontal Movement Devices
-Taut wire arrangements;
.. Cross-arm devices;
-Inclinometers; and
-Strainmeters.
(iv) Other Measuring-Devices
-Stress meters;
-Surface monuments and-alignment ,markers; and
-Seismographic recorders and seismoscopes.
(i) Stability Analyses
As .at Watana, special precautions have been taken to ensure stabi 1 'ity under
earthquake 1 oading by the use of processed free draining rockfi 11 in t'he
saturated zones of the dam, the incorporation of very wide filter zones,
and the removal of all unconsolidated natural material from beneath the
dam.
Static and dynamic stability analyses of the upstream slopes of the Watana
dam (Section 12.6), have confirmed stable slopes under all conditions for a
24H:1V upstream slope and a 2H:1V downstream slope.
Since these same slopes have been used for the Watana dam and the construc•
tion materials are essentially similar, it was considered unnecessary to·
carry out further analysis for the specific details of the saddle dam to
confirm feasibility, though such analyses will be required during the final
design phase.
13-13
13 .. 7 .~ Primary Outlet ~aci lities
(a) Gen.eral
The p.rfme function of the outlet facilities is to provide for discharge
through the main dam of routed floods with ·UP to 1 :SO years recurrence
perjod at .the Devil Canyon reservoir.· .. Downstream erosion is to be mintmal
and nitr·ugen supersaturation of the releases is to be restricted as much as
possible as in the case of.the Watana development .. A further funct.ion of
the releases is t_o provide an emergency drawdown for the reservoir, should
maintenance be necessary on the main dam or low level submerged structures,
and a 1 so to act. as a diversion during the 1 atter part of the construction
period as described in Section 13.4.
The facilities will be located in ·the lo.wer portion of the main dams as
shown on Plate 76A, and will consist of seven free dischnrge valves set in
the lower part of the arch dam!'
(b) Outlet
The discharge valves will be fixed· cone valves located at two elevations:
the upper grouping, consisting of four 102-inch diameter valves~ will be
set at Elevation 1050, and the lower grouping of three 90-inch diameter
valves wi 11 be set at Elevation 930. The valves will be installed radially
(normal to the dam centerline) with the upper set centered ott a point
slightly downstream of that corresponding to the lower valves.
·rhe fixed cone valves will be installed on individual conduits passing
through the dam, set from the downstre·am face, and protected by upstream
ring follower gates located in separate chambers within the dam, as shown
on Plate 76A. The gates wi 11 serve to i so 1 ate the va 1 ves to all ow main-
tenance. Monorai 1 hoists· wi 11 be located above each valve and gate assem-
bly to provide for their withdrawal and maintenance. The gates and valves
will.be linked by a 20 foot high tunnel running across the dam and into the
1 eft abutment where ac~ess wi 11 be provided by means of a verti ca 1 shaft
exiting through the thrust block. Although secondary access will be p~o
vided via a similar shaft from the right abutment vehicle, access and
installation· are both considered to be from the left side.
The valve and gate assemblies will be protected by individual trashracks
installed on the upstream face. The racks will be removable along guides
running on the upstream dam face. The racks wi 11 be raised by a mobile
crane normally stationed at Watana but employed for,both sites.
(c) Fixed Cone Valves
The 102-inch diameter valves operating at a gross level of 420 feet and the
90-inch diameter valve.s operating at a head of 525 feet have been selected
to be within current precedent considering the valve size and the static .
head on the valve. The valves will be located in individually heated rooms
and wi 11 be provided with e 1 ect ric jacket heaters instal led around the
cylindrical sleeve of each valve. The valves will be capable .of year round
oper·ation~ although winter operation is not co.rltemplated.. Normally when
the valves are closed, the upstream ring follower gates will also'be closed
to limit leakage and freezing of water through the valve seats.
13-l4
/ •... ·
,.f ;
I
I
I
••
-·:
I
I
I
I
I
I
I
••
I
I
I
I
··I,
,. .. · .•. )::: 1~--~~··~~~~--~~.~~~~~~~~--~----~--~----~~~~--~~~~~~=<~~~···~··~ .. ~.·~~~-j~=
I
I·
I
----:---;:-~-=-:-:·;>·-~-:---·c-;--""----~--" -~
I
-1 (d)
••
I
I
I
I
I
I
I
-The v~lves will be operated either by twq .:hydraulic ~operators or' by screw
stef!l hoists" _ The former hav~ been a;ssumed for pre 1 imtnary design purposes.
The valves will normally be -operated remotely fr-om Watana, but local opera-
tion i.s also po$sible~ ·
Thorough research and model studies wi 11 be required for final design of
the valves, particularly fn regard to preventing vibration. In $izing the·
valves it has been assumed that the valve gate opening will be restricted
to 80 percent full stroke to reduc.e~pJ;l$,$_i_qi_1ities of vibration.,
Ring Fo 11 ower Gates
Ring follower· gates will be installed upstream' of each valve and will be
used:
-To permit inspection and maintenance of the fixed cone valves;
...; To relieve hydrostatic pressure on the valves when they are in the. cl<>sed
position; and ·
.. To close against flowing water in the event of malfunction or failure of
the valves. ·
The ring follower gates will have nominal diameters of 102 and 90 inches
and wi 11 be of we 1 ded or cast stee 1 constructi.on. . The gates wi 11 be
designed to withstand the total static head with full reservoir. Existing
large diameter, high head ring follower gates are suimlarized in Table.
12.6.. . c .
The ring follower gates will be designed to be lowered under flowing water
conditions with maximum head, although normally they will be raised under
b·alanced head conditions only. Valved bypass piping will be used to equal-
ize the pressure on both sides of the ring follower gate before raising.
The gates wi 11 be operated by hydraulic cylinders with a nominal operating
pressure-of 2,000 p.si. Either local or remote operation of the ring
follower gates will be possible.
A grease system will be installed in each gate for injection of grease
between the gate leaf and the gate body seats to reduce frictional fort~s
when the gates are operated.
I
I
(e) Trashracks
I
I
I
I
A steel trashrack ·wi 11 be installed at the upstream entrance to each 'water
.passage. The racks wi 11 prevent debris from being drawn into the d·i:scharge
valves~
The bar spacing on the racks wi 11 be approximately 7_ inches, and the racks
will be designed for a maximum differential head of about 40 feet. The
maximum net velocity through the racks will be approximately ft/s.
Provision will be made· for monit·oring h~ad loss across the racks.
. 13-15
_I
0
{f} Bulkhead Gates
" ,.
Provision w·iJl be made for installing intake bulkhead gates at the upstream
entrance to each of the conduits for the f1xed cone valves._ Embedded·
guides will be ins:talled at each· conduit entrance extending to above rnaxt-
rnum norma 1 water> 1 eve 1. ·
the bulkhead gates will be installed only under balanced head conditions
using a gantry crane. The gates for the ·upper valves will be 12 feet
square and wi 11· be. 10 feet square for the 1 ower va 1 ves.
Each gate will have a downstream skinplate and seal and will be designed to
withstand full differential head with maximum reservoir water level. One
gate of each size has· been assumed and the gates will be stored at the dam
crest level. ·
A temporary cover-will normally be p 1 aced in the bulkhead gate check at
trashrack level to prevent debris from getting behind the trashracks at the
front of the valve conduit.
The crane' for handling the.bulkhead gates will be an electric travelling
gantry type crane located on the main dam crest at Elevation 1468. The
estimated crane capacity is 40 tons. The crane will have a single point
lift hoist mounted on a moveable trolley. The hotst pickup will be incor-
porated into a follower for handling the bulkhead gates. "
13.8 -Main Spillway
(a) General
The main spillway at Devil Canyon will be located on the right hand side of
the e.anyon (see Plate 77). The upstream contro1 structure will ·be adjacent
to the arch dam thrust block and will discharge down an inclined concrete-
lined chute~ constructed down the steep face of the canyon, and then over
the flip bucket which will traject flows downstream and into the river
below.
The right side location for the spillway facilities is considered prefer-
rable to the left because of the superior quality of the rock~. with its
1 ower degree of weathering~ and the downstream a 1 i gnment of th.e river·~
which allows for spillway discharges paralleling the direction of flow.
The spillway will be designed to pass the 1:10,000 year routed flood at
Watana in conjunction with the outlet facilities, giving it a design capa•
city of·ll5.,000 cfs, which will be discharged over a total head drop of 550
feet. No surcharge will occur above the normal maximum reservoir operating
level of 1,455 feet. This will be below the discharge of 150,000 cfs over
600 feet on the Mica Project, developed for British Columbia Hydro., Canada,
which has operated successfully for a number of years ..
13-16
I
I
I
I
I
I
I
I
I
I
I
I
I·
I
I
I
I
I
I '•
0 ·--~~
<.!:--. . .. ~ '
-:<· •l
.,
I
I
I
I
I
I
I
·•-
1
I
I
I
I
'I
I
--1
I
I
(b) Approac)L.Channel and Control Structute
The approach channel wi11 be· excav.ated. ttl a depth. of approximately 100 feet
in the rock with a wi.4th of just over 130 feet and .an invert ~levation of
1375,.which will produce a flow velocity of approximately 11 ft/s under
design dischar-ge. · . · "----~ --· ---
The control structure wi1 1 be a three unit concr-ete structure set at the
end of the channe 1. · Each unit · wi 11 ·house a· 54 foot high by 30 foot wide
gate which will sit on t()p of an agee-crested weir and, in conjunction with
the other gate. units, wi 11 control the f1 ows passing though the spillway.
The gates wi 11 be fixed wheel gates operated. by individual rope hoists._
Eat:h gate will be contained with a separate monolith unit consisting of an
ogee overflow weir, piers, and integral roadway deck. The box configur-a·
tion of the unit will give the individual monoliths stability during earth-
quake motion, and the "split pier" construction, with each unit having its
separate piers, Will allow for some relative motion with no stress trans-
ference between units during seismic events and less chance of gate seiz-
ure. Model tests will be necessary during the final design stage to deter~
mine final geometry and dimensions of pier noses, crest. slope, and pier
lengths. The main access route will cross the dam and the control struc-
ture deck upstream of the gate hoists. The main dam grout curtain and
drainage system will pass beneath the structure. ·
{c) Spillway Chute
The spillway chute will cut across the steep face of the canyon for a dis-
tance of approximately 900 feet and will 'terminate at Elevation 1000~ The
chute will taper uniformly over its length from 122 feet at the upstream
end to 80 feet downstream.
The slope of the bedding planes on the r1ght abutment above the chute will
be at an approximate angle of 55° or more. Because of_ the i nstabi 1 ity
_along :these planes, the rock above the spillway will be cut back parallel
·to the bedding and the face will be reinforced with steel mesh and rode
bolts. ,
The chute itself will" be concrete-lined W'ith invert and wall slabs anchored
btack to the rock. The profi 1 e of the chute wi 11 be such that the invert
s 1 abs wi 11 be founded on sound rock. Pa·rt way down the chute on the side
closest to the river the depth of cut will be insufficient to provide the
supporting rock to the slabs; hence, the side wall will take the form of a
gravity section over approximately a 200 foot 1 ength.
The velocity at the bottom of the chute wi 11 be approximately 150 ft/s. In
order to prevent cavitation of the chute surfaces~ air will be introduced
into the discharges. Air will be drawn in along the chute via an under-
lying aeration gallery 'and offshoot ducts extendin-g to the downstream side
of a raised step running transverse to the chute, as shown on Plate 77 ..
The chute will be underlain by a series of box drains at the rock/concrete .
interface which will drain through drilled holes to an inclined rock tunnel
below running the length of the spillway.
13 ... TT
. -·~1
i
' i
(d) Flip Bucket
The spillway chute will terminate in a mass contrete fl'fp bucket -founded on
sound rock at E1 evation 1090. approximately 100 feet above the river. The ·
cu~ve of the flow ,surface· of the bucket will be adjusted to confine the
issutng dfschargej but at present i~ is ass·utned to be<cylindrica1 and will
be modified··at the fin_al d~si'gn stage following model tests-.-A grouting
drainage gallery will be provided wi.thin the bucket to allow for foundation
consolidation and relief of uplift pressures.
The jet issuing from the bucket will be traj_ected downstream and parallel
to the river below.
(e) Plunge Pool
The iiJ1lact area of the issuing spillway discharge will be limited to the
area of the downstream river surface to prevent excessive erosion of the
canyon side walls. This will be done by modification of the flow surface
of the flip bucket as described above. Over this impact area the alluvial
material in the riverbed will be excavated down to_ sound rock to provide a
plunge pool in which most of the inherent energy of the discharges will be
dissipated, although some energy will alreaqy have been dissipated by fric-
tion in the chute, in dispersion, and friction through the air.
It was considered necessary to excavate the river material to determine the
general area of any downstream erosion and also to prevent excessive eros-
ion and random downstream deposition of material which might occur if dis-
charges were allowed to·excavate their own pool.
13.9 -Emergency Spillway
(a) General
The emergency spillway will be located on the left side of the river beyond
the rockfill saddle dam. It \'lill be set within the rock underlying the
left side of the saddle and will continue downstream for approximately
2,000 feet. ·
An erodible fuse plug, consisting of impervious material and fine gravels,
will be constructed at the upstream end and will be designed to wash out.
when overtopped by the reservoir, releasing floods of up to 160,000 cfs in
excess of the combined main spillway and outlet capacities and thus pre-
venting overtopping of-the main dam. ·
(b) Fuse Plug and Approach Channel ·.
; .•...
' ;"-
. .
::····· .· .. .-·
I
I
I
I
I
I
I
I
••
••
I
I
The approach channel to the fuse plug will be cut in the rock and will have I
a width of 310 feet and an inv.ert elevation of 2170. The channel will be
cr-ossed by the mai nc access road to the dam on a bridge consisting of can-
crete piers., pref':ast beams, and an in situ concrete bridge deck. The fuse 'I•
plug will close the approach channel and will have~ maximum height of 31.5
feet with a crest elevation of 1465.5. The plug will be located on top of
a flat-crested weir excavated in the rock and protected with a concrete ,1·
slab. Since the rock slopes quite steeply at the channel location,. it is
I
. 13-18
;..
·I
I
I
I.
I
I
I
I
I
I
I
I
I
I
I
I
I
··,·
'~·
desirable to "keep the. spillway chute :as narrow as possible to reduc~ the
excavation quality. For this reason a drop se·ction downstream of the plug
will be introduced to ·;ncrea,se the discharge coefficient'S, at the plug ·
sectlgns and thus enable a reduction in the length of the plug. ·
The plug will be-traversed by a pilot channel with an 'invert elevation of
1464, and will have a similar zoning to that described in Section 12.10 for
Watana.
(c) Discharge· Channel
The channel will n~rrow downstream and lead to a high tributary valley
above the Susitna River. This channel-wi11 rapidly erode under high flows
but will serve the purpose of straining the initial f1ows in-the dir""eCtion
of the valley.
13.10-Devil Canyo!l Power Facilities
(a) Intake -
The intake structure is located on the right bank-as shown on Plate 83.
Separate intakes are provided for each of the units which will operate for
reservoir levels between Elevations 1455 and 1405. Each intake has a
single intake gate, a set of steel trashracks, and provision for placing a
bulkhead upstream from the gate as shown on Plate 83. A trave 1 i ng gantry
crane on the intake deck at Elevation 1466 w:ill service all four intakes.
r) The mechanical equipment i.s described in more detail below.
The intake is 1 ocated at the end of a 200-foot ... l ong un 1 i ned approach chan-
nel. The structure is founded deep in the rock. The rock 'face between in...,
takes will be lined with concrete to stabilize the rock surface. The. '9rOUt
curtain and drainage holes will continue beyond the main dam and beneath
. the structureQ
The 60-foot spacing of the intakes was set similar to the spacing of tJJe
turbines, which will give parallel penstocks and allow for easier setting
out ofthe tunnels during construction.
(b) Intake Gates
The four power intakes will have a single fixed wheel intake gate with a
nominal operating size of 16 feet wide by 20 feet high. The. gates \-till
have an upstream skinplate and seal 'and will be operated by hydraulic
or wire rope hoists located in heated enclosures immediately below deck
level. The gates will normally close under balanced head conditionsto
permit dewatering of the penstock and turbine water passages for t-urbine
inspection and maintenance. The gates \'lill also be capable of closing
under their own weight with fu11 flow conditions and maximum reservoir
water level in the event of loss of control of the turbines. A heated air
VP.nt will be provided at the intake deck to .satisfy air demand requirements
when the intake gate is closed with flowing water conditions.
13-19
(c) lntake Bu1khead Gates
. One se.t of lntake bulkhead,_ cons,isting of two ga.te sections will be provided
for closing the intaKe openings. The gate will be used to permit inspec-
tion and maintenance of the tnt.ake gate and intake gate guidese: ·The gates
will be raised and lowered under balanced ·water conditions only. To bal-
. -ance water :pressure when ra.ising the intake bulkhead, the space between the
gate and the downstream control gate will be flooded by means of a ·
follower-operated bypass valve on the top gate section·. An. air valve will
be provided in the top of the gate. The gates will have a downstream skin--
plate and seal on the downstream Side. They wi11 be designated to with-
Stilnd full differential pressure.
()
(d) Trashracks
Each of the four intakes· will incorporate trashracks just upstream from the
maintenance gates. The trashrack opening will have a bar spacing of about
6 inches and designed for a maximum differential head of about 30 feet.
The m~ximum gross velocity through the racks will be about 4 ft/s. Each
trashr ack wi 11 be constructed in two sections for remova 1 by means of a
follower suspended from the intake gantry crane•
(e) Intake Gantry Crane
-
A 60-ton capacity (approximately) electrical traveling gantry crane will be
provided on the intake deck at Elevation 1466 for handling the trashracks~
maintenance gates and intake g3tes. The crane will incorporate a double
point lift hoist mountedon an enclosed trolley. The hoist pickup \'lfll be
incorporated into a follower for handling the gates and trashracks. The
crane will also have a grappling hoist with a grapple of approximately
5-ton capac.ity for removing debris from in front of ~he trashrack.
13.11 -Penstocks
The power plant will have four penstocks, one for each unit. The maximum static
head on each penstock is 638 feet, measured from normal maximum operating level
(Elevation 1455) to centerline distributor level (Elevation 817). An allowance·
of 35 percent has been made for pressure rise in the penstock under transient
conditions, giving a normal maximum design head .of 861 feet. Maximum extreme
head, corresponding to maximum reservoir flood level, is 876 feet.
The penstocks have been designed as individual concrete-lined rock tunnels from
the intake to the powerhouse~· The section 200 feet upstream of the powerhouse
is steel lined.. The inclined sections of the concrete-lined penstocks are at
ss~ to the horizontal.
(a) Steel Liner
It has been assumed that the rock adjacent to the powerhouse cavern wi11 be
incapable of long-term restraint against the forces transmitted from the
penstock hydraulic pressures. The first 50 feet of steel liner will there-
fore be required to resist the design pressure without contributing from
the surroundfng concrete. For the remainder of the steel liner, which
extendes a further 150 feet upstream, allowance is made for partial rock
13-20
/ ·;
.. ~,
:1-J
. '1
·.JI
I
I
I
I
I
I
I
I
I
I
I ·'
' _. _:.
I
I'
I
I
••••
•••••• •••
• j
I
-I
. _>.
••
I
I
I
I
I
.I
I
I
I·
I
I
I
I
I
I
s.upport to .reduce the stee 1 stress. For pre 1 i mi nary design, it is ass.umed
that not more than 50 per.cent of the design head is taken by the .rock
support over this transition length •
. '
Beyond the steel liner, the hydraulic loads are taken solely by the rock
tunnel with a concrete. linE!r. . . ' ' . .
The steel liner is surrounded by a concrete infi11 with a minimum thickness
of 24 inches. A preliminary analyses has evaluated that the optimum inter-
nal diameter of the ·stee.l lining is 15 feet, based on the minimum total
cost of construction and the capitalized value of energy reduction due to
head loss. A tapering steel transition has been provided at the junction
between the stee 1 1 i ner and the concrete 1 i ner to increase the inter na 1
diameter from 15 feet to 20 feet.
(b) Concrete Liner
The penstocks are fully 1 i ned with concrete from the intake to the steel
lined section adjacent to the powerhouse; the thickness of lining will vary
with the design head. The minimum thickness of lining is 12 inches.
Based on preliminary analyses, the optimum internal diameter of the con-
crete liner is 20 feet.
(c) Grouting and Pressure Relief
A comprehensive pr"essure re 1 i ef system is required to protect the
unde.rground caverns against seepage from the high pressure penstocks. The
system will comprise of small diameter boreholes set out in patterns and
curtains ,to intercept the jointing in the rock. ·
.
Grouting round the penstocks will be provided to:
-fill and seal any voids between the concrete infill and the steel liner
which may be left after the concrete placing and curing; and,
-fill joints or fractures in the rock surrounding the penstock;} to reduce
flow into the pressure relief system and to consolidate the r.)ck.
13.12 -Powerhouse
{a) General
The powerhouse complex will be constructed underground in the right abut-
ment. This will require the excavation in rock of three maj'or caverns
(powerhouse, transformer gallery and surge chamber), with interconnecting
rock tunnels for the draft tubes and isola ted phase bus ducts.
An unlined rock tunnel will be required for vehicular access to the three
main rock caverns. A second unlined rock tunnel will provide· access from
the powerhouse-to the foot of the arch dam!.! for routine maintenance on the
fixed cone valves;~this tunnel will also provide construction access to the.
1 ower section of the penstocks •.
13-21
/
Verti,cal shafts wit 1 be required for personnel access by elevator to the
underg,round powerhouse; for oil filled cable from the transformer gallery~
and for surge chamber ventingG
The draft tube: gate gallery and cavern are 1 ocated in 'the surge chamber""
cavern, above maximum design surge level. ·
The general layout of the powerhouse complex is shown on Plates 85, 86 and
87. The transformer gallery is located upstream of the powerhouse cavern;
the surge chamber is located downstream of the powerhouse cavern. The
clear spacing between the underground caverns is at least 1.5 times the
main span of the larger excavation', from geotechnical considerations.
(b) Layout Considerations
The powerhouse is located underground in the right abutment. ~later for
power generation is taken from an intake structure to the right of the main
spillway, and carried through 1 ndi vi dua 1 penstocks to the turbines. \~ater
is discharged to the river by a single tailrace tunnel 6800 feet in length.
The draft tubes and tailrace tunnel are protected against excessive tran-
sient pressure rise by a downstream surge chamber, which also provides
storage for the turbine start-up sequence~
The intake structure is dP;signed for a maximum drawdown of 50 feet and is
located close to the mairt arch dam thrust block for ease of access. The
powerhouse·is located t:J provide the minimum total length of penstock, as-
suming an inclination of 55° to the horizontal for the sloping section of
penstock~ The orientation of the powerhouse has been selected as a compro-
mise between the desired orientation for power flow (E-W) and the geotech-
nical data on known shear zones and joint sets. Minimum clear spacing be-
tween major rock excavations is at least 1.5 times the span of the larger_
excavation. This is considered a conservative estimate for preliminary
design purposes.
The downstream surge chamber will· be constructed as close as possible to
the powerhouse for maximum protection to the draft tubes under transient
load conditions. For this reason the underground transformer gallery will
be located upstream from the powerhouse" The rock around the powerhouse
cavern and transformer gallery is protected against high pressure seepage
from the penstocks by a 200-foot continuous steel-lining and an extensive
pressure relief system.
(c) Access Tunnels and Shafts
Vehicle access to the underground facilities at Devil Canyon is provided by
two unlined rock tunnels. The main access tunnel, 3,000 feet long, con-
nects the powerhouse cavern at Elevation 852 with the canyon .access road on
the right bank. A secondary access tunne 1 runs from the main powerhouse
access tunnel to the foot of the arch dam, for .routine maintenance of the
fixed cone valves. A branch tunnel from the secondary access tunnel will
provide construction access to the lower, section of the penstocks, at Ele-
vation 820. Separate branch tunnels from the main access tunnel give
vehicle access to the transformer gallery at Elevation 896 and the draft
tube gate .gallery at Elevation 908. The maximum gradient on the permanent
13-22
0
I
I
I
I
I
I
I
I
I
I
1:
I
I .,
I
I
I
•••
t
I
I·
I
I
I
I
I
I
I
I
I
I
I
I
I
:1
I
I
(d)
access tunnel is 8 percent; the maximum gradient on the secondary acces.s
tunnel is 9 ·percent. · · .
The <:ross section of the access tunnels is dictated by requirements for
·con'StPuction plant; for preliminary desfgn a modified horses·hoe shape35-
feet wide by 28-feet high has been used ..
The main access shaft is lot:ated at the north end of the powerhouse cavern,
providing personnel access by elevator from the surfaceo · Horizontal tun-
nels are provided from this shaft for pedestrian access to the transformer
gallery and the draft tube gate gallery(> At a higher level access is also
available to the fire protection head tank.
Access to the upstream grouting gallery is from the transformer gallery
main access tunnel, at a maximum gradient of 13.5 percent.
Powerhouse Cavern _.;...a;;,..;;;.;;.;;.;;;..;;.....;;.;;...;..;;;;.;...;..;.
The main powerhouse cavern is designed to accommodate four vertical shaft
Francis turbines, in Jine, with direct couplitig to overhung generators.
Each unit is d~signed to generate 164 MW at 575-foot head.
The overall height of the cavern is governed by the physical size of the
turbine and generator, space requirements for miscellaneous equipment and
s~rvices, the design dimensions of the turbin~ draft tube, the overhead
travelling crane clearance and size, and the rise of the roof arch. The
selected unit spacing is 60 feet; in addition, a 110-foot service bas has
been allowed at the south end of the powerhouse for routine maintenance and
construction erection. The 1 ocal control room is 1 ocated at the nor·th end
of the main powerhouse flooro The width of the cavern allows for tne
physical size of the generator plus galleries for piping, air-conditioning
ducts~ electrical cab1es, and isolated phase bus. The penstock steel-liner
is continuous with the turbine spiral case; no penstock inlet valves al~e
provided in the po~1erhouse. Continuous drainage galleries are provided to
a low-level sump.
Compensation flow of 500 ft3/s is required to the river immediately
downstream of the arch dam, in view of the length of the tailrace. tuooel
( 6800 feet). This flow is provided by two No. 1300 h p vert i ca 1 shaft. imi xed
flow pumps, installed in a gallery below the service bay. Each pump is
rated at 115,000 gpm at 35-'foot h~ad. Water ·is taken from the base of the
surge chamber and pumped 1000 feet to the dam through a steel pipe 1ai<'1
partly in the secondary access tunnel and partly in a separate outlet
tunnel.
Multiple stairway access points are aV<:tlable from the pOwerhouse main
floor to each ga 11 ery 1 eve 1. Access to the transformEr ga 11 e~'Y from the
powerhouse is by a tunnel from the main access shaft or by a stairwaJ
through each of thr! four i so 1 a ted phase bus shafts. Access is a 1 so
avai'lab.le to the dt·~ft tube gate gallery by a t::~nnel from the main access
shaft. ·
A service elevator is provided for access from the service bay area on the
main floor to the machine shop~ and the pumping and drainage ga11eries on
the lower floors. Hatches have been PY"ovidtad through a11 main floors for
in$t.a.11 ati on and_ routine maintenance of pum}'s, valves and other heavy
• < .. equi pm~nt using the main overhead travell ino crane •.
(e) Transformer Gallery
The transformers are located underground in a separate unlined rock cavern,
120 feet upstream of the powerhouse cavern, with four interconnecting
tunnels for the isolated phase bus. There a1re 12 single-phase transformers
in four groups of 3.! one group for each generating unit. Each transformer
is rated at 13/345, 70 MVA. For increased rteliability, one spare transfor-
~r and one spare HV circuit are provided. The station service transfor-
r.~~~rs (2x2 MVA) and the surface faci1iti.es transformers (2 x 7.5/10 MVA) are
located in the bus tunnels. Generator excitation transformers are located
on the main powerhouse float"'.
High voltage cables are taken to the surface in t~'lto cable shafts, each 7
feet-6 inch internal diameter; provision is made for an inspection hoist in
eat;h shaft.
Vehicle ac;cess to the transformer ga.ll ery is from the south end vi a tbe
main powerh~use access tunnel. Personnel acc1ess is from the main access
sha?,; or through each of the four isolated phi3Se bus tunnels.
(f) Surge Chamber
A simple surge chamber has been provided 120 feet downstream of the power-
house to control pressure rise in the turbine draft tubes and tailrace
tunnel under transient load conditions, and to provide storage for the
machine start-up sequence. The chamber.is common to all four draft tubes
and the inlet pipe to the. compensation flow pumps. The chamber design is
gt>verned by an assumed full load rejection surge and the requiremen~ for
; nci pi ent stabi 1 ity under part load operation, together with estimated
floor levels from the tailwater rating curve.
The .:~raft tube gate gallery and crane are 1 ocated in the same cavern~ above
the maximum anticipated surge level. Access to the draft tube gate gallery
is by a rock tunnel from the main access tunnel. The tunnel is widened
locally for storage of the draft tube gates.
The chamger is generally an un 1 i ned rock ex~:avati on with 1 oca 1 i zed rack
support as neces~ary for stability of the roof arch and walls. The guide
b 1 ocks for the draft tube gates will be of reinforced concrete anchored to
the rock excavation by rockbolts.
13.13 -Reservoir
The Devil Canyon reservoir, at a normal operating level of 1455 feet, will be
approximately 26 miles long with a maximum width in the order of 1/2 mile.. The
total surface area at normal operating level is 7800 acres. Just upstream of
the dam, the ~aximum water depth will be approximate1y _feet. The min·imum
reservoir level will be 1405 fe9t during norma.l operation, resulting in a
13-24
,,
·I"
I
I
I
-I
I
I
I
I"
I
I
1-
·1
I
I
I
•••••• •••
'
I
I
I
I
I
I
I
I'
I
I
I
I
I
I
I
. I
maximl!m drawdown of 50 feet. The reservoir will have a· total capacity of
1,090,000 acre feet of whien 420,000 acre feet will be 1 ive storage.
Prior to reservoir filling, 'the area below elevation 1460, five feet above maxi-
mum operating level, will be cleared of all trees and brush, ·A field reconna-is-
sance of the proposed reservoir area was undertaken as part of these studies.
This work inc 1 uded ex.ami nation of aer i a 1 photographs and maps; an aer i a 1 over-
flight of the reservoir and collection of recent {1980 field season) forest
inventory data from the U~ s. Forest Service. As described for the Watana reser-
voir, most of the vegetatal material within the reservoir consists of trees with
very little in brush. The trees are quite small, and the stands are not very
dense. In the Watana reservoir area, an es-timated 3,200,000 cubic feet of \'load
exi.sts averaging approximately 500 cubic feet of low co11111ercial quality, and
some very significant logging problems would be posed by the ste~p slopes and
incised terrain excavated. Approximately 87 percent of the available timber are
soft woods.. The results of the timber reconnaissance studies are described in
more detail in Appendix C3.
The combination of steep terrain, moderate-light tree stocking levels, small
trees, erosive potential of the reservoir slopes, remoteness, and very restrict-
ed access to the reservoirs are major factors affecting the choice of harvesting
systems to be utili.zed for this project as discussed in Section 13.14.
Present market d~mand for the timber at Susitna is low, however, theworldwicie
demand for wood fluctuates considerably. It is anticipated that use of the har-
vested material would be limited to either sale as wood-waste products and as
fuel.
Slash material including brush and small trees, which will be suitable for
either of the above uses, will be either burned in a carefully con· ""Olled manner
consistent wth applicable laws and regulations, or hauled to a dis~~sal site in
and adjacent to the reservoir. Material placed in disposal areas will be cover-
ed with an earthfill cover sufficient to prevent erosion and subsequent expo-
sure.
A number of unstable areas will undoubtedly result during reservoir operation.
These areas will require remedial t~eatment depending on the nature and extent
of the instab~lity. ·
13.14 -Tailrace Tunnel
?
The tailra~e pressure tunnel is provided at Devil Canyon to carry power plant
discharge from the surge chamber to the river. The tunnel has a modified horse-
shoe cross-secti.on with a major internal dimension of 38 feet, and for prelimin-
ary des·fgns it is assumed to be concrete lined throughout with a minimum thick-
ness of 12 inches. The length of the tunnel is '6800 feet ..
The size of the tunnel was selected after an economic study of the cost of
construction and the ·capitalized value of average annual energy losses caused by
friction, bends and changes of section. Since the size of the surge chamber is
related to the effective diameter of the tailrace tunnel, the cost of the surge
chamber was also included in the optimization studies as a function of tunnel
size •
13-25
" ~.-····~ w •• -•
The tailrace portal site has been located at a prominent steep ;,~ock face on the
right bank of the river to provide the required tunnel :cove.r (about 60 feet) in
as short a di,stance as ·po.ssible. The portal provides a gradual transition from
the tunnel modified horseshoe shape to a rectangular cross-section at the.
outlet; it also reduces the maximum outlet velocity to 8ft/sec_, to reduce the
_velocit-y head loss at exit.. Vertical stoplog· guides·-·are provided for closure of
the tunnel if required for tunnel inspection and/or maintenance.
13.15 -Turbines and Generators
(a) Unit Capacity
The Devi 1 Canyon powerhouse wi 11 have four generating units with a nominal
capacity of 150 MW. This is the available capacity with minimum December.
reservo~r level (El. 1393) and a corresponding gross head of 553 fe"et in
the station.
The head on the plant will vary from 605 feet maximum (597 feet net head)
to 550 feet minimum (538 feet net head). Because maximum turbine output
varies approximately with the 3/2 power of head, the maximum unit output
will change with head as shown in Figure 13.2.
The rated head for the turbine has been established at 575 feet, which is
the weighted average operating head on the station. Allowing for generator
1o.sses, this results in a rated turbine output of 225,000 hp {168 MW) ..
The generator rating has been se 1 ected as 180 MVA with a 90 percent power
factor, which corresponds to a power output of 162 MW. The ~enerators will
be capable of continuous operation at 115 percent rated power. Because of
the high capacity factor for the Devil Canyon station, the units will be
operated at or near full load a 'I arge percentage of the time. The genera-
tors have therefore been sized on the basis of maximum turbine output at
maximum head, allowing for a possible 5 percent addition in power from the
turbine. This maximum turbine output (250,000 hp) is within the continuous
overload rating of the generator.
(b) Turbines
The turbines wi l1 be of the vertical shaft Francis type with steel spiral
casing and a concrete elbow type draft tube. The draft tube will have a
single w,ater passage (no center pier).
rne rated output of the turbines wi 11 be 225,000 hp at 575 feet rated net
head. Maximum and minimum heads on the units wi 11 be 597 feet and 538
feet~ respectively. The full gate output of the turbines wi 11 be about
240,000 hp at 597 feet net head and 205,000 hp at 584 feet net head. Over-
gating of the turbines may be possible, providing approximately 5 perct:nt
additional power. For preliminary design purposes, the best efficiency
(best gate) output of the units has been assumed at 85 percent of the full
gate turbine output. This wi11 be reviewed at the time of preparatiar• of
bid documents for the turbines.
J3-26
.·, 1 .. · '
. ' ~
-, .-· ..•. _·.-·
I
I
••
"I
I
I
I'
I
-1
I
1:
I
I
I
I
I
~
I
I
I
I
I
-I
I
I
I
I
I
••
~·
I
I
'I
I
I
I
The full gate and best gate· efficiencies of the turbines will be about 91
percent and 94 percent, respectivel.y, at rated head. The efftciency wi 11 .
be about 0.2 percent lower at maximum head and 0.5 percent lower at minimum
head. The pre.liminary performance curve· for the turbine is shown in Figure
13.3 ..
A speed of 225 rpm has been selected for the unit for preliminary design
purposes. The resulting turbine specific speed (N 5 ) is 37.9. As shm·m
in Figure 12.23, this is wtthi n present day_ practice for turbines operating
under 575 feet head. The considerations for selection of turbine speed are
briefly discussed in Section 12"16.
On the basis of information from turbine manufacturers and the studies on
the power plant layout, the centerline of the turbine distributor has been-
set at 30 feet below minimum tailwater level. The final setting of the
unit will be established in conjunction with the turbine manufacturer after
the contract for the supply of the turbine equipment has been awarded.
The mechanical/structural designs of the turbines wi 11 be basically the
same as for Watana. Because of the relatively short penstocks and the
surge tank location trrmedi ately downstream from the powerhouse, the hydrau-
lic transient characteristics of the turbines are favorable. Assuming nor-
mal gen-erator inertia (H = 3~5 MW-Sec/MVA), a preliminary analysis has in-
dicated the following:
-Water starting time (Tw) ............................. 1.2 sec.
-Mechanical starting time {Tm) •• ,. •...••.•..•.••..•.. 7.6 sec.
-Regulating ratio (Tm/Tw) ........................... 6.3
-Governor time.-....................................... 5.0 sec ..
-Speed rise on full load rejection . • . • • • • . • .. • • . • • . . 35 percent
-Penstock pres~ure use on full load rejection ••••.•• 20 percent
The regulating ratio is above the minimum recorrmended by the USBR for good
regulating. Also~ unit speed rise and penstock capacity pressure ri:se are
within normal accepted values. Because of the relatively short distance
between the turbine and the tailrace surge tank and the deep unit setting,
there should not be any problems with dr·aft tube column separation.
As discussed in Section 12.16 for Watana, the units will be capable of op-
eration from about 50 to 100 percent load. Considerations for draft tube
surges and corresponding power swings as mentioned for Wataoa also will
apply to Devil Canyon.
As with Watana, the relationship between generator natural frequency and
the possi b 1 e draft tube surge frequency is desi rab 1 e and wi 11 require study
in later design stages. 'Becaase of the high capacity factor for the. Devi 1
-Canyon units~ part load operation for these turbines is not as critical as
at Watana; therefore, the possibility of problems with power swings is
somewhat less of a concern than at Watana.
(c) Generators
The four generators in the De vi 1 Canyon powe.rhouse wi 11 be of the vertical
shaft, overhung type directly connected to the vert·i ca.l Francis turbines.
13-27
(d)
The generators will be similar in construction and design to the Watana
g~nerators and the gener.al features described in Section 12.16 for the
stator, rotor, excitation system, and other details which apply for the
Devil Canyon generators.
The rating and characteris:ties of the generators are as· follows:
Rated CapacitY:
Rated Power:
Rated Voltage:
Synchronous Speed:
Inerti-a Constant:.
Short Circuit Ratio:
Efficiency -at Full Load:
Governor System
180 MVA, 0.9 power factor with overload rating
of 115 percent.
162 MW
15 kV» 3 phase, 60 Hertz
225 rpm
3.5 kW -Sec/kVA
1.1 (minimum)
98 percent (mini mum)
A governor system with electric hydraulic governor actuators will be pro-
vided for each of the Devi 1 Canyon units. The system wi 11 be the same as
for Watana.
13.16 -Miscellaneous Mecryanical Equipment
(a) Compensation Flow Pumps
The two pumps for providing minimum discharge into the Susitna River be-
tween the dam and the tailrace tunnel outlet portal wi 11 be vertical mixed
flow or axial type located in the powerhouse service bay below the main
a erection floor, ~s shown on Plate 87. Each pump will be rated at 250 cfs
(115~000 ga 1/min} at 35 feet total head, and wi 11 be driven by 1,400-hp
induction motors. The preliminary pump and motor data is summari.zed in
Table 13.3.
A single pump intake will be ··located in the surge chamber with an 8-foot·-
dlameter intake tunnel leading to the powerhouse. The intake tunnel will
bifurcate into individual pump intake conduits within the powerhouse. The
pump discharges wi 11 converge into a single pump discharge tunne 1 ..
Butterfly type valves wi 11 be installed in the intake-and discharge 1 ines.
of e.ach pump to permit isolation of a pump for inspection or malntenance.
Trash screen guides and a trash screen wi 11 be provided in the surge cham-
ber at the pump intake. It will be possible to remove the trash screen us-
ing the draft tube gate crane discussed below. The width of the guides
. 13-28
I
I
I
'
I'
I
I
I .,
I
I
I
I
I
I!
I
••
I
I
II
I
I·
I
I
I
I
I
I
I
I
I
I
I
I
I
I
••
I
· .. I
will be select-ed so that one of the turbine draft tube gates may be
installed in the intake to permit_ d~wateri ng the pump intake tunne 1 for
inspection and/or maintenance .of the tunnel or the intake butterfly valves.
Stop 1 og guides and a set of stop 1 ogs will a Tso be provided at the down-
stream end of the pump discharge tunnel to allow.the discharge tunnel to be
dewat.ered. The stoplogs: will be handled with a mobile crane and a
follower.
Pumping operation wi 11 be continuous; therefore, pumping equipment wi 11 be .
conservatively designed to provide efficient operation with minimal main-
tenance. Crane access will be provided. for the pumps, motors, and valves
to permit equipment servicing.
In the detailed design stages, consideration should.also be given to
turbine-driven rather than electric motor-driven pumps. A header from at
least two of the main turbine penstocks would supply water to the turbines,
with the turbine draft tubes connected to the pump discharge.
{b) Powerhouse Cranes
.
Two overhead type powerhouse cranes will be provided at Devil Canyon as at
Watana. The estimated crane capacity will be 200 tons.
(c) Draft Tube Gates
·Draft tube gates wi 11 be provided to permit dewatering of the turbine water
passages for inspection and maintenance of the turbines. The arrangement
of the draft tube gates will be the same as for Watana, except that only
two gates will be provided, each 21 feet by 21 feet. At the time of start-
ing of Unit l, one gate will be installed in Unit 4 with the other gate
available for Unit L. Bulkhead domes.wil'! te installed in Units 2 and 3.
{d) Draft Tube Gate Crane
A crane will be installed in the surge chamt:-er for installation and removal
of the draft tube gates. The crane will be either a monorail (or twin
monorail) crane or· a gantry crane~~ For the preliminary design, a twin
monorail crane of approximately 25-ton capacity has been assumed. The
crane will be pendant-operated and have a two point lift. A follower will
be used with the crane for handling the gates. The crane runway will be
located along the upstream side of the surge chamber and will extend over
the intake for the compensation flow pumps, as well as a gate unloading
area at one end o·f the surge chamber.
(e) Miscellaneous Cranes and Hoists
In addition to the powerhouse cranes and draft tube gate cranes, the fol-
lowing cranes and hoists will be provided in the power plant:
- A 5-ton monora.il hoist in the transformer gallery for transformer mair~
tenance;
,_Small overhead jib, or A-frame type hoists in the machine shop for
handling material; and
13-29
,I
-·A-frame or monorail hoists in other powerhouse areas for handling small ··1·
equipment.
(f) Elevators
. Access . and service eleva tors will be provided for the power p1 ant as
follows:.
... Access elevator from the control building to the powerhouse;
-Service elevator in the powerhouse service bay; and
-Inspection ho4sts in cable shafts.
The elevators will be as discussed in Section 12.17 for Watana.
(g) Power Plant Mechanical Service System~
The power plant mechanical service systems for Devil Canyon will be essen-
tially the same as discussed in Section 12.17 for Watana, except for the
following:
-There will be no main generator breakers in the power plant; therefore,
circuit breaker air wi 11 not be required'!t The high-pressure air system
wi 11 be used only for governor as we 11 as instrument .air. The operating
pressure will be 600 to l,OOOpsig depending on the governor system oper-
ating pressure. An air-conditioning system will be installed in the
powerhouse control room.
-For preliminary desigrJ purposes only, one drainage and one dewatering
sump have been provided in the powerhouse. ihe dewatering system wi11
also be used to dewater the intake and discharge lines for the compensa-
tion flow pumps.
(h) Surface Fac_ilities Mechanical Service Systems
The entrance building at the top of the power plant wiil have only a heat ...
ing and ventilation system. The mechanical services in the standby power
building will include a heating and ventilation system, a fuel oil system,
and a fire protection system, as at Watana.
(i) Machine Shop Facilities
A machine shop and tool room will be located in the powerhouse service bay
area to take care of maintenance work at the plant. The facilities will
not be as extensive as at Watana. Some of the" larger components will be
transported to Watana for necessary machinery work.
13.17 -~cessory Electric.~) Equipment
{a) General
.' !
The accessory electrical equipment described in this section includes the
following main electrical equipment:
13-30
I .
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
¥ ~ ~
I
I
I
I
I
I
I
I
I
I
••
I
I
I
_I
I
I
"(b)
0
.. Main generator step-up 15/345 kV t~ansformers;
-Isolated phase bus connecting the generator and transformer"S;
-345 kV oil-filled cables from the transformer terminals to the switch-
-yard;.
-Control systems; and
-Station service auxiliary ac and de systems.
Other equipment and systems described include grounding, lighting system
and communications.
The mai-n equipment and connections in the power plant .are shown in the
Sing·le Line Diagram, (Plate 88}. The arrangement of equipment in the
PO\'Ierhouse, transformer gallery, and cable shafts is shown in Plates 85 to
88.
General Design Censiderations for
Transformers and HV Connections
(i) General
Twelve single-phase transformers and one spare transformer will be
located in the transformer gallery. Each bank of the three single-
phase transformers will be connected to one generator by i so 1 a ted
phase bus located in bus tunnels. The HV terminals of the transfor-
mer will be connected to the 345 kV switchyard by 345 kV single-
phase oil-filled cables installed in 800-foot long vertical shafts.
There will be two sets of three single-phase 345 kV oil-filled
cables installed in each cable shaft. One additional set will be .
maintain2d as a spare three-phase cable circuit in the second cable
shaft. These cable shafts will also contain the control and power
cables between the powerhouse and the surface control room, as well
as emergency power cables from the d·iesel generators at the surface
to the underground facilities.
As described·in Section 12.18 for the Watana power plant, a number
of considerations led to the choice of the above optimum system of
transformation and connections. Different alternative methods·and
equipment designs were also considered. In surrmary, these are:
-One transformer per generator versus one transformer for two gen~·
erators;
-Underground transformers versus surface transformers;
.. Direct transformation from generator voltage to 345 kV versus
intermediate step transformation to 230 kV or 161 kV3 and thence
to 345 kV;
-Single phase versus three-phase transf()rmers for each alternative.
method considered; and
-Oil-filled cable versus solid dielectric cable or.SF6 gas-
insulated bu~. ·
Reli.ability considerations are based on the general reliability
requirements for generation and transmission described in Section 15
regarding ·the forced outage of a sing)e generator, transformei"', bus
or cable in addition to planned or scheduled .QUtages in a single
13-31
(c)
(d)
1.. , ... , ..... .
contingency_situation,~ or a subsequent outage of equipment in the
double contingency sitU.ation. In the first case~ the system should
be capable of readjustment after the outage for loading 'llithin
normal ratings and, in the second case,-within emergency rati n~s.
The one transformer per generator scheme was selected since. the __
operation of the Devil Canyon power plant wi11 e-ssentially -be a con-
tinuous base-load type operation; also the smaller riu.--nber of units
at Devil Canyon compared to Watana will allo~l a transformer ga11ery
of reasonable length for a unit generator-transformer scheme.
A~ at Watana, transport limitations for both dimensions and weight
wi 11 preclude the use of the 1 arger size three-phase transformers;
hence, single-phase transformers will be used. One distinct advan-
tage of single-phase transformers is that a spare transformer can be
provided at a fairly low incremental cost.
For the same reasons as given in Section 12.18 for Watana, surface
transforme\ ... S and the double-step transformation scheme {15/161 kV
generator-transformer, 161 k V cab 1 e and 161/345 k V auto-transformer
at the switchyard) were ruled out. The direct transformation
(15/345 k\1) scheme with 345 kV oil-filled cables is considered a
better over a 11 scheme.
Main Transformers
The transformers \'lill be of the single phase, two-winding, oil-immersed,
forced-oil water-cooled (FOW) type. A total of twelve single-phase
transformers and one spare transformer wi 11 be provided, with rating and
characteristics as 'follows:
Rated capacity:
High Voltage Winding:
Basic Insulation Level
(BIL) of HV Winding:
Low Voltage Winding:
Transformer Impedance:
70 MVA
345/ 3 k V, .grounded Y
1300 kV
15 k V, Delta
15 percent
The design and construction details are identical to the transformers at
Watana as described in Section l2a18.
Generator Isolated Phase Bus
Isolated phase bus connections will be located between the generator and
the main transformer. The bus will be of the self-cooled, welded aluminum
tubular type with design and construction details g2nerally similar to the
bus at the Watana powr~r plant. The rating of the main bus is as follows:
Rated current:
Short circuit current
momentary:
Short circuit current
symmetri ca 1 :
Basic Insulation Level
(BIL):
13-32
9,000 amps
240,000 amps
150,000 amps
150 kV
I
I
I
I
I
I
I
I
I
I
I
••
I
I
I
I
I
I
I,
I
I
,~I
a·
I
I
I
I
I
I
I
I
I
I
I
I
I·
I
I
I
.,: ::.-,.··
(ia) 345 kV Oil-Filled Cable
The general design considerations leading to the choice of the 345 kV Oil-
filled cable for the connections between the transformer HVterminals and
the 345 kV switchyardat the surface are the same as described in Section .
12.18 for the Watana plant.
The cables will be rated for a continuous maximum current of 400 amps at
345 kV ±.5 percent. The cables will be of single-core construction with oil
flowing through a central. oil duct within the copper conductor. The cable.s
will be installed in the 800-foot cable shafts from the transformer gallery
to the surface. No cable jointing wi1:1 be necessary for this installation
length.
(f) Control Systems
( i} Genera 1
The Devi1 Canyon power plant \<if 11 be designed to be operat@d as an
unattended plant. The plant will be normally controlled through
supervisory control from the Susitna ftrea Control Center at Watana~
The plant will, however, be provided with a control room with suffi-
cient control, indication, and annunciation equipment to enable the
p 1 ant to be operated during emergencies by ore operator in t\ie con-
trol room~ In addition, for the purpose of testing and commission-
ing and maintenance of the plant, local contr·ol b()ards will be
mounted on the powerhouse fl oar near each un~~t.
Automatic load-frequency control of the four units· at OeviT Canyon
will be accomplished through the central computer-aided control
system 1 ocated at the Watana Ar'~t:t ;Control Center.
The power plant will be provh~t:.'.J with 11 black start11 capability
similar to that provided at Watana, to enable the start of one t-.nit
without any power in the powerhouse or at the switchyard·, except
that provided by one emergency diesel generator. After the. start-up
of one unit, auxiliary station service power will be established in
the power plant and the switchyard; the remaining generators can
then be started ant: after the other to bring the plant into full
output within the hour. ·
As at the Watana powe~ plant, the control system will be designed to
permit local-manual or local-automatic starting, voltage adjusting,
synchronizing, and loading of the unit from the powerhouse control
room at Devil Canyon.
The protective relaying sy.stem is shown in the main single line
diagram, Plate 88, and is generally simi 1 ar to that provided for the
Watana power plant.
13-33
(g)
,;;.·
I_
(h)
----..,;:-;:-:-.--;:. ,-:· --·-.
Station Service Au xi. 1 i ary AC and DC Systems
( i ) AC Auxi 1 i ary System
The station service system wi 11 be desi ~ o-· achieve a r·e li ab 1 e
and economic distribution system for the r plant .and the switch-
yard and surface facilities. The auxiliar.;' system will be similar
to that in the \~at ana power plant except that the swi tchyard and
surface faci 1 it i es power wi 11 be obtai ned from a 4.16 k V system
supplied by two 5/7.5 MVA, OA/FA, oil-immersed transformers connect.-
ed to generators Nos. 1 and 4, res;:H~ctively. The 4.16 l<V double-
ended switchgear will be located in the powerhouse. It will have a
normally-open tie·breaker which will prevent parallel operation of
the two sections. The tie breaker will close on failure of one or
the otJer ·of the incoming supplieso The 1400 hp compensation f1ow
pumps \\\ill be supplied with power directly from the 4.16 kV system.
Two 4· .. J5 cables installed in the cable shafts will supply power to
the St'i"face facilities.. ·
The 480 V station service system will be exactly similar to the
Watarta system described in Section 12.18, and will consist of a main
480 V switchger.r, separate avxi 1 i ary boards for each unit, an essen-
tial auxiliaries board, and a general auxiliaries board. The main
480 V switchgear will be supplied by two 2000 kVA, 15,000/480 V
grounded v-1ye sealed gas dry-type transformers. A third 2000 kVA
transformer will be maintained as a spare.
Two emergency diesel generators, each rated 500 kW, will be
connected to the 480 V powerhouse main switchgear and 4.16 kV
surface switchbo.ard, respectively. Both diesel generators will be
located at the surface.
An uninterruptib le high-security power supply wi 11 be provided for
the supervisory computer-aided plant control systems.
(ii) DC Auxiliary Station Service System
The de auxiliary system will be similar to that provided at the
\~atana plant and \'lill consist of two 125 V de lead-acid batteries.
Each battery system wi11 be supplied by a double rectifier charging
system. A 48 V de battery system wi 11 be provided for supplying the
supervisory a!'ld communications systems.
(iii) B1 ack Start Capabi 1 ity
As at the Watana power plant, the Devil CatJ.YOn power plant wi 11 be
provided with "black start 11 capability which wi11 enable the plant
to start un in a completely "blacked out" condition of the power
plant and/or the power system.
Other Accessory Electrical Systems
The other accessory electrical systems including the grounding system,
1 ignti ng system, and powerhouse communications system will be similar in
general design and construction aspects to the: systems described in Section
12.18 for the Watana power plant.
13-34
I
I.
.I
;-;
I
I.
I
I
I
I
I-
I
I
I
I
I
I
J.
I
,Ji
-'--' .I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
.I
I
13.18 -Swi tchyard Structure.s and Equipment
To follow
13.19 -Project Lands
Project lands acquired for the project will be the minimum necessary to
constru<. ~ ·access and site faci li ties, construct permanent faci li ties, to c 1 ear
the reservoir, and to operate the project ..
Appendix C contains 1 and status background information re 1 ati ve to the Susi tn a
Project, together with an inventory of private and public lands required for the
project. A 1 arge amount of pub 1 i c 1 and in the Devi 1 Canyon area is managed by
the Bureau of Land Management. There are large blocks of private Native Village
Corporation tands along the river. Other private holdings consist of widely
scattered remote parcel5. The state has selected much of the federal land in
this area and is expected to receive a patent.
0
13-35
:; ·,
TABLE 13.1: WATANA PE!~K WORK fORCE AND CAtllP/VILLAGE DtSIGN POPULATION
Calendar Year-
1992
1993
199~.
1995
1996
1997
1998
1999
2000
Yearly Peak Force
180
730 ,, '
1635
2455
3180
3180
2000
770
455
Camp/v~Ilage Oes~gn
200 aoo
1800
2700
3.500
3500.
2200
850
500
------------------------------------------------------------
I
I
I
I
I
I
I
I
I
I
'I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
···1· .. -
Dam -
loguiri
(1985)
Vaiont
(1961)
Mauvoisin
(1957)
Chirkei
(1975)
El Ca1on
(1964
Contra
(1965)
Glen Canyon
(1964)
Mohamed Reza
Shah Pahlavi
(1963)
Almendra
(1910)
Vidraru-Arges
Gocekaya
Morrow Point
Pacoima
TABLE 13.2: ARCH DAM EXPERitNCE
Locatiort
Height
ft(tn)
Crest length
ft(m)
""' l ...
Georgia, 892 2,513 USSR. (272) (766)
Veneto, 858 624
Italy (262) (190)
Valais, 777 1,706
Switzerland (237) (520)
North Caucusas, 764 1,109
USSR (233) (338)
Yoro/Cortes, 741 1,253
Honduras (226) (382) .
Ticino, 722 1,246
Switzerland (220) . -{380)
Arizona, J10 1,560
USA. (216) (475)
Khouzestan, 666 696
Iran (2(13) (212)
Salmanca, 662 1,860
Spain (202) (567)
Romani·~ 548 588
(167) (292)
Turkey 521 1,620
(159) (494)
Colorado 465 720
California 372 589
(113) {180)
TABLE 13.3: COMPENSA TIO~ FLOW PlffP DATA
'vertical, axiat1 or
mi)(ed flow
Rated head (total dynamic level) .,. ........................ o 35 ft
Rated discharge ·······~······~····••••••••••••••••••••••• 115,000 gal/min
Pump input !It ••• ~ ••••••• "' ........... c ••••••• ~--§ •••••••••••• •--• 1 ,300~ hp
Speed ••••••••••••• , ••••.•••• if~ •••••••.•• ~ ••••••••••.•••••••• •~ • 400 r_pm
In.,eller diameter ......................................... 51 in (approx.)
H~tor
Type • • • • • • • • • • • • • • • • • • • • • • • • •. • • • • • ... • • • • •. • • • • • • • • • • • • •.• • • verticaJl induction
Rated power ••••••••••••••••••••••• o...................... 1.,400 hp
Speed ••••• -••• • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • 400 .rp~n
Voltage ••o•••·e••••••••••··--·•••••••••••••••••••••-.e••••••• .4,160 V
~~o. phases •••••.• ~ ••••••••• ·• o ••••••••••••••••••••••••••• c .. 3
F~equency ••••••••••••••• -....................... '81 •• ,. • • • • • • • • • 60 hz
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I'
I
I
I
I
I
.I
I
I
I
I
I
I
TABLE 13.4: PffELIMINAAY UNIT DATA
1 GENERAL DATA
2
'Number ·of units e&e't99e•••Gee.eo•e.eO•ea•.;:,eeil!•e•4t•·••••..tf'•
Nominal unit output ••••••••••••••••••••••••••••••••••••
Headwater levels
.. normal maxif11Jm ••••••••••••••••.••••••• , ........... ~· ..... .
-miniml..ID * • ,. .•••••••.•• ~ .••••••••.•• • ••••.••••••.•.••• .-•••.••••
Tailwater levels
minimum ............................. • •••••••••••••••••••
normal •• & .......... d .................... e • a •••••• 0 , ••••••
maximum •••••••••• ~ .... ft •••• ·e ........................... 9
TURBINE DATA
4
150 MW
El 1445
El 1390
El 860
EL 840
El 838
Type· • •••••••••••••••••••••.• ~ •••••• _. ••••.•••• q. ~~~ • • • • • • • • • vertical .·Francis
Rated net head· • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • .• 5 1S ft
MaxitnUm head •••••• a ...................................... ~ 597 ft
Minimllll head •••••••••••••• ·• ................ c........... .. . 538 ft
Full gate output:
at rated level •••••••••••••••••••••• ., .................. 225,000 hp
-at maximum 'head ....................... o••••••••••••••• 240,000 hp
-at minimum head .......................................... 205,000 hp
Best gate output ........................................... 85% full gate output
Full gate discharge at rated head ........................ 3,790 cfs
~eed ............................... 10 •••.•••••••.•••• ~ • • • • • 225
Specific speed •••••a•••••••••·GJ~··••••••••••••••••••••• 37.9
Runner discharge diameter ................................. 135 in
Runaway speed ••••• ~ ....................................... , ••• c. 395 rpm
Centerline distributor ............................... 4 •• 820
Cavitation coefficient (sigma) •••••••••••• ,. •• o••••••••• 0.089
3 -GENERATOR DATA
Type •• ~· ••••••••••••• ~ •••••••• ~ ••• ••.• •••••••••••• e •••.•.•• ve·rtical·modified
l.lllbrella
R~ted output •••••• Ct •••••••••••••••• e •• , • • • • • • • • • • • • .• • • • 180· ·M't*A·
Power factor· •••••••.a••••••••••·•••••o•••····•••••-e••••••• 0.9
Voltage • • • • • • • .. • • • • • • • • • • •. • • • • • • • • • • • • • • • • • • • • .• • • • • • • • • 15 kV
Inertia constant (H)* ••••••••••••••••••••••••••••• .. •••• 3.5 MW-sec/MVA
Synchronous.speed ·~s••••••••••••••••••••••••••••••••••• 225 rpm 6 2 Flywheel effect (WR )* •••••••••••••• .. •••••••••o•••••••• 54 x 10 lb-ft
Heaviest lift ••••••••••••••••••••••••••••o••••••••••••• 750,000 lb
* Including turbine
"
50 --. . .
"""""""'
~
~· 30
u
< . ~ ~
-.
~ .
,. ~
~ ,,""'
..tt# .. -,-
~ ~""" . .
' lO •
I
'
eao -890 900 910 920 8!0 140 960
HEADWATER EL:EVATION (FT.)
. I
DEVIL ·CANYON DIVERSLON
· · RATING CURVE
FiGUR!:; i!.l ]i}
~--------------------------------------------------------------------------------··
I
I
I
I
I
I
I
I
I
:I
I
I
I
I
I
I
I
=~·
!~' ' ,_,
'
580
540
520
100
--.
-
usc lo GENERATOR
RA1 'POWER -
.0
-
RESERVOIR EL. 1455 I ----;' I -
I IL WEIGHl .J ED AVERAGE H ~AD
BES r GATE 7-I
/ J ,,
~ GENERAl PR RATED POWER ...-
I I BEST EFFIC 'f.NCY-j FULL GATE !
"""'IR /e. 1400 MINIMUM DEC fs::ur:u::~ HEAD HI l---
-1!5 ) MW
'
.
120 140 160 180 200 220
UNIT OUTPUT-MW
I}
DEVIL CANYON..;.. UNIT OUTPUt
FlGURE' l3.2
~''("';~
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
t'","
I
I
:.,:·
-?fl. -·~
Z" t.&J•
• "· • ~ • •• .... • ' •••• > •
Q . L~----~~--------~--------~------~---------r--~4000 b: 80-
"-1.5.1
lU z
ai
§
}--en
"" 10 1-------+-----t-----+--~;c_.-+-----+----t 3000 8
~~-----+--------~--------+-------~--------~--~1000
40,000 80,000 120,000 160,000 200,000
TURBINE OUTPUT {HP)
DEVIL CANYON -TURBINE PERFORMANCE
(AT RATED HEAD)
240,000
FlGURE · 13 .. 3 ..
1&.1 .(!)
cr
·<~;·
·~
Z5
I
I
I
~,I
I
I
••
I
I
I
I
~I
I
I
I
I
I ' .
....,.,.---,~.-------;:.,-:-·--:--·--:;·· •.. , ·~-' _, --;~-_,-~ ~ -.-t·-· ·_ .. -
. ·'
14 -TRANSMISSION FACILITIES
Th~ objectiv~ of this secti.on is to record and describe the studies performed to
.select a power delivery system from the susitna River basin generating plans to
the :ffiaj or load centers in Anchorage and Fatv .. ba.nks. This syst.em wi 11 be com ...
prised of transmission lines, substations, a dispatch center, and means of com•
munications.
The major topics of the tra.nsmission studies include:
-Electric system studies;
-Transmission corridor selection;
-Transmission detailed route selection;
-Tr ansmi ss ion towers, hardware and conductors;
-Substations; and
-Dispatch center and communications.
Further discussion of the importance of these studies in determining the method
of operation of the Railbelt System is presented in Section 15.,
In the following text, each of the major topics will be discussed with respect
to previous studies, methodology~ additional data obtained, and conclusions
arising from the studies.
14.1 -llictri c. System Studies ,,
Transmission planning criteria were developed to ensure th.e design of a re1i.·:able
and economic electrical power system, with components rated to allow a smooth
transition through early project stages to the ultimate developed potential.
Strict application of optimum, long-term criteria would require the installation
of equipment with ratings larger than necessary ,at excessive cost. In the
· interest of economy and long-term system performance, these criteria were tem-
porarily relaxed during the early development stages of the project. Although
allowing for satisfactory operation during early system development, final
system parameters must be based on the ultimate Susitna potential.
The criteria are i·ntended to ensure maintenance of rated power flow to Anchorage
and Fairbanks during the outage of any single 1 i ne or transformer element. The
essential features of the criteria are:
Total power output of Susitna to be delivered to one or two stations at
Anchorage and one at Fairbanks;
-"Breaker-and-a-half~' switching station arrangements;
-Overvoltages during line energizing not to exceed specified limits;
-System voltages to be within established-limits during normal operation;
... Power delivered to the loads to be maintained and system voltages to be kept
within established limits for system operation under emergency conditions;
14-1
-·:..-
.p-,,
I .
f
..
~ •.. -
-Transient stability during a j-phase line fault cleared by breaker action with
no r~closing; and ·
-\~here performance li.mfts are exceeded, the most cost effective corrective
measures are to be taken.
(a) Exi sti ns Ststem Data
Data have been compiled in a draft report by Commonwealth Associates Inc.,
dated November, 1980, and entitled •tAnchorage-Fah·banks Transmission Inter-
tie -Transmission System Data". The contents of this report have be.en
included, with minor revisions, as Appendix B of the Susitna Hydroelectric
Project Planning Memorandum-Preliminary Transmission System Analysis (1).
Other system data were obtained in the form of single-line diagrams from
the various utilities.
(b) f._ower Transfer Requirements
The·Susitna transmission system must be designed to ensure the reliable
transmission of power and energy generated by the Susitna Hydroelectric
Project to the load centers in the Railbelt area. The power transfer re-
quirements of this transmission system are determined by the following fac-
tors:
-System demand .at the various load centers;
-Generating capabilities at the Susitna project; and
-Other genet~ation available in the Rai lbelt area system.
Most. of the electric load demand in the Railbelt area is located in and
around two main centers: Anchorage. and Fairbanks. The largest load center
is Anchorage, with most of its load concentrated in the Anchorage urban
area. The second 1 argest 1 oad center is Fai rbank.s. Two sma 11 1 oad centers
(Willow and Healy) are located along the Susitna transmission route. The
only other significant load center-s in the Railbelt region are Glennallen
and Valdez. However, their combined demand is expected to be less than 2
percent of the total Railbelt demand in the foreseeable futureo A survey
of past and present load demand levels as well as various forecasts of
future trends indicates these approximate load levels at the various
centers:
Load Area
Anchorage -Cook Inlet
Fairbanks -Tanana Va 11 ey
Glennallen -Valdez
Percent of Total
Railbelt Load
78
20
2
Considering the geographic location and the currently projected magnitude
of the total load in the area, transmission to G1ennallen-Vtildez is not
likely to be economical in the foreseeable future. If it is ever to be
economical at all, it would likely be a direct radial extension either from
14-2
I 7: -,..
1--·-._·.
.. -~:
·. I·
I
I
I
I
I
I
I
I
I
I
I
I
I
:~:::.~
I
I
;_:,
I
I
I
I
I
;I
I
I
a·
I
'I
I
I
I
~.
I
I
I
I
I
Susitna or. from Anchorage •.. In either case its relative magnitude is too
small to h·ave sign<ificant influence on either the viabil_ity or-development
characteristics of the Susitna project or the transmission from Susitna to
the Anchorage and F'airbanKs a·reas.
As a result, it .has been assumed for study purposes that approximately 80
.percent of the generation at Susitna will be transmitted to the Anchorage
area and some 20 percent to Fairbanks. To account for the uncertainties in
future local load growth and local generation development, the Susitfta
·transmission system was designed to be able to transmit a maximum of 85
percent of Susitna generation to Anchorage and a maximum of 25 percent to
Fairbanks.
The potentfal of the Susitna Hydroelectric Project is expected to be devel-
oped in three or four stages as the system load grows over the next two
decades. The transmission system must be designed to serve the ultimate
Susitna development, but staged to provide reliable transmission at every
intermediate stage. Present plans call for three stages of Susitna ·
development, at Watana in 1993 followed by a further ·340 MW in 1997, and
600 MW at Devil Canyon in 2002.
Development of other generation resources could alter the geographic load
and generation sharing in the Railbelt~ depending on the location of this
development. However, currenf studies indicate that no other very large
projects are likely to be developed until the full potential of the Susitna
project is utilized. The proposed transmission configuration and design
should, therefore, be able to satisfy the bulk transmission requirements
for at least the next two decades. The next major gen~eration development
after Susitna will then ·require a transmission' system determined by its own
magnitude and location.
The resulting power transfer requirements for the Susitna transmission sys-
tem are indicated in Table 14.1.
(c) Transmission Alternatives
Because of the geographic location of the various centers, transmiss.ion
from Susitna to Anchorage and Fairbanks will result in a radial system ,con-
figuration. This allows significant freedom in the choice of transmission
voltages, conductors, and other parameters for the two line sections, with
only limited dependence between them. In the end, the advantages of stan-
dardization for the entire system wi 11 have to be compared to the benefits
of optimizing each section on its own merits. Transmission alternatives
were developed for each of the two system areas, including voltage levels~
number of circuits required, and other parameters, to satisfy the necessary
transmission requirements of each area.
Havi_ng estab 1 ished the peak power to be de 1 i vered and the distances over
which it is to be transmitted, transmission voltages_and number of circuits
required were determined. To maintain a consistency with standard .ANSI
volt~ges used in other parts of the United States, the following voltages
were considered for Sus itna transmi ssi.on:
14-3
. ' ''(~'\ ....
-c, ..
. .
~ Watana to Devil Canyon and on to Anchor~tge: 500 kV or 345 kV
345 kV or 23.0 kV • Devil Canyon to Fairbanks:
(i) _· Susitna to Anchorage _
( i i)
Transmission at either of two different voltage leve,ls (34'5 kV. or
500 kV) caul d reasonably· provide the necessary pawet transfer capa ...
bil ity over the dista·nce of approximately ·140 miles between Devil
Canyon and Anchorage. The required transfer capability of 1,377 MtJ
is 85 percent of the. ultimate generating capacity of 1, 620 MW. At
500 kV, two circuits would provide more than adequate capability.
·At 345 k V, either three circuits uncompensated or two circuits \'lith
series compensation are required to provide_ the necessary rei iabi 1-
ity for the single contingency outage criterion. At lower voltages
an excessive number of parallel circuits are required, while above
500 k V two circuits are st i 11 needed to provide service in the ever•t
of a line outage.
Susitna t6-Fairbanks -.~~~~~--~~
Applying the same reasoning used in choosing the transmission alter-
natives to Anchorage~ two circuits of either 230 kV or 345 kV were
chosen for the section from Devil Canyon to Fairbanks. The 230 kV
alternative requires series compensation to satisfy the planning
criteria in case of a 1 ine outage. ·
(iii)· Total System Alternatives
The transmission section alternatives mentioned above were combined
into five realistic total system alternatives. Three of the five
alternatives have different voltages for the two sections. The
principal param:!ters of the five transmission system alternatives
analy~ed in detail are as fo'llows:
Susitna to Anchorage Susitna to Fairbanks
Number of Number of
A 1 ternati ve · Circuits VoltaJe Circuits VoltaJe
(kV (kV
1 2 345 2 345
2 3 345 2 345
3 2 345 2 230
4 3 345 2 230
5 2 500 2 230
Electric system analyses, including simulations of line energizing, load
flows of normal and emergency operating conditions, and transient stability
performance, were carried out to determine the technical feasibility of the
various alternatives. An economic comparison of transmission system life
cycle costs was carried out to evaluate the relative economic merits of _
each alternative. All five transmission alternatives were found to have
acceptable performance characteristics. The most significant difference
was that single-vo'ltage systems (345 l<.V, Alternatives 1 and 2) and systems
without series compensation (Alternative 2) offered reduced complexity of
14-4
>I.,
~--
1. __ 1
'. -~
__ ._
} '1-
'\ ," ..,
I
I
I
I
I
I
I
I
I
I
'I
I
I
I
I
I
•.
I
I
I
I
I
I
•~
I;
••
~.
I
I
I
"I
I
I
I
···I
I
(d)
design and operation and therefore were likely to be marginally more
reliable. The present-worth life cycle. costs of Alter-nativ~s l through 4
were all within o·ne. percent of each other. Only the cost of the 500/230 kV
scheme .(Alternative 5) was 14 percent above the others~-A surtmary of the
lif.e cycle cost analysis for the various alternatives is shown in Table
14.2. Full details of the technical and economic comparisons are explained
in Reference 1.
A technical and economic comparison was also carried out to determine
possible advantages and disadvantages of HVDC transmission, as compared to
an ac system, for transmitting Susi tna power to Anchorage and Fairbanks.
As outlined 'in detai 1 in Reference 1, HVDC transmission was found to be
technically and operationally more complex as well as having··higher life
cycle. costs.
Configuration at Generation and Load Centers
Interconnections between generation and load centers and the transmission
system were developed after reviewing the existing system configurations at
both Anchorage and Fairbanks as we 11 as the possi bi liti es and current
development plans in the Susitna, Anchorage, Fairbanks, Willow, and Healy
areas.
(i) Susitna Configuration
Preliminary development plans indicated that the first project to be
constructed would be Watana with an initial installed capacity of
680 MW~ to be increased to approximately 1020 MW in the second
development stage. The next projectt and the last to be considel ... ed
in this study, would be Devi·l Canyon, with an installed capacity of
600 MW. ·
(ii) Switching at Willow
rransmission from Susitnaoto Anchorage is facilitated by the intro-
duction of an interr,~ediate s~-1itching station. This has the effect
of reducing line energizing overvoltages and reducing the impact of
1 i ne outages on system st abi 1 i ty. Willow is a sui tab 1 e locat 1on 'for
this intermediate switching station; in addition, it would make it
possible to supply local load when this is justified by development
in the area. This 1 oc a 1 1 oad is expected to be 1 ess than 10 percent
of the total Railbelt area system load~ but the availability of an
EHV line tap would definit~ly facilitate future power supply.
(iii) Switching at Heal~
A switching station at Healy was considered early in the analysis,
but was found to be unnecessary to satisfy the p 1 anni ng criteria.
The predicted load at Healy is small enough to be supplied by local
generation and the existing 138 kV transmission from Fairbanks.
14-5
c
(i v) Anchorage Configuration
In its 1975 report on the Upper Susitna River Hydroe1 ectr•i c Studies
(2), the United States Department of the Interior Corps of Engineers
favored a transmission route termi_nati ng at Point MacKenzie.
The 1979 Economfc Feasibi 1 ity Study Report for the Anchorage-Fair-·
banks Intertie by Ittternational Engineering Company, Inc. (IECO) (3)
recomme_nded one circuit from Susitna terminating at Point MacKenzie
and another passing through Palmer and Ek1utna substations to
Anchorage along the eastern side of Knik Arm.
Analysis of system configuration, distribution of loads, and dev-el-
opment in the Anchorage area led to the conclusion that a transfor-
mer station near Palmer would be of little benefit. Most of the
major loads are concentrated in and around the urban Anchorage area
at the mouth of Knik Arm. In order to reduce the length of sub-
transmission feeders, the transformer stations should be located as
close to Anchorage as possible.
The routing of transmission into Anchorage may be chosen from the
following three possible alternatives:
.. Submarine Cable Crossing From Point MacKenzie to Point Woronzof
This. waul d :require transmission through a very heavily developed
area. It would also expose the cables to damage by ships 1
anchors, which has been the experience with existing cables,
resulting in questionable transmission reliability.
-Overload Route North of Knik Arm via Palmer
This may be most economical in terms of capital cost in spite of
the long distance i nvo 1 ved. However, approval for this route is
unlikely since overhead transmission through this developed area
is considered environmentally un~cceptable. A longer overland
route around the developed area is considered unacceptable because
of tne mountai noL~s terrain.
-Submarine Cable Crossing of Knik Arm, In the Area of Lake Lorraine
and Six f.ti 1 e Creek -
This option, approximately parallel to the new 230 kV cable under
construction for Chugach Electric Association (CEA), includes some
3 to 4 miles of submarine cable and requires a· high capital cost.
Si nee the area is upstream from the shipping lanes to the port of
Anchorage it will result in a reliable transmission link, and one
that does not have to cross environmentally sensitive conservation
areas.
The third alternative is clearly the best of the three options. The
details of this configuration are as follows:
14-6
0
I
10
I
I
I
'I
I
I
I
I
·I
·a
I
••
I
I
I
I
I
I
I
I
I
I
I
I
I
I
1\
I
I
I
I
I
I
I
-1' \ -'
t:
submarine cable crossing. To reduce cable costs the crossing could
be constructed with two cable. circuits plus one spare phase. This
option requires a switching station at the west terminal of Knik
Arm.. A switching station at the west terminal would clearly require
increased costs c:md .complications for construction and operation as
a result of poor access. It would also require a separate location
for the tap to supply MEA.
Plans are present 1y underway for a bridge crossing at Kni k Arm for
both railway and road traffic. If these plans should be realized,
transmission costs and complications could be significantly reduced
by routing the tables across the bridge.
(v) Fairbanks Configuration
Susitna power for the Fairbanks ar·ea is recommended to be delivered
to a single EHV/138 kV transformer station located at Ester.
(e) Recommended Transmission System
The configuration of the recommended tra.nsmission system, (Alternative 2)
is shown on the single-line diagram (Figure 14.1). The.main characteris-
tics of the recommended systems are summarized in Table 14.3.
14~2 -Corridor Selection
(a) Methodology
Development of the proposed Susitna project will require a transmission
mechanism to deliver electric power. to the Railbelt area. The pre-building
of the Intertie system&~ill result in a corridor and route for the Susitna
transmission lines between Willow and Healy. Therefore~ three areas re-
quire study for corridor selection: the northern area to connect Healy
with Fairbanks; the centra 1 area to connect the Watana and Devi 1 Canyon
damsites with the Intertje; and the southern area to connect Willow with
Anchorage.
The corridor selection methodology followed the Susitna study plan formula-
tion and selection methodology. Previous studies, existing data, aerial
reconnaissance and limited field studies. formed the data base. Using the
selection criteria discussed below, corridors 3 to 5 miles wide which met
these. criteria were selected in each of the three study areas. These cor-
ridors were then evaluated to determine which ones met the more spec-ific
screening criteria discussed below. This screening process resulted in one
corri dar in each area being des:i gnated as the recommended corri dar for the
transmission line. For a more detailed discussion of study methodology and
the selection and screening criteria, refer to the Transmission Line
Corridor Screening Closeout Report (~eptember, 1981), hereafter referred to
as the Closeout Report.
14-7
(b) Previous Studies
The two reports reviewed which contained the most information relevant to
the transmission line .stu~n es were;
-The Susitna Hydroelectric Project Interim Feasibi1ity Report, prepared by
the U.S. Army Corps of Engineers. (hereafter referred to as the COE re-
port); and
-The Economic Feasibility Study for the Anchorage-Fairbanks Interti e by
International Engineering Company, Inc./Robert W. Retherford Associates
(hereafter referred to as the IECO/RWRA report).
The COE report consisted primari 1 y of an eva 1 uati on of alternative corridor
locations to aid in the selection of those which maximized reliability and
minimized costs. Utilizing aerial photographs and existing maps, general
corridors connecting the project site with Anchorage and Fairbanks were
selected. This study was general in nature and was intended only to
demonstrate project feasibility.
The IECO/RWRA report utilized the COE report as background information for
both economic feasibility determination and route selection. The corridor
selected by IECO/RWRA was very similar to that selected by the COE with
further definition. The route selected was based on length, accessibility
and environmentai compatibility. The report also presented a detailed
economic feasi bi 1 i ty study for the Anchorage-Fairbanks transmission study.
(c) Selection Criteria and Selection Results
( i) Cri teri a
The objective of the corridor selection conducted by Acres was to
select feasible transmission line corridors in each of the three
study areas:
-The northern area, to connect Healy with Fairbanks;
-The central area, to connect the Watana and Devfl Canyon damsites
with the Intertie; and
-The southern area, to connect Willow with Anchorage.
Technical, economic, and environmental criteria were developed in
order to select corridors within the three areas. These criteria
are 1 i st ed i n Tab 1 e 14 . 4.
Environmental inventory tables were then compiled for each corridor
selected, listing length, number of road crossings, number of river
and creek crossings, topography, soils, land ownership/status,
existing and proposed development, existing rights-of-way, scenic
quality/recreation, cultural resources; vegeta~ion, fish, birds,
furbearers, and big game. These tables are included in the Closeout
Report. ?
14-8
I
I
J
I
I
I
I
I
I
I
I
I
I'
I
I
~a,
·-
1
I
I
I
I
I
'I
I
I
I
'I
I
••
I
I
I
I
I
I
I
I
I
( i i ) Results ·
UtilTf!.ing existing information, 22 ctJrridors were selected based on
their· ability to meet technical, economic and environmental crit.eria
as listed in Tabl~-14.(.
Four of the corridors are in the southern study area, 15 ·;n the cen-
tral area, and four in the northern study area. Three of the corri-
dors in the southern study area run in a north-south direction while
one runs northeast to Palmer, then northwest to Willow. Corridors
in the central study area are in two general groups: those running
from Watana Damsite west to the proposed Intertie and those running
north across the Denali Highway and the Chulitna River. Corridors
in the northern study area r·un either west or east to bypass the
Alaskan Range, then proceed north to Fairbanks.
See Figures 14.2, 14.3, and 14.4 for the location of these
corridors.
(d) Screening Criteria and Screening Results
( i) Criteria
The objective of the screening process was to screen the previously
selected corridors to determine which best meet additional techni-
. cal, economic, and environmental criteria as listed in Table 14.5.
The rationale for selection of these criteria is explained in
Appendix E2.
In addition to these criteria, each corridor w&s screened for reli-
ability to determine if a line could be operated with a minimum of
power operation. Six basic factors were considered in relation to
r e 1 i ab i l i t y:
-Elevation: Lines located at elevations below 4000 will be less
exposed to severe wind and ice conditions which can
interrupt service.
-Aircraft: Avoidance of areas near aircraft landing and takeoff
operations will minimize the risk of collisions.
.
-Stability: Avoidance of areas susceptible to land, ice, and snow
slides will reduce the chance of power failures.
-Existing
Power
Lines:
Avoiding existing transmission lines will reduce the
possibility of lines touching during f.ailures and will
facilitate repairs.
-Topography: Lines located in areas with gentle relief will be
easier to constru~t and repair~
.. Access: Lines located in reasonable proximity to tran.sporta-
tion corridors will be more quickly accessible and~
therefore, more quickly repaired if any failures
occur.
14-·9
-;-;; . .:.::.
( i i)
The screening criteria-and reliability factors for each corr-idor
were evaluated utilizing topograpt:ic maps, aer\'al photos, aerial.·
overflights~ and published materi ~ls •. Each corridor was then ·
assigned four -ratings (one· each for technical, economic and environ ...
.mental consideratto.ns, and one overall summary rating.} Ratings
wer-e defined a·s fo1lows: -
A ... recommended
. C -acceptable but not preferred
F .oc unacceptab 1 e
From the technical point of view~ rel i abi 1 i ty was the main objec-
tive: An environmentally and economically sound corridor was re-
jected if a line built in· the corridor would be unreliable. Thus~
any line whicn received an F technical rating was assigned a sumnary
rating_ of F and eliminated from further consideration.
Similarly, because of the critical importance of environmental con-
siderations, any corridor which received an F rating for environmen-
tal impacts was assigned a summary rating of F, and eliminated from
consideration.
Results
Table 14,.6 summarizes the comparison of the corridor screened in the
southern, centra 1 and northern study areas. One corridor in each of
the three study areas received A ratings for a 11 three categories. .
These three corridors and the rationale for their A ratings are dis ...
cussed be low. For a description of a 11 22 corridors and the ration-
ale for their ratings, see Appendix E2.
-Southern· Study Area
Corr.'idor Two -Willow to Point MacKenzie vi a Red Shirt Lake
• Oescri pt i on
Corri. dor AOFC, consisting of Segments ADF and FC (Figure 14.2),
comnences at the point of intersection with the Intertie in the
vicinity of Wi 11 c~w but immediately turns to the southwest,
first crossing the railroad, then the Parks Highway, then
Willow Creek just west of Willow. The land in thevicinity of
this part of the segment is very flat, with wetlands domi·nating
the terrain.
Southwest of Florence Lake, the proposed corridor turns,
crosses Rolly Creek, and heads nearly due south ti passing
through extensive wetlands west and south of Red Shirt Lake.
Th.e corridor in this area parallels existing tractor trails and
crosses very flat lands with significant amounts of tall-grow-
ing ve_getation in the better drained locations.
14-10 ~
l.o
.," .. , -; l'
1:. "
·~l :: 1-,· .. ·~.
'r_,_ •
t~-:
I
I
I
I
I
I
I
I
I
I
I
I
·-.1'-·.·.t.
' .
I
I
I " '
1··;·.· '
' '
~ " i. ·.. :
I'
. •, ~ ~'
1
I
I
I
I
I
I
·~-.1· '
'." ./ ..
I'
·-
1
I
I
,I
' i; .'
I
I
-:,·, ~'t._ ;;
",'}
Northwest of Yohn Lake~ the corrid~r segment turns to the
southeast, passing Yohn Lake and My Lake befor-e crossing the
Little Susitna River •. Just south of My Lake~ the :Corridor
turns in a southern direction, passing Mi. dd1e Lake and east of
Hor·seshoe take before finally inte~se:cting ttre ,~xist1ng B~luga
230 kV transrnissi orf line at a spot just .north of Maci<enzi e
Point. From here., the corridor parall~ls MacKenzie Point 1 s
existing transmission facilities before crossing under l<nik Arm
to emerge on the eastern shore of Knik Arm in the vicinity of
Anchorage. The 1 and in the vicinity of this segment is
extremely flat and wet. '!t supports stands of tall-growing
vegetation on the higher or better drained areas.
. Technical and Economi ca 1 Rating
Corri do.r ADFC crosses the fewest number of rivers and roads in
the southern study area. It has the advantage of paralleling
an existing tractor trail for a good portion of its length~
thereby reducing the need for new access roads. Easy ace ass
wi 11 allow m·a; ntenance and repairs to be carried out i'n minima 1
time. This corridor also occurs at low elevations and is
approximately one-half the length of Corridor Onec
• Environmental Rating
This corridor crosses extensive wetlands from Willow to Point
MacKenzie. . At higher elevations or in the better drained
sites, extensive forest cover is encountered. Good agricul-
·tu~al soils have been identified in the vicinity of this corri-
dor; the state plans an Agricultural Lands Sale for areas to be
traversed by this corridor. The corridor also crosses the
Susitna Flats Game Refuge. The presence of an existing tractor
trail near considerable portions of this corridor dimiriishes
the significance of some of these constraints. Furthermore,
its short length and the fact that it crosses only one river
and eight creeks increases its environmental acceptability.
-Central Study Area
Corridor One-Watana to lntertie via South Shore, Susitna River
. ~script ion
This corridor originates at the Watana Dam site a·nd follows the
southern boundary of the river at an elevation of approximate 1 y
2,000 feet from Watana to Devil Canyon (Figure 14 .. 3). From
Devil Canyon, the_~grridor continues along the southern shore
of the Susitna River at an elevation of about 1,400 feet to
where it connects with the lntertie, assuming the Intertie
follows the railroad corrigor. The land surface in this ·area
is relatively flat., though incised at a number of locations by
tributaries to the Susitna River. The relatively flat hills
are covered by discontinuous stands of dense, tall-growing
vegetation .•
14-11
• Techn-ical and. Economical Rating
Corrido,r One is one of the 'shortest corridors considered. It-
is approximately 40 miles, long, making it economic-ally favor-
able. No technical restrictions were obser-ved along the entire ·
le.ngth of this corridor.
. Environmental Rati-ng
Because of its short length, environmental disturbance caused
by transmission line construction would be reduced. The more
noteworthy constraints are those identified under the catego-
ries of land use and vegetation. Corridor One would require ·
the development of a new right-of-way between Watana and Devil
Canyon with some opportunity existing to utilize the COE-
developed road for access between the In terti e and De vi 1
Canyon. The potential does exist in this cor.ridor to use the
proposed access road. Wetlands and discontinuous forest cover
occur in the.corridor, especially in the eastern third of the
route. Access road development, and the associated vegetation
clearing, present additional constraints to this corridor.
-Northern Study Area
Corridor One -Healy to Fairbanks vi a Parks Highway
Description
Corridor One (ABC) , cons 1St i ng of Segments AB and BC, statts in
the victnity of the Healy Power Plant (Figure 14.4). From
here, the corridor heads northwest, crossing the existing
Golden Valley Electric Association Transmission Line, the
railroad, and the Parks HighWay before turning to the north ,and
paralleling this road to a point due west of Browne. Here, as
a result of terrain features, the corridor turns northeast,.
crossing the Parks Highway once again as we 11 as the existing
transmission line, the Nenana River~ and the railroad, and
continues to a point northeast of the Clear Mi ssi 1 e Ear 1y
\~arning Station (MEWS).
Continuing northward, the corridor eventually crosses the
Tanana River east of Nenana, then heads northe-ast, ftrst Crtls-s-~
· ing Little Goldstream Creek, then the Parks Highway just north
of the Bonanza Creek Experimental Forest. Before reachi rig the
drainage of Ohio Creek, this corridor turns back to the north-
east:s crossing the old Parks Highway and heading into the Ester
Substation west of Fairbanks .•
Terrain along this entire corridor segment is relatively flat,
· with the exception o.f the foothills north of the Tan an a River.· .
Much of the. route, especially that portion. between the Nenana
and th·e Tanana River crossings, is very broad and flat. It has
standing water during.·the summer months and, in some places, is
overgrown by dense stands of tall-growing vegetation. This
corridor segment crosses the heavily wooded foothills northeast
of Nenana. ·
'14-12
,,
I·
·J--·~. . .---.
I
I
I
I
I
I
I
I
I
I
I
I
t:
I .<•
1_,"
''.
~·
••• . .· .
·. ·•'
1·, ...
. . ,'
I
I
I >;; .,
I
I '•
I
I
I
I ~.
·I
I
I
I
I
I
I
(e)
An option to the ,above not shown in-the figures has been
considered., closely paralleling and. sharing rights-of-way with
the existing Hea1y ... .fairbanks transmission 'line. ·While it is
· usually attracti v~ to parallel extsti.ng corridors wherever -
.. p.osstble·, this op.tion necessitates a great number of· road
crossings and results in· an extended 1 ength of 1:he eorri dor
paralleling the Parks Highway .. A potentially significant
amount of highway-abutting land would be usurped for contain-.
ment of the right-of-way. The combination of these features
precludes this corridor from further evaluation.
o Technical and Economical Rating
•
This corridor crosses the fewest water courses in the northern
study area. Although it is approximately four miles longer
than Corridor Two, it is technically favored because o.f the
existence: of potentia 1 access roads for almost the ·entire ·
length ..
Envi ro~!]nta 1 Rating
Because it parallels an -existing transportation corridor for
much of its length, this corridor would permit line routing.
that w.ould avoid most visually sensitive areas. The three pro-
posed road crossings for this corridor (as-opposed to the 19
road crossings of the Healy-Fairbanks transmission line) could
occur at points where roadside qevelopment,exists, in areas of·
visual absorption capability, or· in areas recommended to be
opened to long-distance views.
Four rivers and 40 creeks with potential for impacts are
crossed by this corridor. It crosses the fewest number of
water courses of any route under consideration in the northern
study area. In addition, the inactive nest site of a pair of
peregrine falcons occurs within this proposed corridor.
r.; As with visual impacts, land use, wildlife, and fishery re ...
source impacts can be lessened through carefully route location
and utilization of existing access. Impacts on forest clearing
can be lessened through the sharing of existing transmi sslon
line corridors.
Cone lusi ons
A review of previous reports, other existing information, and aerial over-
flights was used to select corridors for consideration in this study.
These corridors were screened against certain technical, economic and envi-
.ronmental criteria, resulting in one recommended corridor in each of the
southern, central and northern study areas. The corridors shown in Figures
7.1 through 7.8 of Appendix E2 are .believed to best meet the technical,
economic and environmental criteria; therefore, these corridors are the
best locations in which to p 1 ace the Susitna transmission 11 nes.
..
14-13
14.3 -Ro.ute Selection
'-}::-.:::·-~-:-.:.-:....:.. ~-,
(a) _Method() lQSl:
After identifying the. pr:efer.red transmJsston line corridors, the next $tep
in the route selection process involved the analysis o:f the data as gather-
ed and presented on the base map~ overlays were compi.ted so that various
cmi:itr:aints affecting construction or maintenance of a transmission facil-
ity could be viewed on a single map. The map is used to select possible
routes within each of the three selected corridors. By placing a 11 major
constraint-s (e.g., area of high visual exposure, private lands, endangered
species~ etc.) on one map, a route of least impact was gelected. Existing
faciliti.~s, such as transmission lines and tractor trails within the study
area, ~t:re also considered during the selection of a least impacted route.
Whenever poss-ible, the routes were selected near existing or proposed
access roads, sharing whenever possible existing rights-of-way.
The ~ata base used in this analysis was obtained from the following
sources:
-An up-to-date land status study;
-Existing aerial photos;
-New aerial photos ·conducted for selected sections of the previously rec•
ommended transmission line corridors;
-Environmental studies including aesthetic considerations;
-Cl imatologi ca 1 studies;
-·Geotechnical exploration;
-Additional field studies; and
~ Public opinions.
{b) Selection Criteria
{i) Criteria
·The purpose of this section is to identify three selected routes:
one from Healy to Fairbanks, the second from Watana-Devil Canyon
damsites to the intertie!! and the third from Willow to Anchorage.
The previously chosen corridors were subject to a process of refin-
ing and evaluation based on the same technical, economic, and envi-
ronmental criteria used in corridor selection (see Table 14.15).. In
addition, special emphasis was concentrated on the following
points:
-satisfy the regulatory and permit requirements;
-selection of routing that provides for minimum visibility from
hi-ghways and homes; dnd . .
-avoidance of developed agricultural lands and dwellings ..
-
14-14 .
1' .
. \f.;..:,.,·
'
.I• ..•.
' . '
I
I
I
I
I
I
I ·.
I
l
I
I
I' r~ ; .,
'
I
I ·~ill
I
I
':1··.
·.. i
'· ..•.. :.
" ..
. .
«
I
I
t·
I . . '·'*' ~
' I
I
1·.·!
. '
'!'>,, .,
I ~ .
I ......
·., .... ~;;_ -,t '. ~
I
··''':,,.
·, ,_7-
1
I
;.
Figures 1 througb 14 in J.\ppendi~< E3 show the selected trartsmission
line. rout.e fo.r t~e· three areas of study; hatnely, the southern study
ar,ea; the central study area; and the northern study area. · As a
first step, the. 3...,to~S ... mi le-widt.h corridor previously selected for
each of the three study areas was narrowed to a half-mile-width
corridor based on the previous. criteria" The preliminary centerline
of the right-of-way is shown in the figures. This centerl tne
represents a right-of-way width of 400 feet. This width is adequate
for three~ single-circuit, parallel lines with tower structures
having. hori zonta 1 phase sp.aci ng of 33 feet. However, between the
Devil Canyon damsite and the intertie, the width of the right-of-way
is 700 feet which is needed to accommodate five sing le~ci rcuit
1 i nes.
Southern Study Area -Willow to Point Mackenzie
Via Knik Arm._C_r_o_ss_,_·n_g ___ ~---------
Description
Evironmental Considerations
. Techni ca 1 and Economic Considerations
This .route crosses a very few number of rivers and roads.. It paral-
lels the existing tractor trail for a considerable portion of its
length, thereby reducing the need for new access roads.
At Knik Arm, the route parallels a proposed 230 kV transmission line
and will share (if possible) its right-of-way. This will avoid
pi aneering new right-of-way especially in an area full of dwellings
and other· constraints. ·
14-15
Central Study Area -Wat~na to Intert ie
·oescrlptio.IJ
Environmental Considerations
Technical and Economic Considerations
This is the shortest route among the studied ones in the central
study area~ which makes it economically feasible. The route paral-
lels a proposed access road almost through its entire length north
of Sus itna River for the sect ion between the two d amsites and south
of it between Devil Canyon and the Intertie. This wi11 add to the
reliability and economical considerations.
Northern Studx Area -Healy to Ester
Desc~iption
14-16
. . ·~
I
I
I
I
I
..• 1·.·
•.
I
I -~·
I
:I
I · .........
I
I
I " .~
I
I
1·, -, .
.
l c ~:
. .
• ""<"'~ ,.. ~ •• ._,~ _..,, : ........... ;:;·,~_.;.-~ ··•r~· ~· ~-•·-'·· ·.
-
I
.
I~"·~
.;''I
I.
I
I
I
' I
I ., ..... ·
1
~
I -.,,
I
I ~..,.,.,..:
I
I
.. ,.'._,
'
. 'i
I
.
I '
t\ '' . . :.::.:: ..
. . .
(c)
. .
··Environmental Considerations
Technical and .Economic Considerations
The existence of access roads for almost the entire length makes
this selected route technically and economically feasible. -Its
short length will add to its economical considerations.
Route Soil Conditions
-~
(i) Description
Baseline geological and geotechnical information has been compiled
through photointerpretation and terrain unit mapping .(Reference 4).
The general objective was to document the conditions that would
significantly affect the design and construction of the transmission
line towers. r~ore specifically, tpe objectives included the deline-
ation of and forms of various origins, noting the occurrence and
distribution of significant geologic factors such as permafrost,
potentially unstable slopes, potentially erodible soils, pqssible
active fault traces, potential construction materials, active
floodplains, organic materials;, etc.
Work on the air photointerpretation consisted of several activities
culminating in a set of terrain emit maps delineating surface mater-
ials and geologic features and conditions in the project area.
The first activity consisted of a review o'f the 1 i terature concern-
ing the geology of the Intertie corridors and transfer of the infor-
mation gained to high-level photographs at a scale of 1:63~000. In•
terpretation of the high-level photos·created a regional terrain
framework which helped in the interpretation of the lo~1-leve1
1:30,000 project photos. Major terrain divisions identified on the
high-l.evel photos werethenused as an aerial guide for delineation
of more detailed terrain units on the low-leve,l photos. The primary
effort of the work was the interpretation of 140-p 1 us photos cover-
ing about 300 square miles of varied terrain. The land area covered
in the mapping exercise is shown on map sheets and displayed in
detai 1 on photo mosaics .
14-17
-' -...: ' . . . . -. ·~·
. . ~·'
( i i)
As part .of the terrain analysis, the. various bedrock .units.· and domt-
nant lithologies· were identified usirt9·; pub1 ished u'"s. Geologjcal
Survey reports •. The extent of these. units was,· .roughly delineated on
the: photographs, and using exposure patterns, 'shade, texture, and
other features of the rock unit as they appeared on the photographs,
~ ' 1
I
t
unit boundaries were drawn.. Terrain unit symgols denoting the vari ....
uus lithologies were utilized on the m.aps~ ·
Physical characteristics and typical engineering properties of each
terrain unit were considered-and a large chart for each corridor· was
developed. The charts identify the terrain units. as they have been
mapped and characterize their properties in numerous categories.
This allows an assessment of each unit's _influence on various pro;..
ject features.
Terrain Unit Analysis
-
The terrain unit is a special purpose term comprising the land forms
expected to occur from the ground surface to a depth of about 25
feet.
The terrain unit maps for the proposed Anchorage to Fairbanks trans-
mission 1 ine show· the aerial extent of the specific terrain units
which were identified during the air photo investigation and were
corroborated in part by a limited onsite surface investigation. The
units document the general geology and geotechnical characteristics
of the area.
The north and south corridors are separated by several hundred miles
and not surprisingly encounter different geomorphic provinces and
climatic conditions. Hence, while there are many la~dforms (or in-
dividual terrain units) that are common to both corridors, there are
also some landforms mapped in just one corridor. The 1 and forms or ·
individual terrain units mapped in both corridors were briefly des-
cribed.
Several of the landforms have not been mapped ind~pendently but
rather as compound or complex terrain units. Compound terrain units
result when one 1 and form over 1 ies a second recognized unit at a
shallow depth (less than 25 feet), such as a thin sheet of glacial
till ov~rlying bedrock or a mantle of lacustrine sediments overlying
till. Complex terrain units have been mapped where the surficial
exposure pattern of two 1 andforms are so intricately related that
they must be mapped as a terrain unit complex, such as some areas of
bedroc:k and colluvium. The compo-und and complex terrain units were
. described as a composite of individual landforms comprising them ..
The stratigraphy, topographic position, and aerial extent of all
units, as they appear in each corridor, wer·e summarized on the
terrain unit properties and engineering interpretations chQ.rt ..
14-18
,~...._ ":
I
-·~
I
I
I
I
I
I
a+
,.;:;._ ......
I ._~-, .....
I
t~
~~· ,;, ,__.,
I
I
I
I
I ' '
,.
I
t
I
"-•-"
I ... ~
I
' \[>'
I
'W'
I '-'
I ,,
~
I "':,"'
I -...~-
1
.' :.,...,
I -..;..
I
I
r.:.l.·.· .. ·
(
"-··'
.. ·a·.·.· __ ..
< '•'
t
{d) Conc1uslons
A study of existing information and aeri a1 overflights, together with add'i-
-ti onal aeri a 1 coverage, was .used to 1 ocate the recommended route i n each ,of
the southern; central, and nort-hern study are.as.
Additional environmental information and land status studies made it pos-
sible to align tha routes to avoid any restraints.
Terrain unit_ maps describi-ng the general material expected in the area were
prepar.ed specifically for transmission line studies and were-used to locate
the routes away from unf avor,ab le soi 1 conditions whenever possible .
The route shown in Figure 14 .. 5, represents the general location of-the
-r_ecommended ·route. It also shows the existing surface transportations.
Fi gores 1 through 14 of Appendix E3 are be 1 i eved to best meet the techni-
cal, economical~ and environmental criteria. In these figures, a half-
mile-wide corridor is located; within this width, the centerline of th~
proposed route is located. The centerline represents 400-foot-w-idth right-
of...;way for the entire length of the transmission line, except in the seg-
ment of the 1 i ne between Devil Canyon and the Interti e where the right-of-
way is 700 feet. This segment has five single-circuit lines parallel to.
each other.
14.4 -Towers~ Foundations and Conductors
A transmission line intertie between Anchorage and Fairbanks is planned by APA.
The intertie will consist of existing lines and a new section between Willow and
Healy. The new section wi 11 be bui 1 t to 345 kV standards and wi 11 be fu 11y com-..
patible with Susitna requirements.
(a) Transmission Line Structures
(i) Selection of Tower Type
Because of unique soil conditions in Alaska, with exten~ive regions
of muskey and permafrost, conventional self-supporting or rigid
towers will not provide a satisfactory performance or solution for
the proposed transmission 1 i ne. ·
Permafrost and seas on a 1 changes in the soi 1 are known to cause 1 arge
earth movements at some locations, requiring towers with a high
degree of flexibility and capability for handling relatively large
foundation movements without appreciable loss of structural integri-
ty.
The guyed tower is exceptionally suitable for these types of condi-
tions. The recommended· type of structure for this study is there-
fore the hinged-guyed steel x-tower (Figure 14.6).
14-19
1
I
j
I
I :
'i
J
ij' I :r
''.·j' '' I,
.r
The design features incTi.ide hinged connections between the 1 eg
members and the foundations ·which, together with the longitudinal
guy system, provide for large flexibility combined with excellent
stability in the direction of the line. -Transverse stability is
provided by the wide leg base which also accounts for relatively
srna 11 and manageable footing reactions~ -. -
In add 1 ti on to the above features of the selected structure, the
following are important favorable points:
-The x-frame structure provides for 1 ess vi sua1 impacts than other
structures. This results from the simplicity of design;
-It requires little maintenance except for guy adjustment when
needed;
-Towers caul d be stored in remote areas \'lith out concern for
replacement when needed because of vandalism or deterioration;
-Easy for construction., a typical tangent structure consists of
only six major components with bolted connections. All bolts are
loaded in shear to eliminate any special consideration for
torquing of assembly nuts during field assembly; and
-Economically attractive, especially when considering all the
engineering features it combines.
The tangent tower shown in the figure represents the majority of
structures required for the transmission line (about 90 percent of
the tot a 1 structures) . However, speci a 1 types of structures. may be
needed to satisfy a specific field requirement, wire stringing or
line angles.
Also, it is anticipated that the section between Knik Arm and
University Substation will consist of double-circuit,
self-supporting, single-pole type structures.
(ii) Design Parameters
-Clearance to Ground: A ground clearance of 32 feet may be ade-
quate against expected field tolerances.
-Spans: A wind span of 1400 feet, and weight span of 1600 feet is
recommended for the tangent structure. This will combine economy
and fl exi bi 1 i ty in spotting the structures. Longer ·spans, how-
ever, may be needed to overcome specific site location.
~ Structure Height: An average height of 85 feet is expected
thrOughout the transmission lines.
-Insulations: The center phase is a V-string assembly to control
sw1ng and to provide for smaller phase spacing .. The side strings
are of the single type.
14-20
i',l· ,· ', '.:::;;,--.
. :·1·.·.:;_· . ,
' .
.
.
I
I '-... : ~'
I ,, ... ~
I
"'""
I
I
' .I:-
i
>.,,·t:'"
I ·,~~
I ·~-
'" ~ \.'1."'~:/
I
.a:. -.
I .j,;·«.,.-.
' 1
..... ,
I
I
I
1,•-
;
'
' l
I
I
..... 1'"'
'
'
' ~-1' -i
I
I ._,.
I ...,_
I --...~
I
I '·
I
t
-Shield Wires: The shield wire when needed will be 3/8 inch by 7
strand, EHS stee 1 •
(iii J .Loading .Condi tj ons~
Climatic studies for transmission lines were performed to determine
likely wind and ice loads based on historical data. A more detailed
study inc.orporating additional climatic data was performed to con-
firm or modify the obtained data.
Details of the climatic studies for· transmission lines may be found
in Appendix B-6.
The design loads acting on wires and structures are mainly based
upon weather conditions. Four cases of loadings are thus establish-
ed f?r the tower design. These are:
-National Electrical Safety Code (NESC) Heavy Loading: This con-
sists of 1/2-inch radial ice around the wires, 40 miles per hour
wind at 0°F. This wind speed produces four pounds per square inch
pressure acting on the project areas of cylindrical surfaces.
-Extreme Wind Loading: Consists of 140-mph wind (produces 50 psf
on cy1 i ndri ca 1 surf aces), no ice at 32° F.
-Heavy_Ice: Consists of one-inch ice with no wind at 32°F.
-Longitudinal Pull: No ice, no wind, 32°F. The pull pressure
should be applied at any one conductor attachment.
The first case~ NESC heaving loadings, will be applied to the major
part of thel)transmission line; however, in certain areas where the
weather conditions are more severe, the seccnd and third case may be
applied.
Loadings Upon Supporting Structures
-Description of Loads
• Vertical loads on supporting structures and foundations shall be
their own weight plus the superimposed weight which they sup-
port, including all wires, ice coated when specified.· The ef-
fective vertical span for wires shall be determined with proper
consideration of the effect of support at different elevations.
The weight of ice shall be determined based on 57 pounds per
-cubic foot.
• Transverse Loading (perpendicular to the wires). This shall be
determined from the following conditions: wind in conjunction
with ice and wind without ice. The effective span of determin-
ing the wind on wires shall be equal to one-half the sum of the
adjacent spans between supporting structures.
{b)
Where a change in direction of wir£rs occurs~ .. a transverse.~loadlng~
upon the suppo~t i ng structure sha 11 be a resultant .1oad equal to ·
. the vector sum of the maximum transverse wind load and the
resultant load imposed by,the wires because of their change in
·direction~ In obtaining these loadings, a wind direction shall· be
used which will. give the maximum resultant load.
-Longitudinal Loading
Proper ~llowance should be made for longitudinal loads which may
be produced on structures by wire stringing operations and con-
struction techniques.·
-Application of Loads
The vertical, transverse, and longitudinal loadings p~eviously
specified re 1 ate to loadings on the wires and structures. The
component of these loadings should be considered acting simultane-
ously.
(iv) Results
The hinged-guyed steel x-type tower is selected as the basic struc-
ture for the project because of its flexibility and performance in
withstanding the unique soil conditions in Alaska. Also, the
x-tower is rated very favorably concerning reli ability, maintenance,
construction, economy, and aesthetics.
Tower Foundations
(i) Geological Conditions
A generalized terrain analysis was conducted to collect geologic and
geotechnical materials data for the transmission line corridors" a
re 1 ati ve ly 1 arge area. The engineering characteristics of the ter-
rain units have been generalized and described qualitatively~ When
evaluating the suitability of a terrain unit for a specific use, the
actual properties of that unit should be verified by onsite subsur-
face investigation, sampling, and laboratory testing.
The three main types of materia 1 s a long the transm·i ssi on line are:
-Good materia 1, which is defined as materia 1 which permits augered
excavation and allows installation of concrete without special
form work;
-Wet 1 and and permafrost materia 1 which requires additional design
details providing additional depth~ and
--Rock materia 1 is defined as :nateri a 1 in which dri 1.1 ed-in anchors
and concrete footings can be used. ·
14-22.
I
I ·~·
I
'
' I
I ~
I
' ,,
I •.._. ,, ...
~.-
' I ·.,~
I
I
I ·7
I
«
I; :~ ' -: .>~
I
.
•· A ,
.
I
·,-_ ·~
t
I
t
I
.
i::
~·
I ~ ....
I -
:,.!
. !
I
I ""'-'
I
I . 1.lo.-t
I
I
Based on aerial, topographic., and terrain unit maps, the following
is recorded~
-For the sbuthern study·· area: ·Wet 1 and and .perme}rrost m~teri a 1 s
constitute the major part of this area. Some ro.ck and good foun-
dation materials are present in this area in a very small propor-
tion.
-For the central sb.1dy area: Rock foundation and good materials.
were observed· in most of this study area.
For the northern study area: The major part in this area is the
wetland and permafrost materi a 1 s. Some parts-'!Tave good rock
materials.
(ii) Types of Foundations
The recommended two-1 egged x-frame tower is hinged at the foundation
attachment connection for longitudinal freedom and restrained by
fore and aft guying to an equa 1 i zing yoke. This arrangement wi 11
result in relatively smaller loads on the foundations. The recom-
mended types of foundations are shown in Figure 14~ 7. These are:
-Rock Anchor
This type of footing is specified whenever good quality rock is
encountered near the ground surface. The concrete piers are
grouted into the rock with reinforcing bars; permissible bearing
values with this type ·of footing are high. The entire hole, as
shown in the figure, can be drilled using the small diameter hole
size without casing. This type of hole is easy and quick to drill
and presents little or no problems.
The minimum depth of these holes is approximately eight feet, with
the entire depth grouted to insure adequate anchoring below the
maximum frost depth.
-Pile Foundation
Most of the t1ansmi ssi on line towers wi 11 be supported by pile-
type foundations consisting of heavy H-pi1e beams driven to vari-
able depths, depending upon the soil conditions.
This type of footing is considered in the three study areas \vhen a
good bearing stratum does not occur at normal footing depth or at
a reasonable distance below. The piles driven to firm strator
until the required penetratiori is reached would be less costly
than other types of footings for the same type of soil. A minimum
safety factor of 2 is recommended to be applied against uplift.
The uplift resistance is always dependent on the skin frittion be-
tween pile and soi 1. The safety factor may be varied if soi 1
tests or pile loading tests indicate that uplift resistance of the
pile is greater or less •
14-23
... :~.:·c~.~ ~-~~~~-~~~~~ •. ~~~~--~ -~-' •··~·-;0-.·~~ ~~~--~--~~~~---~~ ~-~~-~-~~. ,~. ~-~~~~--~~~~~ ••• ~.--~--~~-:~:r~~-·~-. ~: ~~~·-~-~~ ~~~-~-~~~~-:~t7~~-~;·~~ .~.~~-~r_~~---~~-a~:~~.~~]~~---~~ · , ' ,• ~ . -~ -~ ~ --. ~ ': ~. . '':i~~.:~-
-~ ~ .
'> . . • ·I·< ··~
(c)
(iii} P .. e~i gn Crite~i a
The greater part of the combined maximum ·reactions on transmission
structure footing is usually from temporary load.s such as broken
wire, wind, and ice. ·with the exception of. heavy-angle; dead-end,
or terminal structures, only ~ part ~''f the total reaction is of a
permanent nature. As a consequence, the permi ssi b 1 e soi 1 pressure
as used in the design of building foundations may be considerably
exceeded for footing for transmissio~ structures.
The permissible values of soil pressure used in the footing design
will depend on the structure and the supporting soi 1. The basic
criterion is that displacement of the footing is not restricted be-
cause of the flexibility of the selected x-frame structure and the
hinged connection to the tower footing-~ The shape and configuration
of the selected tower are important factors in foundation consider-
ations.
Loads on the tower consist of vertical and horizontal loads and are
transmitted down to the foundation and then distributed to the soi 1.
In a tower p 1 aced at an ang 1 e or used as a dead-end in the 1 i ne, the
horizontal ·1 oads are responsi b 1 e for a 1 arge portion of the 1 oads on
the foundation. In addition to the horizontal shear, a movement is
also present at the top of the foundation, creating vertical down-
load and uplift forces on the footing.
To select and detai 1 design the most economical type of foundation
for a specific tower location, soil conditions at the site must be
known. Soi 1 i nvesti gati on wi 11 furnish this needed information. A
soil boring is a guide to the type of soi 1 and its strength in re-
sisting the forces on the tower. ·,he cost of soi 1 borings is sma 11
compared to the line cost per mile. The primary purpose of soil
borings is to assure an adequate and safe foundation.
Conductor Requirements
( i) Conductor Size
Based on the transmission and power transfer requirements at the
various stages of the Susitna development, economic conducfor size-s
were determined.
The methodology used to obtain the economic conductor size and the
results obtained are outlined in Appendix El. Also included in the
,£\ppendix are the capitalized costs of transmission line losses.
When determining appropriate conductor size, the economic conductor
is checked for radio interference (RI) and corona performance. If
RI and corona performance are within acceptable limits, then_the
economic conductor size is used. However, where the RI and corona
performance are found to be limiting, the conductor selection is
based rin these requirements.
14~24
...,:_ ·i
~ .. ;:
I
-·-•• ·• c
'~ \ .
' :.;~
;
I
I·
I -......
I
f,
'
I
I
I "-'':;
I
I ~
I
t ..-;.-
1
I
,....,*"J
' I
• 9
I ' ... ,
t
' I .
i
" .-,.~
. ·'
. \<
I ~
I ,,
I
I
I
t .), .. -
. ' .
t.
I
-~,
t
t ...
t
/
· ( i l) Recomme_nded S'i ze.s .·
~=~~r~h,!i's~sit~a~ ir~ansmYs~fan·~fias~--seen-'(fivlded, '·fnta . three:·=9"eo 9raphi ca1
sections each of Which has particular loading and environmental re-
qui remefits. ·
In the section from De vi 1 Canyon to Will ow to Knik Arm, there wi 11 ~
be three circuits ultimately. Line loadings are sufficiently heavy
to have a significant effect on conductor se.lection, and the eco-
nomic choi~e is considered to be: ·
- 2 x 954 kcmil, 45/7 Alumi.num Conductor Steel Reinforced (ACSR),
11 Rail" conductor .. This conductor will also be used on the two
circuits between Watana and Devil Canyon.
In the section from Knik Arm to University Substation, the line
loading is also heavy, but environmental considerations dictate the
use of two circuits instead of three and the ~conomic conductor size
is:
- 2 x 1,351 kcmil, 54/19 ACSR!t "Martin 11 conductor.
In the northern section from Devi 1 Canyon to Fairbanks, 1i ne load-
ings are light and the economic conductor size would be smal 1 er than
that allowed by RI and corona considerations. In this section the
minimum conductor that can be used is:
- 2 x 795 kcmi 1, 26/7 ACSR, 11 Drake 11 conductor.
However, since the intertie between Hillow and Healy will be pr.e-
built with 2 x 954 kcmi 1, it may be advantageous to standardize by
constructing all of the remaining Susitna transmission to Fairbanks
with 2 x 954 kcmil. This could be determined later.
14 .. 5 ... Substations
To follow
14.6 -Dispatch Center and Communications
(a) Existing Railbelt Dispatching Facilities
The main generating·and load centers are located in Fairbanks and Anchor-
age. Both areas operate independently of each other. It is proposed by
APA to conne.ct the two systems by a tie in 1984 which wi 11 be operated at
138 kV. The power· transfer capability wi 11 be approximately 70 MW.
Golden Valley Electric Association (GVEA) and Fairbanks Municipal Utility
System (FMUS) constitute the two major producers of electrical power 'in
Fairbanks. Although both utilities are intertied at 69 kV, they each pro-
vide their own dispatching. GVEA is responsible for maintaining frequency
in the Fairbanks area.
14-25
Chuga~h Electric Association (CEA), Anchorage Municipal Light and Power
(AMLP) and Matanuska Electric Association (MEA) are the utilities which·
serve the Anchorage area and· a11 areas north of Anchor.age, including Willow
and Palmer. MEA is mainly a distributing utility and imports its powe.r
from CEA and the Alaska Power Administration. --AMLP generates and -
---. distributes power in the Anc;.horage area. · CEA generates and -distributes
electric pow.er in Anchorage and adjoining areas. ~ach utility provides its
own control center facilities and is intertied at the 115 and 138 kV
level ..
Presently, CEA provides frequency control for the Anchorage are~o CEA has
its own system control center· that provides dispatching and supervisory
functions over its generating and substation facilities.
(b) 1993 Rai lbelt Power System
The introduction of Susitna hydroelectric power in the Railbelt area will
require several hundred miles of transmission lines fr6m the Susitna River
basin to Anchorage and Fairbanks. In fact, the ultimate development will
require approximately 850 miles of transmission, 5 switchyards and 2 hydro
generating stations at Watana and Devi 1 Canyon. Thermal generation at
Fairbanks and Anchorage will still be in operation. The installed genera-
tion capacity will be. over 2,000 MW at that time.
To operate such an en 1 arged Rai 1 belt system, a contra 1 system or energy
management system (EMS) wi 11 be required. This system wi 11 i flsure seeuri ty
of the 345 kV transmission lines and switchyards/substations operations.
The system will also exercise remote control and efficient dispatching of
the generating units in the Railbelt.
(c) Energy Management System Requirements
To provide an efficient and secure dispatching system for the Railbelt, the
following functions are proposed:·
( i) Supervisory Contra 1 and Oat a Acqui si ti on ( SCADA) Subsystem
Includes real-time system data acquisition; remote control of power
system devices; data base and data base management; data processing;
operation data logging and report generation; and man/machine int~r
face requirements.
(i i) Generation Control Subsystem
Includes automatic control of hydro and thermal units in the Rail-
belt area to maintain interconnected system frequency and inter-
change scheduling; economic unit operation; generation reserve eval-
uation; and monitoring of system generation performance~
(iii) Power Scheduling and Load Forecasting Subsystem
Includes the forecasting of system load and the scheduling of the
power system generation to meet load requirements in the most eco-
nomical and reliable way.
.•.•... :1
'!
t~
'---~·~ .·/':
.i' .If ::
!
i
f
I -~ ..
t
I
I
I
I
I
t
I
f
I .....
t
I
. .· ....... t.::
. ~~--·'-~---; ...... ,;;,_" ~~·~~.;;~·~~ ... .'<.~h.;..;...;_>••-'-·~· '""'"·"·-·'"'"'d
14-26
c_ ..
t
·I·· ~--'
t
I
t·
I
I
t
. I
I -
t~~-_
<' It
i•. ~~
' I
' t
-1-; ' '
t
.. ,._.
'. '
(i v) Energy Accounti_ng Subsystem
Includes collection, recprding, and processing of data power trans-
actions-among various utilities in the interconnected sys-tem; .al~o
the cost information and the savings/losses resulting from purchase/
sale. of power.
(v) sxstem Security_Subsystem
Includes the ability to evaluate system performance based on p.resent
and predicted system conditions and the ability to evaluate the
impact of probable contingencies (loss of generation, Toss of a
transmission line, etc.).
(vi) System Support Subsystem
Includes on-line/off-line functions that could be performed by EMS
to support engineering, accounting, and system operation organiza-
tions.
A more detailed description of the functional requirements is covered by a
report entitled, 11 Energy Management System (EMS) ... System Requirementsn
dated December, 1981 .
(d) Energy Management System Alternatives
An evaluation of alternative system configurations showed that two differ-
ent .approaches to generation centro 1 are possi b 1 e:
-Alternative I provides indirect control of generating units; and
-Alternative II provides direct control of generating units.
To formulate and evaluate the alternative configuration, the following
cri teri a were used:
-Configurations must fulfill functional requirements discussed above in
paragraph (c);
-Configurations must be technically, economically, and operationally main-
tainable through the life of the" systems (10 to 15 years); and
-Configuration must be technically feasible~ as well as proven.
(i) Alternative I System Configuration
The Alternative I system configuration is typical of the current
offerings of several EMS equipment manufacturers (see Figure 14-12,
EMS Alternative I System Configuration). The configuration is based
on the assumptions that:
-An in-plant, computer-based control system, located at Susitna
Hydroelectric Control Center will be provided;
14-27
-The Susitna in-plant control system wi11 directly control a11 hydro
gen~rating units and. the switching stations at ':Watana and Devil
"CatlYQ!J:.,._::gMs will determine generation participation requirements on
the unit level, but the units will be pulsed by the in-p1ant
system. The supervi sary contra l _actions for Watana and Devil
Canyon generating stations will be initiated at-EMS level, but the
control functions wi 11 be ·imp 1 emented by the in-plant contra 1
'System;
The northern and southern computer-based systems will receive gen-
eration participation requirements from the EMS, but participation
allocation and direct unit pulsing will be accomplished by these
systems; and
-EMS will directly monitor and control the following 345 kV substa~
ti ons:
. Ester;
. Willow;
. Kni k Arm;
• University; and
. Others, as required.
(i i) Alternative II System Configuration
The Alternative II system configuration is also typical of current
offerings of several EMS equipment manufacturers (see Figure 14 .. 13,
EMS Alternative II Syst·em Configuration). The configuration is based
on the assumptions that:
-An in-plant, computer-based control system, located at Watana, will
be provided to monitor generating units performance and control the
units;
;. All Watana and Devil Canyon generating units will be controlled
(raise and 1 ower) direct 1 y by EMS from system contra l center at
w·i llow; ·
-All northern and southern area generating units will be dir.ettly
controlled (raise and lower) by EMS, Willow Control Center; and
-The switching stations at Watana and Oevi 1 Canyon and the other"
four 345 kV substations will be directly monitored and controlled
by the EMS Control Center.
(e) Communication Requirements
Effective operation of Et4S is very dependent on transfer of data and
immediate response of supervisory functions such as control and
telemeter1ng. Various communication systems to determine the most reliable
and cost-effective communication media were evaluated.
14-28
' t •c
•'
~·< -·,
.
•... ·· .. --
I
il
t
' t
I
~~-~ v
-
I
t
t
t
t
I
I
l ..... ~
\ "·' 1 .• -; .
t
' 1-..
-.
l
' t
' t
I
'
' t
t
I
' I
I
' t
( i ) Power Line Carri. er
This system is dependent on the state of the transmission line and,
therefore, wi 11 not be avai 1 able when the 1i ne ts. down ..
( i i ) -Te 1 ephone
Telephone companies provide data transmission services but the
service is very erratic and unreliable for EMS applications.
(iii) Microwave
A microwave s.ystem is the most re l i ab 1 e and cost-effective sol uti on
for the EMS communications. It is-highly desirable to install a·
looped system for power system operation.
Microwave systems are line-of-sight propagation and have an average
standard transmission path of approximately 35 to 40 miles in an area of
flat terrain.. The cost was estimated for appro xi mate ly 17 towers and
repeater stations and without having the benefit of a detai 1 communication
analysis. A microwave system is recommended for this application;
(f) System Software Requirements
The EMS should be provided wi 11 all the software required to satisfy the
functional requirements described in paragraph (c) and all the software
functions described below.
The software should be the general-purpose operating system~ developed and
tested by a major computer supplier and verified through many i nsta 11 ati ons
in real-time applicationso It should provide a reliable, high-performance
environment for the concurrent execution of multiuser, time-sharing~ batch,
and time-critical applications. This software will consist of the fo1lowiftg
major components: ·
-Executive services;
-System fai lover and system restart;
-Diagnostic programs;
-Programming services;
-Special data base, CRT display, and log/generation compilers;
Engineering support; and
-Special I/O handlers.
Fortran compatibility of the software is essential, as most of the power
application programs will be written in a high-level language.
(g) Control Center Facilitx
The facility wi 11 be the nerve center of the APA po.-Jer system operations of
345 kV transmission network and the electric power generation. All deci-
sions concerning the operation and maintenance of the power system will be
implemented through this complexo The importance of this facility dictates
that its location be selected with a great deal of care.
14-·29
f
f.
· ( 1) Location of Site
The control center.must be located on a site that provides high
security against disruption of power system operations by human
intervention or by acts of God, Acts of human intervention that must
be considered are civil disturbances and terrorist activ"iti es.
Natural disturbances that could occur are floods, fires, and land-
slides.
Several additional factors that have a bearing on the suitability of
a site are:
-Land availability;
-Housing availability;
-Transportation accessibility;
-Educational facility availability;
Climatic conditions;
-Power avail-abi 11 ty; and
-Centralized location in the power system.
All of the above factors and the fact that a major S\.c-:itchyard is
already 1 oc ated in the area make it appropriate to recommend Wi 11 ow
as the location for the EMS center.
Willow has additional qualifications as a possible capital site. The
Willow center could also be the headquarters for the maintenance
staff for the transmissio~ n~twork between·Susitna and Anchorage.
The t~illow site also ha.: flat lands between it and Anchorage which
also reinforces the reconu-nendation to use microwave as the communica-
tion media.
(ii)· Control Center Building
The EMS control center building can be located on the same site as
the Wi 11 ow swi tchyard. The construction of this bui 1 ding wi 11
require special facilities. This is all described-in the 11 Energy
Management System (EMS) -System Requirements" report.
Figure 14.4 provides a conceptual layout of the Willow Control
Center. Thfs layout is based on a one-level building having a
total space of 14,537 ft2.
(h) Staffing Requirements
The functional organization of the EMS Control Center must efficiently and
cbmprehensively support all aspects of the operation and control of the
Railbelt's power system. This also includes not only the day-to-day opera-
tions, but also the coordination of power transmission and generation and.
the ongoing training of personnel to improve efficiency and effectiveness.
14-30
I
t
I
' I
t
I
t
t
I
·I
t
' I
I
.'j
I
I
'
I
t
f
' I
' I
'
' I
I
' I;
' I , ,
t
(i) fewer SysternOperattons Staff
The fo11owing operating staff are recommended:
... ,;One <:hi ef . operator; '"
-·Five ·senior operators;
-Nine load operators;
-One engineering technician; and
-One clerk.
The above organization can mai ritai n a 24-hour operation for 365
days a year.
(ii) Computer Applications
The computer applications section should be managed by a supervisor
of software applications. Reporting to this supervisor-should be
at least three additional software engineers charged with the
duties of mai ntai ni ng SCADA, Generation Contra llt and System Secur-
ity software programs.
(iii) Power Coordination
The power coordination group will be responsible for evaluat1ng
unit commitment runs, preparing interchange schedules, and perform-
ing after-the-fact power accounting, etc. This group will include
one supervisor, one power production specialist, one budget
specialist, two· power system engineers/analysts, two statisticians,
and one power scheduler. ·
{iv) EMS Maintenance Graue
The EMS maintenance group will be responsible for maintaining the
EMS hardware and software. As a minimum, this group should
include:
-One system hardware engineer;
-Two system software engineers;
-Two hardware technicians;
-Two RTU maintenance technicians; and
-One commu·n; cation maintenance technician.
{j) Budgetary Cost Estimates
This paragraph provides overall budgetary cost estimates for the develop-
ment procurement, system testing, and installation of EMS Alternatives I
and I I. Costs for the EMS Contra 1 Center and Microwave System are also
provided. These costs are representative of what Energy and Control
Consultants estimate as the middle price bids of such a project ..
The cost estimates for these configurations, microwave system, and EMS Con-
trol Center .are given in January., 1982 dollars for a fixed-price contract
that includes milestone pajfllents. Table 14.7 shows comparative cost
estimates.
1'4-31
f
(k) Recommendations
Alternative t, shown in rigure 14-.5., is recommended for the Railbe1t Energy
Management System as th.e most cost .... effecti ve and deslrab 1 e, system approach.
Uniike Alternative Il, Alternative I system approach allows generation
· control of the southern (Anchorage) and northern (Fairbanks} areas to
remain under theit respective utilities. Alternative I also encourages the
formation of regional control centers for each area. Presently, this is
· the trend in power system control to decentralize in large geographical
areas.
Alternative I is also marginally less costly than Alternative II.
Microwave is recommended as a communicating medi urn. Once provided, th i s
system will perform the following additional functions:
-Provide a transmission media for protective line relaying; and
-,"rovide reliable voice communications between the various stations. This
is very important in power system operations. ·
It is recommended that the El'1S Contra 1 Center be 1 ocated at Wi 11 ow within
the Willow Switching Station compound. This location has many advantages
and is centrally located in the southern Railbelt power system. It would
also be reasonable to designate this location as a maintenance center for
the transmission system. This area has lots of land for expansion. There
a 1 so appear to be some p 1 ans to provide. a highway crossing at Kni k Arm.. If
these plans materialize, Willow would only be one hour away by highway from
Anchorage.
14-32
"·
I
'
;1.·.· . .
t
I
t
I
t
t
t
t
I
' I
'
' t:
t
I . . .
I
I
I
I
t
I
' I
I
' a
I
t
'
'
' ,.
I . '". ,..,
LIST OF REFERENCES
(1) Susitna Hydroelectric Project Planning Memorandum -Subtask 8.02
Preliminary Transmission System Analysis Acres, 1981
(2) Upper Susitna River Hydroelectric Studies Report on Transmissic..,n System
U.S. Department. of Interior Corps of Engineers, 1975
(3) Anchorage-Fairbanks Tr ansmi ssi on Interti e Economic Feasibility Study Report
IECo, 1979
(4) Terrain Analysis of the North and South Intertie Power Transmission
Corridors, Prepared for Acres American, Inc~, and the Alaska Power
Authority by R&M Consultants, Inc., Anchorage, Alaska.
14-33
TABLE 14 .. 1: POWER TRANSFER REQUIREMENTS (MW)
INSTPLLED CAPACITY TRANSFER RE;QUIREMENT
Susitna to Susitna to
Year Watana Devil C"anyon Total Susitna Anchoraqe Fairbanks
1993 680 --680 578 170
1997 1020 -1020 867 255
Jll 2002 1020 600 1620 1377 405
TABLE 14.2: SUMMARY OF LIFE CYCLE COSTS
TRANSMISSION ALTERNATIVE 1 2 3 4 5
Transmission Lines 1981 $ X 10 6
Capital $156.70 $159.51 $133.96 $140.94 $159.27
Land Acquisition 18 .. 73 20.79 18.07 20.13 18 .. 65
Capitalized Annual Charges 127.34 130.14 107.43 112.83 126.91
Capitalized Line Losses 53.07 54.50 64.51 65.82 42.82
Total Transmission Line Cost $355.84 $364.94 $323.97 $339.72 $347.65
Switching Stations
Capital $114.09 $106.40 $128.:32 $120.64 $154.75
Capitalized Annual Charges 121.02 113.30 135.94 128.22 165.02 --
Total Switching Station Cost 2)5.11 219.70 264.26 248.86 319.77 --0
Susitna life Cycle Cost $590.95 $584.64 $588.23 $588.58 $667.42
I
t
I
I
t
I
I
t
I
t
t
I
t
t
I
I
'
' I .. ...
•.. ,, ... , ...
..
'
'.
: .. -, ..
I
t
'
' I
I
I
I
t ., ..
t
I
I
.. I
' .. ,
TABLE 14.3: .TRANSMlSSlON SYSTEM CHARACTERISTICS
NUmBer or-NUmber-& !h ze
Line. Section Len9th Circuits Voltage of Conductors
(ml.) (kVJ (komi!)
Watana to Devil Canyon 27 2 3~5 2 by 954
Devil Canyon to Fairbanks 189 2 ;345 2 by 795
Devil Canyon to Willow 90 3 345 2 by 954
Willow to Knik Arm ;;a 3 345 2 by 954
Knik Arm Crossing* 4 J 345
Krtik Arm to University
Substation 18 .2 345 2 by 1351
*Submarine Cable
!i
TyPe
1. Technical
-Primary
-Secondary
2. Economical
-Primary
-Secondary
3. Environmental
-Primary
-Secondary
TABLE 14e4: fECHNICA!-t ECONOtUC, AND EN.VtRONMENTAL C!HlERlA
USED IN CORRIDOR SELECT ION.
Criteria
General Location
Elevation
Relief
Access
River Crossings
Elevation
Access
River Crossings
Tinbered Areas
Wetlands
Development
Existing Transmission
Right-of-Way
Land Status
Topography
Vegetation
Selection
Connect with lntertie near Gold Creek, Willow,
and Haaly. Connect Healy to fairbanks. Con-
nect Willow to Anchorage.
Avoid mountainous areas.
Select gentle relief.
Locate in .pro)(imity to existing transportation
corridors to facllitate maintenance and repairs.
Minimize wide crossings.
Avoid mountainous areas.
Locate in proximity to existing transportation
corridors to reduce construction costs.
Minimize wide crossings.
Minimize such areas to reduce clearing costs ..
Minimize crossings which require special designs.
Avoid existing or proposed developed areas.
Parallel.
Avoid pri~ate lands, wildlife refuges, parks.
Se:i.ect gentle relief.
Avoid heavily timbered areas.
, .. ,, I
I
t
I
t
I
' I
I
I·
I
I
I
t
I
I
I
' I
"i ,::
I
t
I
·t
I
I
I.
I
I
I
I
I
I·
I
I
I
I
TABLE ·14. 5; TECHNICAL, ECONOMIC AND ENVIRONMENTAL CRITERIA
USED IN CORRIDOR SCREENING . .
Technical
Primary
Secondary
Economic
Primary
Secondary
Environmental
Primary
Topography
Climate and Elevation
Soils
Length
Vegetation and Clearing
Highway and River Crossings
Length
Presence of Right-of-Way
Presence of Access Roads
TopO-Jraph}'
Stream Crossings
Highway and Railroad.Crossiqgs
Aesthetic and Visual
Land Use
Presence of Existing Right-of-Way
Existing and Proposed Development
Secondary , ··
Length
Topography
· Sdils
C.ultural Reservoir
Vegetation
Fishery Resources
Wildlife Resources
TABLE 14;6: SUMMARY OF SCREENING RESULTS
RATINGS
Corridor E:nvo . Econ. Fecfi •
-Southern Study Area
······(;-] ABC' C~ c c
*(2) ACfC A A A
(3) AEFC F c A
-Cental Study Area
*(1) ABCO A A A
(2) ABECO F c -C
(3) AJCF c c c
(4) ABCJHI F F F
(5) ABECJHI f F F
(6) CBAHI f' c F
(7) CEBAHI F F c
(8) CBAG F F c
(9) CEBAG F F c
(10) CJAG F F c
(11) CJAHI F c c
(12) JACJHI F F c
(13) ABCf A c A
(14) AJCO c .A A
(15) ABECF F c c
-Northern Study Area
*(1) ABC A A A
(2) ABOC c A c
(3) AEOC F c F
(4) AEf F c F
A = recommended
C = acceptable but not preferred
f = unacceptable
*Indicates selected corridor.
I ;
Summarx
c
A
F
A
F
c
f'
F
F
F
F
f
F
f
F c c
F
A c
F
F
-,
I ·~·· .
t
I
t
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
'I
I
I
I
I
I
TABLE 14"' 7; EMS ALTERNATIVES I AND H
COMPARATlVE COST ESTIMATES
EMS Project
· Hardware
Software
Auxiliary
Internal (APA costs)
Susitna In-Plant Control System
Hardware
Software
Auxiliary
Internal (APA costs)
Microwave System
EMS Control Center Building
TOTAL
Alternative I
$ 2,942,000
:3,956,000
1,210,000
3,416,000
$ 11,524t000,
$ 1,131 ,ooo
1,200,000
750,000
1,770,000
$ 4,851,000
$ 4;920,000
$ 3,853,140
$ 25' 148,140 -
Alternab.ve 11
$. 31072,000
4,200,000
1,:350,000
3,606,000
$ 12,228,000
$ 1,094,000
1,200,000
700,000
1,875,000
. $ 4, 869 ,ooo
$ 5,100,000
$ 3,853,140
$ 26,050,140
•.
.···.f' .. · ·.· : ;
-. ' .
-... ····1. ··.
-.
~> ..
•. e,,
-·
I
••
':1
I ,,
I
I
I
-·
·I
.. ,.
:I .,
1-
·01.
r------------, ....... ? . . I.
. .,.._ SUIMMICE c:AILE ff
I UNDEil KNJK ARM' .
·...--+-------( •f-n--
....___~-Hr--
' I . .. I
·~ L---
75MVA
3-45-l~l<V
--1~-fr--
II MI.
-KNIK 'ARM
--,
J
I
I
I
I
WILLOW
UNIVERSITY
(ANCHORAGE)
·v..l...v 250 MVA
r~ 345-
1 ·.·'""r" 115l13.:S t(\1
,L .. 1 ..
4y-' .,...
""""-r-~ 1
I
\ l l . -
1 I
4 X 150 MW UNITS • ' . "'!
26 MI.
WATANA
I I
CHUGACH ELECTRfC -ASSOCIATION ANCHORAGE MUNICIPAL 6 x 170 MW UNITS·
LtGHT 8 POWER
.RAILBELT 345 K.V TRANSMISSION-SYSTEM. SINGLE LINE DIAGRAM
195ML
SHUNT
REACTOR
ESTER
{FAIRBANkS )
DEVIL CANYON -
STAGING. LEGEND
---1993
----2002.
FIGURE 14.1
I
I
I •
I
I
I ,.
I
I
I
I
I
I
-.., ..
"
--~--"ALTERNATIVE TRANSMISSION LINE CORRIDORS
SOUTHE,RN STUDY A.REA
•
loiT,!oltlll~~ .......
,/-
,....."" DEHAI.f J
~Tfj· c ~~~------~ ..
L E.G E _N 0
----STUDY CORRJDOR
• • • • •• • • • • • • • • I NTERTlE (HYPOTHETICAL)
0 5 ID ~~~----iiiiiiil SCALE IN MILES
FIGURE .14.'2 ..
I
I
I
I
I
I.
I
I
I
I
I·
I
I ,,
I
I
I
I
' ·I
ALTERNATIVE ":·~NSMlSSlON LINE CORRIDORS
Ctk!NRAL STUDY AREA
•)
LEGEND
---STUDY CORRIDOR
• • • • • • ~ • • • • -• • I NTERTI E {HYPOTHETICAL)
0 5 10 ...-; d
SCALE IN MILES
FIGURE 14.S 1111·
I
I
•••
. '
I
••
I
··I
'I
I
I
I
I
.I
;I
I
I
:I ,.
I
@)· .
I ';
i~
~.
J
t
#
·~
r:/
'!
-~.; • :~ ... 1
·'-\ 1 .
~-LDCA~I(m MAP 't .,._) __ .... ~............ ..~-.,.....---., ... , ... , .... ___ .......,.
LEGEND
----STUDY CORRIDOR
............... I NTERTIE
(HYPOTHETICAL}
0 5 . 10
SCALE IN MILES
.. II'
1
I
. ··.· ....•.•.... ··· "' '·'
I
I.
I
I
I
I
I
I
I'
I
•••
I
:I
.>.
.,I ...
ALASKA
RAU .. ROAD
'~ .. z:.··
••• ••••• ••••• •
ALASKA RANGE
L.E G END
--0--HIGHWAY
· 1 f RAILRCAO
-·-·-TRANSMISSION LINE ROUTE
0 P.ROPOSED SWITCHING STATION
\
DENALI \
NATIONAL PARK \\
\
\
\
\
\
\
\
\
\
\
\
\
\
. WATANA
GOLD CREEK
THE PROPOSED INT.ERTIE 'FROM Wlt,.LOW T() HEALY WILL BE CONSTRUCTED
FOR 345 KV CAPABILITY AND INITIALLY OPERATED AT 138 KV.
fF SUSITNA IS PROVED FE~SISLE, THE FULL 345 t<V CAPACITY WILL BE UTlLlZED,
ANCHORAGE TO
PROPOSED TRA,NSMISSlON LlN ROUT£
TALKEETNA MOUNTAINS i
ALASKA
RAILROAD
12 24 MlLES SCALE O.!!·~~~iiiiiiiiiiiiiiiiiiiiiiiiliiiiil
·;a_
I
--=1'
I
I
-1
-I
~.
I
I
I
I
I
I
:1
I;
·I
I
•••
:
0
. -I
"' Q)
'1][
~--~3~3~·~~o-"~~~~~~~~3~3~·t~~~--~
.....
:t:
C)
Uj
% ,.._-GUY
l&l
(!)
< 0:::
Ul
~
DETAIL B
DETAIL A
-.
RECOMMENDED 345 J<V TANGENT TOWER
\
·l
\
'
I
' I ,..,
T
~ . . ..... ~
=====-v~~~~~ . .
\
-· · · ·. · · . · ·.-. CiiAMSER HOLE TO
: ACCEPT HEAVY DUTY
GUY-THIMBLE WITH 1
1 GUY-WIRE, a PREFORMED
GUY ATTACHMEttT
DETAIL A
rr---Tl
l! J J ;J II
11
I
I
DETAIL B
CLEARANCE iO PILE
X -FRAME GUYED STEEL TOWER
' .
'
..
I
' r
SECTtON A-A.
SECTION B-8 · -CHAMBER HOLE TO
ACCEPT HEAVY DUTY
·GUV THIMBLE WITH
(;UY WIRE 8 PRE-
FORMeD -GUY ATTACH-
MENT
1
'1
..
I
I
I
I
I
I
I
I
I
I
I
I
.
I
I
I
I
I
.I
,I
< ~ \.,~ • .. •
l
2 EACH, 3. CU. YD. CONCRETE
2 ANCHORS 1 H REBAR
lN 4 11 9.' X e:.o" HOLES
2 .EACH t
l ANCtiORt lt ~t:eAH .
IN 4"0 X 8~0 u HOLE$
ROCK FOUNDATION
.... -. -·. . ;~II .ST·EEL 4::: t.At.,;n·, · ' · ·
H-PILES, 2.5'-o" LONG
2 EACH 1
MULTI-HELIX ANCHORS
15!..0" LONG
STANDARD FOUNDATION
2 EACH,
MULTI-HEUX ANCHORS
G 301-011 LONG
2 EACH, 12 u STEEL
H-PILES, 50' LONG
WET LAND FOUNDATION
TRANSMISSlON TOWER FOUNDATION CONCEPTS
FIGURE 14.7 [iii
- - -•.. ·~-·-:····-- -·--· ·-- - - --; - -
NORTHERN AREA
CONTROL SYSTEM
G ------RTU
COMPUTER COMPUTER
~-----------------~
PERIPHERALS
MAN/MACHINE
INTERFACE
COMMUNICATION SUBSYSTEM
SOUTHERN AREA
CONTROL SYSTEM
G -------RTU
SUSITNA HYDROELECTRIC
CONTROL CENTER
G ------· RTU RTU r-------RTU·
SUBSTATION
RTU 5
ENE_RGY fvlANAGEMENT SYSTEM, ALTERNATJVE I~ SYSTEM CONFIGURATION· FIGURE 1.4.12 ~~~~~
- - - - - - - - - - - -·-- -·c-- - -
NORTHERN AREA
CONTROL SYSTEM
G -·-----RTU
COMPUTER
PERIPHERALS
MAN I MACHINE
INTERFACE
COMPUTER
COMMU~·,cATlON SUBSYSTEM
SOUT!:ERN AREA
CONTROL SYSTEM·
G ------RTU
--RTU
SUSITNA HYDROELECTRlC
CONTROL CENTER
G
I ~ RTU ---.·-.----RTU5 f--rru ~---·---RTU
WATANA/DEVIL CANYDN SUBSTATION RTU
SUBSTATIONS
ENERGY MANAGEMENT SYSTE:M, ALTERNATIVE n, SYSTEM CONFIGURATION FIGURE 14.131 MIR I I
---· - - - ----- ---- - - - - - - -
l
..
..... rE--------------------170 _______ ...__ _________ ___,~
0
MECHANICAL AND
FACILITY SUPPORT
)200 SQ. FT.
CONFER. · . TRAIN lPROG.
ROOM ROOM
400 SQ. FT. 400 SQ. FT.
OFFICE AREA
1500 SQ. FT
20
FEET
40
COMMUNICATION STORAGE
ROOM 300
600 SQ. FT. SQ. FT.
EMS EQUIP;
MAlNT. ROOM
900 SQ. FT.
HALL 7.5 FT. WIDE
BATTERY ROOM
350 SQ. FT.
UPS ROOM
350 SQ.FT.
EMS EQUIPMENT
ROOM
1500 SQ. FT.
ENG. KITCHEN a MEN
LAV.
450
SUPPORT LOUNGE
LAV. a
KITCHEN
350 SQ. FT.
DISPATCH
AREA
650 SQ. FT.
DISPATCHING
· ARENA
1500 SQ. FT.
LOBBY
450 SQ .. FT.
WOMEN
LAV.
450
MAI-.AGEMENT
AREA
l
t {(0
~ ;(ij)
l
L•
600 SQ. FT. 900 SQ. FT. . SQ. FT. SQ. FT. 637 SQ. FT.
TOTAL; 14,537.5 SO. FT.
WILLOW SYSTEM CONTROL GENTER,
FUNCTIONAL LAYOUT
.___
ENTRANCE.
FIGURE 14.14 l111R]
\1
I
't·
I
I
I
I
I
I
.,._
••
I
.. I.
I
I
-I
I
I
I
" I
-~ NORTHERN AREA .
...... ~..-~ CONTROL _SYStEM .. ...,_.. ____ ............ ..,... ___ __,
-~-FAIRBANKS
TO
ENERATORS
--
ESTER
SUBSTATION
WILLOW
SUBSTATION
KNIK ARM
SUBSTATION
UNIVERSITY
SUBSTATION
SOUTHERN AREA
l t----------__,
ENERGY MANAGEMENT
SYSTEM
WILLOW CONTROL
CENTER
I
I
J •
---
.. .-CONTROL SYSTEM-r------------~
ANCHORAGE
TO
GENERATORS
WATANA
SWITCHING
STATION
TO GENERATORS
J~ j ~ H A l ~~-· ~'
SUSJTNA
HYDROELECTRIC
CONTROL CENTER
l
t -' .
1( w 11' ~
TO GENERATORS
DEVIL CANYON
SWITCHING
STATION
ENERGY MANAGEMENT sYsTEM; ALTERNATIVE r, · IAPBmJ
CONFlGURATION BLOCK DIAGRAM · F.lGURE 14.15 U [I)
0
I
,I
I
I
I
I
I
I
I
·. .
•••••
I
15 -PROJECT OPERATION
This s..ection describes the operat·ion of the Watana anr. 'lil.Canyon power plants
int.he Railbelt electrical system. Under current cona;t1ons in the Railbelt a
tota,,1 of nine utilities share responsibility for generation and distribution of
electric power with limited interconnections.
The development in Sections 6, 8, and 14 of th~ Susitna project, size and
schedule of on-1 i ne dates, and the associated transmission 1 ine requirements \'las
necessarily based on the assumption that a single entity \'lould eventually be set
up to optimize and control the dispatch and distribution of electric power· from
all Ratlbelt sources. It is not the ptirpose of this report to discuss how thts
entity should be s,'::ructured or come about. However, it is important to note
that the Susitna project will be the single most significant power source in the
System. Careful cr,nsideration is therefore essential of the dispatch and dis~
tribution of power from all sources by the most economical and reliable means.
The general principles of reliability of plant and system operation, plant oper-
ation and reserv::ir regulation, stationary and spinning reserve requirements and
maintenance programming are also discussed. Estimates of dependable capacity
and annual energy production for both Watana and Devil Canyon are presented ..
Operating and Maintenance facilitles and procedures are described and the pro-
posed performance monitoring system for the two projects is also outlined ..
15 .. 1 -Plan~ and System Operation Requirements
The two plants comprising the Susitna Project wi 11 represent about 75 percent of
the system capacity, having an i nstall!!d capacity of 1620 MW in a total system
installed capacity of 2100 MW in the year 2010. In vis_'W of its large capacity
and the extent of its influence on the operational characteristics of the power
system, it is appropriate that the Susitna project operation should be discussed
within the framework of general power system operation considerations of economy
and security.
Planning studies discussed in Section 8 were primarily concerned \vith selection
of plant installed capacity such that an optimum installation could be provided
to meet projected generation requirements over the life of the project, which
may be. considered as 50 years or more.
The main function of system planning and operation control to be discussed in
this section is concerned with the allocation ·of generating plant in the system
on a short-term operation basis so that the total system load demand is met by
the available generation at minimum cost consistent wit~ the security of
supp rY.
The ger1era 1 objectives are generally the same for I ong-term planning or short-
term operational load dispatching, but with important differences in the latter
case. In the shor-t-term operational case the actual state of the system dic-
tates system security requirements overriding economic considerations in load
dispatching. An important factor arising from economic and security considera-
tions in the system planning and operation is the provision of reserve capacity,
both as stationary reserve as well as spinning reserve.
15-1
~-
The basis of system operation is the demand to be met at any moment and is the·
aggregate of all consumers• demands in the interconnected Ra11belt System.
Figure 15.1 shows the daily variation in demand during typ,;cal winter and summer
wee.kdays afld the sea~onal v.ari ation in monthly peak demands for estimated loads
in a typi (Ca1 year (fhe year 2000).
15.2 -Gent~ral Power Plant and System Rai lbelt GriJeria
The power plants and electrical system are planned and constructed in such a
manner that they can be operated so that the more probable contingencies can te
sustained without less of-load. Less probable contingencies are also examined
and the consequences to the system are determined and evaluated. The more prob~
able contingencies are usually defined as those that occur once in 5 or 10
years. The less probable contingencies have a probability of occurrence of once
in 50 or 100 years. Automatic load relief (ancl when necessary in extreme cases,
load shedding) is provided to minimize the probab;lity of total shutdown of area
load which becomes isolated by multiple contingencie.: The following are basic
reliability standards and criteria gene~dlly adopted in the industry for power
systems. Further details are described in Section 8 for generation p~itnning, in
SectiGi•S 12 and 13 for the power plants, and in Section 14 for thf! transmission
system.
(a) Installed Gene-rating Capacity
Sufficient generating capacity is installed in the system to insure that
during ee.ch year the probabi 1 i ty of occurrence of load exceeding the avai 1-
able generating capacity shall not be greater than one day in ten years
(LOLP of 0.1).
The factors affecting the calculation of probability iryclude the ci-laracter-
istics of the loads, the probability of error In load forecast, the
scheduled mai ntenanc·e requirements for generating units, the forced outage
rates of generating units, limited energy capacity of plants, effects of
interconnections, and tr·ansmission transfer capabilities. The calculation
of LOLP is done in the generation planning studies described in Section 8.,
(b) Transmi ssio~.System Cap:abi lity
The high-vcltage transmission system in general and that associated \'lith
the project in particular, should be operable at all load levels to meet
the following unscheduled single or douLle contingencies without instabil-
ity, cascading or interi"uption of load:
-The single contingency situat~on is the loss of any single generating
unit, transmission line, transformer, or bus (in addition to normal
scheduled or maintenance outages) without exceeding the applicab1s ;emer-
gency rating of any facility.
The double contingency situation is the subsequent outage of any remain-
ing equ·i pmen't, 1 i ne o;· subsystem without exceeding the short time emer-
gency rat 1 ng of any faci 1 ity.
15-2
·I·_-.-
. l_
I
I
I
I
I·
I
aJ
I
I
I
I
••
'I
II
I
I
I
I
••
I
I
I
I
I
I.
I
I
I
I
I
I
'··
I
.. ~·-
1
I
' •. : . ;
~, ,': . : :''
' . ' .'~ ;-.,·.·:< ' .: .
(c)
ln the. s;1ngla contingen_cy situation, the power system must be capable of
readjustment so that all equipment will be loaded within normal ratings,
and in the double contrnuency situation within emergency ratings for the
probable duration of thi:! outage.
During ahy contingency:
-Sufficient rt=active power (MVAR) capacity with adequate controls are
installed to maintain acceptable transmission voltage profiles.
Q
The stability of the power system is maintained without loss of load or
generation during and after a three-phase fault, cleared in normal time,
at the most critical locatio~.
It is impossib 1 e to anticipate or test the system for all contingencies
_that can· occur in present or future confi gur at ions.. Typi ca 1 ex amp 1 es of
the less probab 1e contingencies are:
-Sudden loss of the entire generating capability of anypm'ler plant for
any reasons.
-Sudden loss of all transmission lines on a single right of way.
-Sudden dropping of a very 1 arge 1 oad at a major load center.
The abov~ stated general principles of reliability and security have been
generally applied in the design of the project and system and are described
and referred to in greater deta i 1 in the v·ari ous pertinent sections of this
report. These principles constitute the basis of project operation plan-
ning and criteria described in this section and else'lhere in the report.
Summary
Operational reliability criteria thus fall into four main categories:
( i)
( i i)
(iii)
Loss-of-lead probability (LOLP) of 0.1 or one day in ten years, is
maintained for the recommended plan of Susitna project and,$ystem
operation through the year 2010 (Section 6, Generation Planning).
The single and double contingency requirements are maintained for
any of the more probable outages in the plant or transmission
system.
System stability and voltage regulation are assured from the elec-
trical studies (Section 14, Transmission Facilities). Detailed
studies for load frequency control have not been done, but it is
expected that the st i pu 1 a ted cri teri a wi 11 be met with the more than
adequate sp·:nning re-s·erve capability with six uni.ts at VJatana and
four units at Devil Canyon .
15-3
·-·
Jl .. ,,
,.~
.. ; . ·~." ·.-·
(i v) ·The loss .Qf all S~~ftna transmiSsion lines on a single ri ght~'of-W(LY
has a low~ level of probabtlity as described in Section-18 JJtider Rlsk
Analysis •. In the event -of the loss of a1l ~1ines 1 the~hydroplant~; at
Wat_ana. and De vi 1 Canyon are best su5 ted to t"estore po~Jer supply ··
qufckly :after. the first line· is reslored since th~y ate· desi'gne() for
.. black start 11 oper.ation. In this t .. espect., hydrQ -plants ar~ super tor·
to thermal plant.~. because of their inherent black start capabi 'Hty
for restoration of supply t~ a large system. ·
15.3 -Economic Operation of Units
The Central Oi spatch Control engineer has the responsibility of deciding which
generating units should be run at any particular time. Decisions are ma:de on
the basis of a number of different pieces of inform_ation, including an ~'~order ...
of-merit•' schedule, short-term demand forecasts, limits of operation of units
and unit maintenance schedules. ·
(a) Merit-Order Schedule
In order to decide which generating unit should run tomeet the system
demand in the most economic manner, the Contra 1 engineer is proV'ided with
information of the running cost of each unit in the form vf an uorder-of-
meritu schedule. The schedule gives the capacity and fuel costs for
thermal units-, and reservoir regulation limits for hydro plant'5.
{b) O~~imum Load Dispatchina
One of the most important functions of the Control Center is the accurate
forecasting of the: 1 oad demands in the various areas of the system. Are a
demand forecasts up to 8 hours ahead of unit loading are based on regional
short-range weather forecasts for an estimate of heating and lighting
demands plus light or heavy industry loads. Short-term forecasting up to 1
or 2 hours ahead is more difficult and remains the key factor to the secure
and economic operation of the system.
Based on the demand, bast c power transfers between areas and an a llawance
for reserve, the tentative amount of generating plant is determined, taking
into consideration the reservoir regulation plans of the hydro plants. The
type and size of the units should ~lsn be taken into consideration for
effective 1 oad di spat chi ng. ·
In a hydro-dominated power system, such as the Susitna case, the hydro
unit wi 11 take up a much greater part of base load operation than in a
thermal dominated power system. The hydro units at Watana typically are
well suited to load following and frequency regulation of the system and
providing spinning reserve. Greater flexibility of operation was a signif-
icant factor in the selection of six units of 170 MW capacity at Watana,
rather than fewer, larger size units.
No significant load following can be done by the Devil Canyon units due to
environmental constraints as described in Section 15.6.
15-4
' ..• ,!
" '•
; .... :
··I·
a:
I
I
,>"1
II
I
I
I'
I
I
I
I
I
I
I
.~
I
;I:
, I
,,.:,.1 . .,.
: . ,,.
I
I
I
I
I
~I
I
I
'I
I
I
I
'.1
·I
( ') -~ c·-_
(d)
Operating Limits of Units
. ~
Ideally; the plant having the lowest fuel costs should be. allo~ated 1oao·
for as long as possible, and the most expensive p.l ant required to meet--the
peak demands for as short a-time as possible. In practice, it ls not~ pos-·
sible to meet this ideal situation due to security requirements for the
system and the characteristics of the generation plant. There are strict
constraints on the minimum load and the loading rates of machines and to
dispatch load to these machines requires a system wide dispatch program
taking these constraints into consiaeration. In general, hydro untts have
excellent startup and load following characteristics, thermal units have
good p.art-loading characteristics.
Typical plant lo.ading limitations are given below:
(i) Hydro Units
~ Reservoir regulation constraints resulting in not-to-exceed maxi-
mum and minimum reservoir levels~ daily or seasonally.
-Part loading 0f units is impossible in the rough zone of turbine
operation (typically from speed-no-load to 50 percent percent
load) due to vibrations, arising. from hydraulic surges.
(ii) Steam Units
-Loading rates are slow (10 percent per minute).
-The units may not be ab 1e to meet a sudden steep rate of rise of
1 oad demand.
-Usually have a minimum economic shutdown period (abou~ 3 hours).
-The total cost of using c·anventional units i.nclude banking~ rais-
ing~ressure and part-load opertions prior to maximum economic
operation.
(iii) Gas Turbines
Cannot be used as spinning reserve because of very poor efficiency
and reduced service life.
-Requires 8 to 10 minutes for norma} start-up from cold. Emer-
gency start up times are of the order of 5 to 7 minutes.
Optimum Maintenance Program
An important part of operational planning which can nave a significant
effect on operating costs is maintenance programming. The program speci-
fi. es the times in the year and the sequence in which p 1 ant is re 1 eased for
!l· maintenance. Monthly, yearly, and 5-year maintenance schedules are pre-
pared. In a large interconnected system with minimum reserves, optimum
I
I
I' \
\
maintenance progratl1mipg" uses h~euristic methods.. The program planning takes
into consideration the avai 1abi 1 ity of trained repair and/or maintenance
personnel. ·Further details of Wat.anaqmdDevi 1 Canyon power plant mainten-
ance. prpgrams are given in subsection 15 ... 8.
15.4 -Qnit Oeeration Securi~y Criteria
During th~ operational load dispatching conditions of the power system, the
security criteria often override the economic considerations of merit-order
scheduling of the various units in the system. It is impossible to anticipate
all the probable conti ngencles of operation, hence an operati anal approach im-
plies the use of conttn:uous on-line data updating the state of the system for
the information of the operator. Projected !=lectrical power system analyses are
then carried out at frequent intervals to estimate the security of -operation.
Also important in consideration of operational security are system response,
load-frequency control and spinning reserve _capabilities.
(a) Power System Analyses
(b)
During the planning stages, system stddies are carried out for all credible
generation and network changes and probable contingencies. The trans-
mission system studies (Section 14) were undertaken to check the more
probable cases for load flow, short circuit and transient stability. The
load flow studies determined the voltage levels and reactive power
compensation for various plant loadings to meet the various load demands up
to the year 2010. The transient stability study determined that the system
was stable due to a transmission system fault resulting in the outage of a
critical transmission line.
Load-fr~quency response studies determine the dynamic stability of the
system due to the sudden forced outage of the 1 argest unit (or generation
block) in the system. The' generation and load are not balanced and if the
pick-up rate of new generation is not adequate, loss of load wi 11
eventually result from under-voltage and under-frequency relay operation,
or load-shedding. The aim of a well-designed high security system is to
avoid load-shedding by maintaining frequency and voltage within the
specified statutory limits.
System Response.and Load-Frequency Control
To meet the statutory frequency requirements, it is necessary that the
effective capacity of generating _plant supplying the system at any givan
instant should be in excess of the load demand. In the absence of detailed
studies~ an empirical factor of 5/3 times the capacity of the largest unit
in the system is normally taken as a design criterion to maintain s.ystem
frequency within acceptable limits in the event of the instantaneo'lis loss
of the largest unit. The factor 5/3 allows for the maximum dip in the fre-
quency of the system. It is recommended that a factor of 1-1/2 times the
lar_gest unit size be considered as a minimum for the Alaska Railbelt.sys-·
tern, with 2 times the largest unit size as a fairly conservative value
(i.e.~ 300 to 340 MW).
15-6
..•. ~-";
~ .~ ·•,
-~ \
1-.• ,,
. · .. ~~ : :_·· ~
. :, c' i
I
I
I
I
I
.<:
I
I~
I
I
I
I
I
I
li
I,
I
-_;;:.I~L
'1 ., .• J"li-;-
\; 'i{:
~~~~~"-'-'='--'-'--~--'-"----'-'--'-'---'~~-~~~~. -~-'--'"-~~--'-'-'-"_._;__""-----'~~-~~'--'---'-~-"-----'------'-~~~· ·~--~ ~.,__ ---~"-~~·~~~
.·.····, . . .
I ' ..
I
I
I
I
I
I
I
I
I
I
I,
I
I
I .~ ,.. :.
Upon sudden loss of ~eneration., .additional power may ini.tfally be derived from
,. the inertia of the rotating. machjnes (in tne first few seconds) and then from
the spinning reserve depending on its governor 4ction in the next 10 to 20
seconds •. · Th.e mi nimurn system frequency should in the. meantime. be' conta:i ned ·with ...
in the specified statutnry limits ti 11 ste~dy state oper.ation is reached.
From preliminary studies regarding the plant response in the system, it appears
that Watana is best suited for system frequency contra l and regulation. The
Watana.unit being the largest in the system could be put under sensitive gover-
nor control. Small generating stations and base-load thermal plant would he
normally given steep governor drops so that they maintain their scheduled power
output despite small changes in frequency.
The quickest response in system generation will come from the h-ydro units. The
1 arge hydro units at Watana and De vi 1 Canyon on spinning reserve can respond in
the turbi ni ng mode within 30 seconds. This is one of the parti cul ar1y important
advantages of the Susi tna hydro units. Gas turbines can only respond in a·
second stage operation within 5 to 10 minutes and would not strictly qualify as
spinning reserve. If thermal units are run part-loaded (example, 75 percent),
this would be another source of spinning reserve. Ideally, it would be
advantageous to provide spinning reserv·e in the thermal generation as well in
order to spread spinning reserves evenly in the system, with a compromise to
economic loadiqg resulting from such, an operation ..
· Det ai 1 ed load-frequency and spinning reserve studies should be done in the
design stage of the project.
(c) Protective Relaying System and D&vices
The primary protective relaying systems provided for the generators and
transmission system of the Susitna project are designed to disconnect the
-· faulty equipment from the system in the fastest possible time. Independent
protective systems are installed to the extent necessary to provide a
_fast-clearing backup for the primary protective system so as to limit
equipment ciamag·e, ·to 1 imi t the shock to the system and to speed restoration
of service.. The relaying systems are designed not to restrict the norma 1
or necessary network transfer capabi 1 iti es ~f the. power system.
15.5 -Dispatch Con,tro 1 Centers
The operation of the Watana and Devil Canyon power plant in relation to the cen-
tral_ dispatch center can be considered to be the second tier of a three-tier
control structure as follows:
-Centra 1 Dispatch Contra 1 Center ( 345 kV network) at Will ow: manages the main
system energy transfers, advises system configuration and checks overall
security.
-Area Contra 1 Center (Generation connected to 345 kV system, for example,
Watana and Devil Canyon): deals with the loading of generators connected di r-
ectly to the 345 kV n-etwork, switching and safety precautions of local sys-
tems, checks security of interconnections to. main system.
15-7.
... District ·or''~ho~d Centers, (138 kV and 1 ower voltage networks): generation and
distributton at lower voltage levels .. ·
For the Anchorage' andJ~,ai.tbanks..o-.area.~ the district center funct 1 ons. are
i ncorpor·ated Jrt the respective ,area contra 1 centers. .
The details of. the CentraJ Oi. spatch Control Center and of the· Watana Area Con-
trol Center are given in Section 14. Each generating unit at Watana and Devi 1
Canyon is started up, loaded and operated and shut down from the Area Contra 1
Center at Watana according to the loading demands from the Central Dispatch Con-
trol Center with due consideration to: c
-Watana reservoir regulation criteria;
-Devi1 Canyon reservoir regulation criteria;
-Tu~"bine l9ading and de-loading rates;
-Part loading and maximum loading characteristics of turbtnes and generators;
-Hydraulic transient characteristics of waterways and turbines
-Load-frequency control of demands of the system; and
-Voltage re~u1ation requirements of the system.
The Watana Area Control Center is equipped with a computer-aided control system
to efficiently carry out these functions. The computer-aided control 'System al-
lows a minimum of highly trained and skilled operators to perform the control
and supervision of Watana and Devi 1 Canyon plants from a single control room.
The data information and retrieval system will enable the performance and alarm
·monitoring of each unit i ndi vi dua 11 y as we 11 as the p 1 ant/reservoir and project
operation as a whole.
15~6 -Susitna Project Operation
A reservoir simulation model ~·1as used to evaluate the optimum method of opera-
tion of the Susitna reservoirs and power plants at Watana and Devil Canyon~
Substantial seasonal· as well as over-the-year regulati.on of the river flow i:s
achieved with the two reservoirs. The simulation of the reservoirs and the
power facilities at the· two developments was carried out on_ a m~mthly basis to
assess the energy_potential of the schemes, river flows downstream and flood
control possibilities with the reservoirs. Details of the computer model are
described 1n Appendix B~2. The following paragraphs summarize the main features
of reservoir operation.
(a) Reservoir Operation
Gross storage val ume of the \~at ana reservoir at its normal maximum operat-
ing level of 2185 feet is about 9.5 million ac/ft which is about 1.6 times
the mean annual stream runoff (MAF) in the river at the dam site. Live
storage of the reservoir is about 4.3 million .. ac!ft (75 percent of MAF).
Devi 1 Canyon Reservoi.r has a gross storage of about 1.1 million ac/ft and
live storage of 0.34 million ac/ft.
. 15-8
.-., .. _ .. ·.
,.,-::.
I ,,
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
'':1······.
' .
I
I
I
I
1·,
!
I
I
I
I
I.
I
I
I
I
I
<I
The reservolr.~imulatton model uses est'imated historical month'iy
streamflows at the damsites for 32 years of avai 1 able records, reservoi ro
characteristfcs~ and _power facility parameters as a basis of estimating the
energy potentials of the developments it Hydrological, en vi ronmenta 1, .
equipment and geotechP.:ical constraints were incorporated in the simulation
to take account of varied requirements.
(i) !1Ydrologica1· Constraints.
The 32-year records of t)ydro1o£iy simulates seasonal and over-the-
year flow characteristics of the river and as outlined in Section
7. 2., i ncl odes a ser·i es of very dry years <..nd correspondingly lower
energy potential in those years~
(1 i) Environmental Constraints
A variety, of environment a 1 cons tr ai nts were deve 1 oped and evaluated
during the course of the study and the following have been incorpor-.
ated in the reservoi.r·operations:
-To reduce fluctuations in downstream river flows and water levels,
no significant daily load following v-1i 11 be attempted from the
Devil Canyon power station;
-To maintain proposed fisherie.s mitigation efforts (further dis-
cussion of flows to minimize impacts on downstream fisheries is
presented in Volume 2 of this report) a minimum flow of 5~000 cfs
will be maintained at Gold Creek at all times during the reservoir
operation.
- A min·; mum flow cf 500 cfs wi 11 be mai nt ai ned at a 11 times in the
river reach between the Devil Canyon dam and the Devil Canyon
tailrace outlet. ·
-Both Watana and Devi 1 Canyon reservoirs wi 11 be operated through-
out spring and summer months to attenuate flood discharges to the
extent possible. This will m;inimize potential damage to fisheries
mitigation efforts due to high flood peak discharges.
(iii) Equipment Constraints
Generating equipment in each pCiwerhouse wi 11 be assumed to function
at not less than 50 percent of maxi mum output to avoid rough opera-
t ion.
(iv) Geotechnical Constraints
The assumed ratio of reservoir drawdown and filling have been lim-
ited to ensure that no serious r~servoir slope stability problems
will occur. In addition, maximum drawdown limits have been deter-
mined which will produce the optimum combination of firm and average
energy. The dra\-JdovJn l~1mits are discussed in more detail in .. · · ·
Appendix 82.
(b) System Demand,and Reservoir Operating Rules
Studies of reser-voir operation ~y~re based ori preliminary demand forecasts
established for the ini~ial development _selecti-on studies (Section 5) and
subsequently ,revised to take account of the system load forecast developed
in power alternativ,e studies by Battelle (Section 5 .. 7). System reliability
criteria (Section 15.2) requires a guaranteed or firm energy to be avail-
ab le from the Wat ana and· De vi 1 Canyon developments. This energy is a func-
tion of hydrology nf the river, reservoir storage and operating procedures.
The reservoir simulation mode-l uses a procedure to maximize the firm energy
potential of the developments, consistent with the various constraints
listed above.
An optimum reservoir opef•ation requirement was thus established by an iter-
ative process to minimize net system operating costs while maximizing firm
and usable energy production during the earlier years of demand growth ..
Four· alternative operating rules for the Watana reservoir (A, B, C and D)
were selected for study, to define the possible range of operation. Case A
represents an optimum power and energy scenario, while Case D reflects a
case of "no imp act on downstream fisheries" or 11 avoidance fl owsn. Gases B
and C are intermediate 1 eve ls of power operation and dovmstream impact.
These essentially define monthly minimum reservoir levels that should be
maintained to provide firm e~ergy consistent with constraints-outlined
above. For feasi bi 1 i ty report purposes, operation mode 1 "A" was adopted
for· project design and approximate fisheries mit i gat i on measures developed
(See Volume II). Details of the computer simulation runs for energy poten-
tial and their impact on project economics may be found in Appendix B"2.
Table 15.1 presents a summary of potential energy generation with different
operating rules for ~Jatana and Devi 1 Canyon developments.
The proposed reservoir operating rule. (Case A) is presented in Fi gur~ 15~1 ..
,This mode of operation represents target minimum levels to be strived for
in the operation of the reservoir"" The target \-Jatana reservoir, level to
be attained at the end of September each year is fixed at 2,190 feet. This
level of 5 feet above the normal maximum operating level is designed to
provide a higher level of winter energy production to meet the greater
winter demand. This is consistent with river hydrology in that significant
floods do not occur in the peri ad from October to early r~ay, and safety of
-structures is not sacrificed.
For th·is mode of operation, the average annu.al drawdown in the Watana res-
ervoir is estimated to be 85, and at-Oevi 1 Canyon 55 feet. However, during
the driest $equence of simu 1 ated ri verf1 ows, Watana waul d be drawndown to
its optimum-minimum level of 2~045 feet (see Section 12.11).
(c) Energy Potential of the Watana-Oevi 1 Canyon Developments
Aver age annua 1 energy potentia 1 of Watana development is 3460 Gwh and that
of Oevi 1 Canyon development is 3340 Gwh. A frequency analysis of the
annua 1 energy potential has been made to derive the firm annua 1 energy
potential or the dependable capacity of the hydro development .
.:, . •.
15-10
I
I . ·~
I
I
I
I
I
I
I
I
I
I
I
I
I
I
., •• < ••
. .
I
i
i
I
I
I
I
I
I
I
I
I
-I
I
I
I
I
I
I
I
I
The federal ~nergy Regulatory Commision. (PERC) in their publicati.on
Hydroelectr1~_,,Power Evaluation (DOE/FERC-0031 of August 1979) defines the
dependable caj):acity of hydroelectric plants as: ... the· capacity which, under
the most adverse flow, conditions of record can be relied upon to carry
~ystem load, provtde dependable reserve capacity, and meet firm power
·obligations taking into account seasonal variations and other
characteristics of the lo~d to be supplied 1•. As described in Section 7.2,
the recorded 1 owest flow in the Susi tn a river at Go 1 d Creek has a
recurrence frequency of the order of 1 in 10,000 years. This is considered
an extreme.ly rare event to be considered in an electrical system
reli abi 1ity evaluation. The critical streamflow sequence of record for the
reservoir operation simulation is the 32-month peri.od between October 1967
and May 1970, resulting in extreme drawdown of reservoirs in 1970 and 1971 ..
This sequence has a recurrence i nterva 1 of 1 in 300 years. .Based on the
Railbelt system studies and previous experience on large hydroelectric
projects~ it was assumed that a recurrence period of the order of 1:40 to
1:50 years dry hydro logi ca 1 sequence would constitute an adequate"
reliability for the electrical system.
An analysis of annual energy potential of the reservoirs showed that the
lowest annual energy generation has a recurrence frequency of 1 in 300
years (See Figure 15.4). The second lowest annual energy of 5400 Gwh has a
recurrence frequency of 1 in 70 years. This 1 atter figure has been adopted
as the firm energy from the developm~nt.
Expressed another way,_the firm energy as defined may fall short of its
v a 1 ue by about 5 percent once in 300 years. This is again a ccnservati ve
interpretation of the FERC definition.
Tpe monthly distribution of firm annual energy as simulated in the reser-
voir operation has been used in system generation p1anning studies. Average
monthly energy based on the recorded sequence hydrology is used in the
economic ana lysfs (Section 18 .1).
(d) Reservoir Filling Sequence
Given the relative sizes of the Watana and Devil Canyon reservoirs, 'it is
apparent that the most si gni fi cant impact on the downstream fl q)w regime
wi11 occur during filling of the Watana reservoir. Since this will be the
first reservoir filled, careful planning is essential.
· ( i) Watana Reservoir Impoundment
Minimum monthly flows that must be maintained in the river below the
dam during filling were established in consultation with fisheries
and other environmental study groups and agencies. Table 15~2 pre-
sents the minimum monthly flow that is cconsidered acceptable for
· river maintenance and fisheries requirements during the fi 11 ing per-
iod. With the above minimum ·flow requirement, it would take at
1 east 2-1/2 years of average stream flow to fi 11 the r~servoi r,.
,-
15-11
Other major considerations in determination of tn'e reservoir filling
-sequence is the level of the fi11 dam construction~ avaiJable flood
discharge facilities and tlie permi ssib 1 e risk of ove_rtopp~Jig the
partially constructed dam during unusua.l floods in the river. It
has been assumed that a min1murnstorage volume will be maintained
behind the part1a11y constructed dam at a11:times during the filling
period so that with available t:rischarge faciltties {low level
o~tlets and service outlet works as they become available) a 1 in
100 year flood could be safely absorbed without overtopping-of the
dam. This figure has been selected as acceptable on 'che assumption
that short-term flood forecasting will be made during the filling
period. Consequently, with car~ful monitoring of snow pack in the
basin and storm tracking, potentially damaging streamflows could be
predicted with ~ufficient warning to lower the reservoir level in
time.
It may be noted that the placement of the fi 11· dam critically
contra ls the reservoir fi 11 i ng in average streamflow years and
restricts earlier filling should wet years be experienced. The
driest recorded streamflow sequence waul d extend the filling period
by one year,
The filling sequence in the years of average streamflow would allow
first power on line by July, 1993. The units could be tested and
commissioned prior to this date. A bonus in power and energy could
be gained with one or two units installed by July 1992 when the
power intake will be submerged sufficiently to allow power
generation utilizing the minimum downstream discharge required.
(ii) Devil Canyon Reservoir
With Watana Reservoir in operation, the fi 11 i ng of the De vi 1 Canyon
Reservoir is relatively easily accomplished. Average monthly power
flows from Watana between the months October through December in a
single year wi 11 fi 11 the reservoir while maintaining the minimu.-n
downstream flow requirements (see Table 15.2).
(e) Operating Capabi 1 iti es of Susitna Units
(i) Turbine Performance
The reservoir operation studies described above, show that the
Watana plant output may vary anywhere from zero with the unit at
standstill on spinning reserve, to 1,200 MW w~en the six units are
operating under maximum output at maximum head. (Note that there is
a limitation in loading of a single unit in the rough zone of tur-
bine op.erati on from above speed-no-1 oad operation to about 50 per-
cent load). The four units at Devil Canyon have a maximum total
output of 700 MW at maximum head.
The operating conditions of the turbines determining its character-
istics are summarized in Tab 1 e 15.3.
15-12
·.
'1··. , -. -
I
I
I
I
I
I
:1
I
I
I
I
I
I
I
I ,,
I
I
I
_I
I .,..,,
I,
·,I;
I
I
I
~
I
I
I
I
I
••
I
I
I
I
I
The turbine design-head corresponds to the ;,weighted average head-.
Based on the predicted da.i ly 1oad curves through the year 2010 and
expected reservoir operation, it is expected that-'each unit at .
Watana is to supply a load averaging between l96.MW and 100 MW .. ·
This is the load which corresponds most closely to the best
efficiency operation of the turbine.
Similarly the Devil Canyon units will supply a load between 174 MW
and 100 MW.
(ii) Expected Unit Performance Characteristics
The rated output of the turbine corresponds to full gate operation
at the rated head. Each turbine should operate satisfactorily at
the maxi mum head. The output of the generator is 1 i mited by its
continuous maximum rating of 115 percent with a maximum temperature
rise of 80°C. The coriti nuous. maximum rating of the generator deter-
mines the maximum output of the unit and it will be necessary to
limit the turbine output to this value accordingly at higher heads.
The expected plant performance at varying heads between minimum and
maximum heads is shown in Figures 12.22 and 12.23. The plant
efficiency with different number of units in operation is shown in
Figures 15.2 and 15.3. In practice, for Watana, the load following
requirements of the plant results in widely varying loading and
resulting efficiency of operation.
(iii) Stability and Governing of Units
The required flywheel effect (inertia) of the unit is governed by
the stability requirements of the units, namely the stabilization of
the frequency for small load fluctuations. The machine inertia also
influen·ces the transient stability of the units during transmission
line tripout following electrical faults in the system, a larger
machine inertia decreasing the initial swing of th~ generators. On
the other hand, a larger machine inertia decreases the natural fre-
quency of oscillation of the machine, and increases the possibility
of resonance with hydraulic surges in the draft tube and penstock.
Electrical transient stability studies of the Railbelt system indi-
cate that the "natural 11 inertia of 3.2 to 3.5 kW-sec/KVA for the
Watana and Devil Canyon generators is adequate for electrical stab-
ility of the system.
The pertinent plant data for stability and governing are given in
Sections 12 and 13 in Watana and Devil Canyon plant respectively.
Pressure rise and speed rise are within normally acceptable limits
of about 40 to 50 percent. A low ratio of the starting time of the
·-wd.ter masses to the mechanical starting time of the unit is an indi-
·cat ien of the hydrau~l i c stabi 1 i ty and acceptable response (prompt i ...
tude time constant) of the governor. Good governing response and
15-.13
stability is 1nd1cated for the Watana and Devil Canyon units, and is
imp-ortant from the overall considerations of system load· fo 11owi ng
and load-fr.equency response of the units.
-{f) Watana Plant Daily Simulati.on Studies
The objectives of the plant daily simulation studies are to present perfor-
mance studies of the se1ected 6-170 MW unit plant at Watana. The studies·
demonstrate 1ts improved performance in comparison with a 4-250 MW plant.
The si mul ati on program was arranged to~
-Study the operation and load following characteristics of the Watana
powerp 1 ant with dif-ferent number and rating of units;
-Determine the effect of minimum and maximum loading constraints·of the
units;
-Determine the effect of critical single or double contingency outages of
units on the amount and type of spinning reserves available in the
system;
-Study the effects of maintenance outages and its impac_t on generation
scheduling and system security; and
--Check the operation of gas turbines and peaking plant.
( i) Computer Simulation Model
To achieve t~e stated objectives, a computer simulation program was
used to simulate Watana power plant and system operation. The
Watana turbines and reservoir are mode led in detai 1 to simulate
closely the reservoir regu1ation and load ~ following
characteristics of the turbines.
The model includes the following principal features:
Turbine characteristics as a function of head, gate opening
(flow), and efficiency are used in the mode 1.
-Minimum loading limitations of the turbine due to rough zone of
operation up to 50 percent of the gate -openings are constraints
for turbine loading and operatio!'l~
-Maximum continuous rating (CMR) of the generators constitutes the
maximum loading of the units. Higher turbine capability at higher
heads is blocked at the generator CMR rating ..
-Predict~d daily system load demand curves are used for two typical
load shapes for winter and summer, respectively. Monthly peak
load variation of the load is taken into account.
-Reservoir characteristics as a function of level and storage.
Average maximum and minimum reservoir levels are constraints for
res.ervoir regulation and operation.
15-'14
,
I
I
1 ..
I.
I ~
I
I
.I
I
l
•••
I
I.
I
I
I
I
I
.I :.~
I
I -
I
I
I
I
I
I
I ,,,_·
I
I
I
I
I
I ....
I
I
I
• :r
-Unit by unit 1oQding B:nd de-load1ng of Watana generators according
to load demand (load~following} is done taking-into iccount all
constraints mentioned above. The program 1 oads 'the units equally
for maximum efficiency of operation •
-.Loading steam plants as·base-load plants, and gas turbines as
peaking plants. .
-Maintenance scheduling of the generating units.
Results of the Simulations
Printouts of the results of the simulations are included in Appendix.
. For each run, printouts are presented for the fo 11owi ng out---p~ut~s in a typical day in each month of the year 2000 (January to
December·):
-Watana plant kW out~ut;
-Watana turbine kW output, with flow and efficiency for each unit;
-Watana turbine utilization, showing number of units loaded;
-Watana reservoir level;
~ Peaking plant kW output;
-Total system load kW demand;
-Total system reserve, including maintenance outage;
-Watana reserve capacity; and
Annual energy output of Watana, thermal plant, small hydro, gas
turbine plants, and overall annual system energy.
Simulation re.sult of a typical December, 2000 day is shown in Fig
15.6. The simulations indicate that the six unit Watana plant
(6-170 MW) has superior overall performance in terms of load follow-
ing, improved overall efficiency and minim~m loading constraints of
the un1ts over the four unit plant (4-250 MW).
The overall reliability of the six unit Watana plant is better than
the four unit plant. During maintenance the six unit plant has a
planned outage of 170 MW, as opposed to 250 MW for the four unit
plant. During peak December loading, a double contingency outage of
two units brings down system reserve to 107 MW for the 6-170 MW unit
p 1 ant and to 1 ess than zero for the 4-250 MW unit p 1 ant for the year
2000 on study.
The. simulations indicate that sufficient spinning reserve of a mlnl-
mum of one Watana unit is available for all pe)ak d·ay loadings for
the six unit Watana plant for the year 2000 on study .
15.7 -Performance Monitoring
(a) Watana Dam
Instrument<:Ltion is installed to enable the performance of the dam to. be.
monitored to ensure that its behavior is within the limits assumed in the
design and to enable any variations beyond those limits to be recognized
quickly so that remedi·al action can be taken without delay.
15-15
lt is essential~hatcontltu.tous monitoring of the 1nstrunentation installed·'
in the dam be carried out' by qualified pe~sonne1 who thoroughly un~erstand
the significance of the readings and more importantly, the signific·an~e of
variations in the readings.
The 1 nstrumentati on is installed to monitor both short-term behavior during
construction and initial filling and long-term-behavior ove-r the life of
the dam .. The short-term is the most significant period when the dam is
subjected to its initial loading and the responsibility for read1ng the
i nst)"uments, reducing the data and eva 1 uat 1 ng the results over that period
normally rests with the design engineer~ It is important that personnel
who wi 11 ultimately be responsible for the monitoring are involved as early
as possible i~ the development~
The most important aspects of the moni taring program and likely maintenance
requirements are outlined below:
(i) Foundation Abutment Pore Pressures and Discharge
Frcm _Drainage Pressure Relief System
Since sections of the foundation are frozen, the grouted cut-off may
not be fully effective and leakage may increase a~ the rock tempera-·
ture increases.
This condi t; on would be ind 1 c a ted by increased discharge from the
drainage system and would be remedied by additional grouting from
_the grouting gallery, possibly combined with additional drainage
holes.
(ii) Quality of Discharge from Pressure Relief System
Any discoloration nf the drainage system discharge would indicate
the carry over of fine material either from the rock foundation or
from the core. The problem area would be located and additional
grouting carried out~ Water quality should also be monitored for
any change in materia 1 content.
(iii) Deformation of the Structure
" Most deformation of the structure as observed by settlements and
1 at era 1 movements is expected to occur soon after construction and
under i nit i a 1 fi 111 ng of the reservoir. Any excessive sett 1 ement
would be made good to maintain freeboard .. Deformation records would
be correlated with such data as reservoir level, heavy storms and-
seismic activity.
(iv) Routine Observations
An essential part of any monitoring program is a regular routine
visual inspection of all exposed parts of the structure and the area
downstream of the dam for any unusual features such as 1 ocal
settlement or other movement, zones of seepage discharge, wet areas,
and changes in vegetation. All exposed concrete surfaces would also
be inspected and records kept of any signs of distress, cracking or
deterioration. ·
15-16
I
I ,,
I
!}
I
I
I
:1··· ..
' .
I
a
I
I
I
I
I
I
I
I .,..
I
'I
'
' '"'\
' I
I
,I
' I
I
I
I
I
I
I
I
I
I
( v) Relict Channel~
·Particular attention_ must be ·paid to monitoring the whole area of
tt·1 r·elitt channel, .including· regular readings of piezometers and
thermistors or S!lrface elevation survey monitoring and inspect1ons
of. the discharge zone for changes in seep,age flo\'Js and any signs of
piping fatlure. ·
15.8 -Plant Operation and Maintenance
The system demand varies throughout the year from a winter (December/January)
peak to a summer (July/August) trough~ and from hour-to-hourthroughout the
dayo The central dispatch center operates with the object of ensuring that
sufficient plant is available at all times to meet the varying load in
accordance with a meri.t-order schedule with due considerations to security.
On the other hand, generating plant must be maintained periodically for various
reasons:
-Preventive maintenance, to ensure safe and reliable operation (performed
either on load or shut down);
-Corrective maintenance, to restore lost efficiency of plant; and
-Emergency maintenance, arising from plant failure.
Stationary inspection~ of plant are also ~equired.
To meet the conflicting requirements of supplying load and maintenance outages,
the plant maintenance program is planned to fit into the electrical system gen-
eration program to determine the amount of plant which can be safely permitted
to shut down for maintenance during each week of the year. Due consideration
must be taken of such factors as:
-Repair capacity in maintenance workshops;
-Delivery of spares and materials;
-Availability of specialized labor, special equipment and similar resources;
-Expiration of statutory inspection periods;·
-Weather conditions, for outdoor installation; and
-Concurrence of outages between generating plant and transmission system.
The plant at Watana and Devil Canyon are high merit-order plants and maintenance
must be organized to minimize outages which affect availability. The expense
and time of inspecting and maintaining the .large and important machines at
Watana and Devi 1 Canyon in good condition, rather than operating the machines
continuously until failure occurs in service can be justified on the basis of
increased reliability and lower overall cost. With this in mind, sufficient
spare capacity.must be availaqle in the Susitna plants and the system .at all
times to cover planned and forced outage of the large units~
(a) Frequency of Inspections and Maintenance
Th.e degree of inspection at the p 1 ants varies from frequent periodic vi sua 1
inspections to a complete major disassembly and thorough inspection at long
intervals of about 10 to 15 years. Certain manufacturers and users recom-
mend a major inspection after the first year of service. Factors influenc-
ing the degre~ of disassembly and period between inspections include the
following:
15-17
.·: '
0
(b)
>' ' ""''-"-.,......---;;--" ~ ..
-Findings at previous visua1 inspections;
-Result$ .. ,of previous tests;
-Hi~story of similar mach-ines; and
.. Frequency or starts, load cycli.hg and overloading during service.
Ready avai 1 ability and access of stored data on the plant computer system
of the records of previous inspection and tests as well as current perfor-
mance trends (such as, for example, abnormal temperature recordings) wi 11
improve over a 11 ma1 ntenance and performance of the units.
Experience records from machines similar to the Watana and Devi 1 Canyon
machines indicate that a minimum maintenance period of 5 to 6 days are
required for each machine, resulting in an outage of 150 to 170 MW capa-
city for an average period of 50 to 60 days in the years. In exceptional
cases, certain machines may be do'lm for greater maintenance periods. It is
therefore re.asonab1e to allow a total of 2-1/2 to 3 months planned outage
as a conservative approach to system generation and maintenance planning
for the Susitna units~ In principle, these outages are scheduled during
the months of June to August when the lower summer load demands make it
possible to release the units for maintenance. The actual outages will be
coordinated on a week-to-week basis with the planned maintenance of the
units in the rest of the system and will take into consideration emergency
shutdowns, breakdowns, delays in construction and maintenance and other
unforeseen contingencies.
Access and Maintenance in the Powerhouse
Techniques developed both in the design and the operation of convention?-1 -
underground hydroelectric powerplants have resulted in underground facili-
ties which are not significantly more difficult to maintain than surface
- p 1 ants. I so 1 ati on of underground i nst a 11 at ions from both penstock water
and from tailrace water is a vita 11y important factor. Downstream water
conduits with manifolds require draft tube isolating devices of appropriate
design. Drainage and dewatering facilities must be highly reliable and of
adequate capacity.
There will be situations where a decision must be made as to whether to
carry out maintenance and repair work on components underground or on the
surface. Many i terns are large and heavy and therefore are best hand led by
the powerhouse crane. Sufficient erection bay space and 1 aydown area
between the generating units are provided for all normai maintenance and
overh au 1 needs .
Transformers wi 11 be moved within the access tunnel and transformer ga 1-
lery by means of whee 1 s mounted on the transformer base~
The greatest demand in 1 aydown space within the powerhouse cavern is 11 kely
to occur during the initial equipment installation process and the 10 to 15
year major disassembly/maintenance. The working area will be siz.ed to
allow the simultaneous placing of turbine and generator components.
~
1.5-18
,,
.. 1_.·
. .
f.
I
I
"I
I ...
I
I
1.;
.
I
I
I .
-:>
.... '
I
1-
, ..
a~
I --.~
I
{\
.' •. \~!::
.· j
I
I
I
I
I
I
I
I
I
I
·I
I
I
I
I
I ,,
·I
(c)
'-
Adequate crane facilities are ~,provided both for installation. and mainten ...
ance. _ The main powerhnus'e: overhead cranes have a_ capacity of about 200
tons and will be equipped with a zo ... ton auxiliary hoist. Small monorail
hoists wi11 be provided as necessary at intermediate levels w1thin
powerplant caverns for handling equtpment likely to require movement for. ,
·l.nspecti on, maintenance and/or replacement._ .
Major Overall Activities
The major activities which require special space and handling considera-
tion 1n the plants include:
Replacing generator stator
winding coils
-Rotor inspection
-Replacement of thru.st-beari ng
assemblies
-Replacement of runner seals
Cavitation damage repair to
runner I i mp e 11 e r ..
-Repair and refi nlshi ng of
water passage stee 1 and
concrete surfaces
'15-19
--performed in situ, may require
removal of rotor assembly~
--performed for losseness, over-
heating or short circuits,
specially after a major trip
out causing full overspeed.
--designs normally permit
removal and replacem~nt of
components without major
di smant 1 i ng, but at major
overhaul intervals removal
and strip down will be
ad vi sab 1 e.
--requires dismantling and
removal of runner component.
This may be possible from
below by removing the draft
tube cone and bottom cover, or
alternatively from above by
removing the he.ad cover and
runner .
. --will normally involve access
through draft tube for in-
spection and minor repair~ but
ultimately requires runner re-
moval as outlined above.
-~access to the penstock will
occur at about 5 to 6 year
intervals. Major painting or
refinishing will probably not
be required until 10 to 12
years after commi s 1 oni ng. More
frequent access will be provid~
ed to the downstream water pas~
sages when isolated by the
draft tube gates and unwatered
by the station dewatering
system.
,~ . ~ ~
.'I ,. "'""~'~ ':~-~~
---1 ,di
(d)
. .
--Generator . circuit b_f'ea:ke.r ·
repair
... Transformer matotenance
---particular -attenti·on-and a
high l~ve1 of majntenance ·are
-requir.ed for. generator circuit
bl"eake.rs.
-~general maintenance will be
carried out in situ in the
transformer gallery .. , Major
overhaul or repair requiring
untaki ng of transformer wind-"
ings will be done in the
powerhouse erection bay with_
adequate crane f aci 1 1 ties.
Maintenance Workshops and Operating/Maintenance Staff
'
The. Watana and Oevi 1 Canyon powerplant are each provided with workshops to
facilitate the normal maintenance needs of each plant. The workshop block
includes operations for fitting and machining, welding, el•ectri.cal, and
relay instrumentation, with adequate stores for tools and spare parts.
TheWatanapowerplant will be provided additionallywith surface
maintenance and central storage facilities to cater to the needs of both
p 1 ants.
Maintenance operation p l anni ngs of both plants are centro 1 ized at Watana.
Staff wi 11 be normally located at Watana and housed at the operators vi 1-
1 age at Watana .. With centra 1 i ~ed contra 1 of the Susi tna project 1 ocated at
Watana, the Devi 1 Canyon p 1 ant wi 11 not have a resident operating . and main-
tenance staff. Prop.er road and transport facilities should be maintained
between Watana aqd Devil tanyon to facilitate movement of personnel and/or
equipment between the plants.
The central maintenance staff should include the following recommended
minimum personnel:
1 -superintendent of maintenance
1 -electrical maintenance engineer
1 -mechanical (and building) maintenance engineer
1 -instrumentation maintenance engineer
6 -assistant maintenance engineers; at peak maintenance will
shift basis
work on a
Both the Wat ana and Dev 11 Canyon power plants are desi· gned to be norma 11 y
operated from the Sus 1 tn a Area Contra l Center at Watana. The oper:ating
staff wi 11 be stationed at Watana and would consist of the following
personnel:
15-20
'=~=-.... ~~) .... ..::.,~~··-=_.· .. ·,"'-~ .·"'"-"'-. :"""""'. -''---'~--'----'-~~~:_____J/t'--~·-'=---'-~
. . .. : ::~
1.1 . ,~1
.i:l ,, ..• ~
t
I
I·
I
t
'"" I
I
I
I
I
I
I·
I
t
···:' . .
:. . ·~ ( . '
. i
·., •. · ..
.......
.·'·1,: •. <' .
;;:1.··
t.r' . :
I
I
I
I
I
I
1
..
. :
. '
.. , ...
I
I
I
I
I
,I
" 1 · "'l sup_er-l ntend~nt, Sasi· tna project operations . ·
·1."" chief :operator
3 ... control ,room operators (on shift basis)
2 ... powerhouse oper.ators .
2: ~·assistant powerhouse operators
2 -computer s.ystem operators
When necessary, operators will trave 1 to Devil Canyon to assist in
oper.ation :and/or maintenance programs at the powerplant.
:•
.15-21
· ..
~ " . ~.,.,~ . ···~.:::: ... .._' ··~~
7
MONTH·
OCT
NOV
DEC
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
TOTAL
NOTE:
' .~
t
.TABLE .15~ 1: ENERGY POTENTIAl Or WATANA .:. DEVIL CANYON -DEV£LOPMEN1'S
:-:. fOR OtF'FE.Rl!Nf RESERVOIR; OPERATING RULES .
..
ENERG.'f POIEN11A.L G W -If
c
0
w _~~. r & ~N. p. 0 N L J ·-wArJ\NJl ~-E V i L C A N Y _Q_Jf
I" lf1 .M i:.NtKEY ·. AVt; .KAl:il::.' i:.l\lt.KI.iY t KM. t.Nt.r ,\.11 .AVt.H ,Aut .t.Nt; Klil
. t;ASt. A c u A c I D A c D A· r.; D
234 200 172 281 2141 178 437 399 J34 511 422 34Ci
270 235 201 348 331 271 502 463 388 543 625 506
-
"'" 322 276 236 445 397 364 598 547 4.58 B17 751 68):
283 242 208 383 '}57 3.25 590 480 403 715 677 618
228 202 173 318 3.35 293 .452 395 330 599 632 561
.235 201 173 276 330 277 470 398 335 532' 629 536
199 165 142 203 214 197 460 332 280 451 419 387-
180 152. 131 _-;: 180 247 174 462 304 286 465 536 399
170 135 111 175 212 191 492 ~23 278 478 485 460
182 209 345 258 267 374 j$7 471 755 521 579 784
170 311 531 344 327 545 321 659 1095 598 679 1095
158 151 155 249 158 166 29) . :326 .:390 463 346 J95
26:32 2479 2578 )459 3389 3354 5394 5099 5332 6793 6781 6168
Cases B and C were similar and only Case C was analy?:ed in detail ..
0
;
_·;·: .• :;,.:
.·;.. '~
' ~-i,. :
'l ·I',
I
I
I
I
I
I
I
I
I
I
I:
··,: a:
II
I
I
I
c~< •• •
••••••• . .
.·1
I
' I
I
I
••
.·~··
I
••
I· ....
I
..• 1.
I
I
-a·· ..
I
TABLE 15,.2; . MINIMUM ACCEPTASLE F'LOWS i3E:LOW __
. ' WATANA. OAM DURING RESERVOIR FILLJNG
MONTH
OCT
NOV
OE:C
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SE:P
MlNIMUM.ACCEPTA8LE.FLOW CfS
2050
900
900
900
900
900
900
4000
4000
6000
6000
4600
•
~ .·-·
rABLE:·15.3: . TURBtNt oPERA.llNG coNOITlONS
Maximum net head
MioimuiTI net head
Design head
Rated head
Turbine flow.at rated head
Turbineefficiency at design head
Turbine-generating rating at rated head
Watana
72S feet
580 feet
680 .feet
680 (eet
3550 feet
91~
181;500 kW
.;
Devil Canyon
59.7 feet
238 feet
575 feet
575 feet
3800 feet
91%
164,000 I<W
·-
;:; . ' .,~
"-~·
I
I
I
I
I
I
I·
I
I
I
I ..:;
I
I
I •
I
I
1
. 'l
.. c-c;..>l
I
~:.
~ .• ;
. I
I
I
I
I
I
-·
I
I
'
••••• •
'I
•••
....... ~... l '
i-. · .. ~.·. ·.•· l>;~ :
I\~ ..f..'-~;.,·
100
90
0 ao
c
0 70 ..J
~. 60 c. ·w
a., 50
-40 .... z.
1&1 30 0 .
a:
~ 20
10
0
. · /
I ~
.. I
~· v -0
.•
-
0 4
<
a 12
HOURS
16
WINTER WEEKDAY
HOURLY LOAD VARIATION
::'.
lOO
' ~
~
'
90
c 80 c
0 ....J 70
.~
··"« 60 w
1L
I&. 50
0
.... 40
z
1&1 30 ()
a:
Ill 20 D.
10
0
20 24
. :.,:;· ·.~· ---...
..
/ ...._,_
'""' ·.
I(
' '\.
~
X
0
-.
I -
I
""" ~
4
..
'
•·
8 12 16
HOURS
SUMMER WEEKDAY
HOURLY LOAO VARJATlON
0
..
20
NOTE: PEAK MW JULY 200Q AD= 658 MW .
TYPICAL LOAD VARIATION
lN ALASKA RAlLBELT SYSTEM
24
..
uoo
1000
900
z 800
2
.z 700 -
0 600 c(
0
-J 500
~
ct
kl. 400 n..
300
200
100
0
( .... : ...
;
' t . ·" ' ' ...... ~ ..
'
I
r
I· I;
:
f
I
J F M
I
........... _..., -
l I
A ·M J ,,-A
· MONTH
LOAD VARIATION
IN YEAR 2000
/
.. v
/ v
..
I )
s 0 D
FIGURE 15 .. ( .......... : ·• .•. ··.····.~~-. ' ' ' F • , , '
i - : . ~ ' : :
' : -~ . .
----...
,·I
_ .•
I
I
I ,,
I
I
I
I
I
I
I
I
I· ,,
I
I
I
86~+---*---~--------~--------~-------+--------;-~
~ 0 ->-0 z w
0 82~+-------~--------~--------~-------+--------~~ fu
U) .... -z ,:::»
78r-+-------~--------~--------r--------+--------~~
~~+-------~~-------r--------+--------+--------~~
100 500 700
PLANT CAPACiTY ( MW}
WATANA-UNIT EFFICIENCY
(AT RATED HEAD)
900 liOO
FIGURE 15.21 Alii l··.
I~··
I c
I 94
--.
I 90
I'
S6
I -~ 0 -
I
I
[J
>-0
I z
I.IJ
2 82 u..
Lt..
I.IJ ,, . ....
z
:::;)
I 7S
I
I 74
I
I
'I
I
I
I
-,:1
~..~.,
I . /\NIT r-2 UNI'fS r-3 UNiTS
/' -\/ ·."V V"
!f \ v v
l/
I
.
,:)
100 200 300 400 500
PLANT CAPACITY ( MW)
DEVIL CANYON-UNIT EFFICIENCY
(AT RATED HEAD)
. I
[4 UNITS
' ~"-
~,
·-
600
flml
FIGURE 15.31 AIIR I
"' ill[. 0
-1& HRS. 1;0625£+01
",... l.Ol.67Ef01 <(
Q 18 t.o;oSE:f1fl -l
<( ~" . , 1;07SOE+01
(.) kJ t-l.0792tt01 -m 20 . a.. "~ 1oOB3jEt01 >•
1-w > 1.0875£101 0 22 z -'""-t ... --. ,. l.<>917Ef01
~ . 1.095SEt01 24
OHRS 1~1CCOH01
1G1C42E1Q1 2
1 .. 10S3Et01
4 1.1125Et01
1.1167Ef01
6 t.120BEt01
la1250Ef01
8 1.1292£+01
1.1333Et01
)-1.1375Ef01 <( 10 c l.1417Ef01 ...t
<( n: 1.145BEt01 0 l&J NOON -m 1.1500H01 0... ~ >-w 1.1542Ef01 1-0 14 -
l&J 1.1SB3Et01
0 1.16Z5H01 )6 1.1667H{)1
1.1703Et01 18 1.1750[~ Crj
20 1.1792EJ01
1 lOJln(li .1. I. U . ..SJ..I ~
22 1.1875rf01
• 1)<1](J111 .lt .1.. • IV
24 HRS. 1.1~53El01
-iii r.dttViif ' ·
l.OB21Ei05
UATANA
6.4956EtQS
6.84o8Ef05
"7:1935EtOS
7,0NO£f05"
6.8608Et'05
6.o276H05
6.~935Et05
5.7272E+05
5.0609E105
• •
-
----...o:~_-........... ____ .... __ _... ____ __;.._ ___ ~ ____ ..;.. __ ,...._ ___ f
_, ______ ~*'!"!1"--~-~---..ii..;~~-------------"--~----~------f:
---~-------------~---------------------t
-------~------~----------------~------+
-------------------------------:---t
------... ---------------------t
. "
' H 'GEt1Kt1 EFF
· iThooonoo 1~6239Etos 9.15sor-.ol
.t.OOOOEtOO .1 .• 7117EfOS 9.07uSE-01
s.ooooEtob n4JB7Etos 9it747E-01
4.0000EtOO. 1.7735£+05 fh9866E-01
4;000()tf00 t~7152EtOS 9e0b61E-Ol
4,0000Et00 · 1.6569Et05 9.H55E-01
4.0000Et00 {.5994Ef05 9.1651E-Ol
4.0000EtOO 1\431BEt05 9,16SBE-01
3 I OOOOE tOO 1.6870£ +OS'. 9.1049£-01
l. . Kir;ia«UIIl l "RESER\T, -4, 47;1EtOS
j 6.'01-44Ef05 ... ..,__.__._-...f
5.6o32EtQS ___ ... .,._ ... -+
.
. , .. • .. 5~.316·5·· ·.E. t. 0 .. 5 __ ..;_ .... .,. s.416oaos ------~t .
x 5 • 64 92Ef05 .__ ........... -•. +
REStRV verses TIHt .·.
. .
'
Ha,d~oor
1 ~.t42BE406
0
INSTAL KWLOAII WAIRES
. ·r.S310Et06 ·9.2956Ef05 ~.2646Et0S
1t53l0Ef06 · 9l6468Et05 :319129Et&S
t.S310Ef0b 9.9935Ef05 3.5656EfOS
1.53l0Et06 · 9.S940Et05 J~664SEt05
1.5310£+06 9.660BEfOS 3.S971Et&S
l r S~BB24EfO:i -----·--+ 1.~310Et06 9.4276Et05 ~~129SEf05
! 6.1165£+05 --------t 1.5310Et06. 9~1935Et05 4.3634Et05
f6.}B2SEt05 ---------------+ t.5310Et06 S.5272EtOS 5.0292Et05
j1,4491E+05 . ----------------. ..... + 1.5310Et06 7•B609H05 5.6951Etos·
4.S959Ef05 ----------+-----------3.0000E+OO 1.6320£+05 9.1560E7'"01 ! 7.6141Ef05--------+----------~.1.S310Et06 7.6959EtOS S,S598Et05
4.6429£+05
4.3900£+05
4.13B1Ef05
4.3549£+05
4.5717.£+05
~ .• 7S97Et05
S.SB47Et05
6.3796Et05
7.1728Et05
7.172SH05
7.172BEt05
7 .1729Ef05
7.2090[105
7.2451£-1·05
7.2S21Ef05
-. 6' .. 0 ... +/\1:' I• t.~ t. v.J a /\"lo9::-~or-t V;JU L I .J
7 °7 10noc-I I..S L I .J
7,67B1Et05
7.4252[{05
7 .1711[·105
t..44C5[~05
5.7250Et05
.
-~----------M~----------t ___________ ,_;,. __________ +
---------------------~ J "
-----------------------+
-------------------------1
-----------~--------------+
--------------------------------+
-----------------· ---------------------+
------------~-----------------~------------+ ----------=------·-------------------------+
--------... --------------.-.-------------------+
-~----------~-----------------------------+
-~---------------~~-------------------------~ J
---------------------·--------~-----------~-~
" ' -______ _..,.,._...,_.,.,~_,. __________________ :--------------+
______ ...., ___ .., ________________ ""'" ____________________ J.
. ~
---·----~----------..... ---______ ,...,. _____ ---oii---~..,--'"'"'!-----.l
' -----... -~--------------------~---'--------------------~ " I
--------------------------------------.. ----~---+ ---~----"""1~---------------.... ----~--ooo! .. ,.-...,. _____ ,..._.._ ___ +
------------------------------------------+
------------------~------------.aot---------t
' --------------------------------+
3.0000E+u0 1•5476£+05 9.J7B9E-01
3.0000HOO · 1.4633Et05 9.1944£-01
3.0000£+00 1a379~£+05 9e125BE-01
3.0000£+00 1.4516£+05 9.1847£-01
3.0000E+OO 1.5239£+05 9.!B54E-01
3.0000Ef00 1.5966Et05 9.1657£-01
4.0000HOO 1.3962Et05 911393£-01
4.0000Et00 h5949H05 S\1662E-01
S.OOOOEtOO h4346Et05 9.1706£-01
5.0000Et00
S.OOOOE+OO
5.0000Et00
S,O@OE+OO
5.0000£t00
S.OOOOEtOO
S.OvGOEtOO
5,0000EtOO
S.uOOOEfOO
S.OOOOHOO
S,OOOOE+OO
S,OOOOEtOO
4.0000Ef00
4,0000E+OO
la.4346Ef05
1.4346£+05
1.4346£+05
1.4'41BEt05
1 • .1\490Et05
1.4564Et05
1 r ... ,6,..f0r ...... ~ .. t.. J
1 d.07BE+OS
l.5862E+05
1.5356Et05
1.4S50Et05
1.4342Et05
L6121Et05
1.4315Et05
-~· ---;......,. .. ""*"" ~-•.-· !=.
9.1705£-01
9.1704E-Ql
S\1704£-01
9.1'}62~-01
9.1S20E-01
9.18BOE--01~
9 .1B33E-01
9.1629E-01
9 •. 16BBE--Ol
9.1826E-01
9.1963E-01
9 1'9_,.. .. ~ •.. 0 --~t:-~.l
9.1619£ -(}1
9.1671£-01
>-(.) z
La.J
o-
i:i:t-
LLZ
La.JW
0 t-Ct:
-W zo..
::::::> _.
i 7.8611Et05 --------.:.-.-· ------.------+ . . h5310Et0o 7.~429Et05 6.1123EtC-5
. S~1200Et05 _____ _;_ ______________ .-.-+ 1.5310Ef06 7.1900H05-6.3649EtOS
· B·37l9Et05 -------------------------t t.5310Et06 6. 93B1EtOS 6.6165EtOS
8.1551Et05 · -------------------------+ 1.5310Et06 ].1549£+05 &..3994Et05
; 7 • 93S3Et05 -~----------~----------+ 1.5310£+06 7 .3717Ef0S &.1822E1-05
f7.7203Ef05 ---... --------------------+ 1.3310Et06 7.5B97EtOS S.'9638Ef05
·.L~ .• ·9
1
2
3
5
0
3E
4
· E+t.
0
05
5
.. -_-_--_-_-_ .... -_-. ---+-----+ 1.5310Et06 a.3B47HOS S.1684EtOS
, ·o 1.5310E+Ob 9.1796EtQS -~.3730£+0.5
· S.3372Et05 -----t 1,5310EHl6 9,972BEtOS ~.S792805i
5.3372£+05
5,3372Ef05
· 5.3371E+05
~ 5.3010Et05 .
jS.2649£t05
: 5.2279Et05
4,8470HOS
. 4t4711E+05
: ~I S790Et05
4.8319£+05
' 5.0B48EtOS
• -----r
---+
------+
---+
-----+
-----+
--t
t
i
--t
----t
5.3389Et05 ------t
6.0615Et05 ---------+
6.7B42Et05 --------------·-+
0
1.5310Et06 9.972SE+OS 3.S7B6Et05i
1.5310EfM 9.972BEfOS ~.S78QEtOS
1.5310Et06 9t9729HOS J.5773E+OS
1.5310Et06 1.0009Et0.0.." 3.:S405Et05
·t,5310Et06 1.0045£+0& S,.SQ3BEtOS
h53H)Et06 1.00S2Et0S 3.~662Et05
1.5310Ef06 h0463Et06 3.vS47Et05
1 t5310£t06 l.OS39Ef06. 2,~70S1E+OS
1.:5310£+06 l.0731Ef0b 2,.S153Et05
1. 531 0Et06 1 t 04 78£+06 3. 0675EfQ5
1.5310£+06 t.0225Ef06 ~ .. 3l98Et0$
1.5310H06 9 .• 9711E+OS 3.5732HOS
h5310Et06 9.2485EtOS ~~2953£+05
! .S310Ef06 8.525BEtOS .S.:01.74Et05:
.. '·~, ··-·--,~ ,,.,.,. .. _ .... _ ~ .... ,~..; "'
...J
~-r o-
1-·3=.
~
~-w
f-0
(f)·<t >-a (f) ..:.I
WATANA PLANT SIMULATION
. DECEMBER 2000
t
I
I , FIGURE 15.4 . liiJ
••
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-I
I
I
16.-ESTIMATES OF.COST
Thi-s section pre.sents estimates of capital and operating costs for the Susitna
Hydroe 1ectri c Proj.Q ··\_,.. comprising the Wa.tana and Devi 1 canyon developments and
associated transmilS\..,,~t and access faci liti·es. The costs of design features and
facilities--incorporated into the project to mitigate environmental impa€ts dur-
ing construction and operation ar·e identified. A cash flow schedule, outlining
capital requirements during planning, construction, and start-up is presented.
The section also includes estimates of the cost of capital funds required during
construction, based on alternative financing scenarios. The approach to the
derivation of the capital and operating costs estimates is described.
The total cost of the Watana .and Devil Canyon projects is summarized in Table
16.1. A more detailed breakdown of cost for each development is presented in
Tab 1 e s 16 . 2 and 16 . 3.
16.1 -Construction Costs
This section describes the process used for derivation of construction costs and
discusses the Code of Accounts established, the basis for the estimates and the
various assumptions mad.e in arriving at the estimates .. For general consistency
with planning studies, all costs developed for the project are in January, 1982
do 11 ars.
(a) Code of Accounts
Estimates of construction costs were deve 1 oped using the FERC format as
outlined in the Federal Code of Regulations, Title 18 (1).
The estimates have been subdivided into the following main cost groupings:
Group
Production Plant
Transmission Plant
General Plant
Indirect Costs
16·--1
Description
Costs for structures, equipment~
and facilities necessary to produce
power.
Costs for structures, equipment,
and f aci 1 iti es necessary to trans-
mit power from the sites to
1 oad centers.
Costs for equipment and facilities
required for the operation and
maintenance of the production and
transmission plant.
Costs that are common to a number
of construction activities. For
this e.stimate only camps• and
electric power costs have been
included in this group~ Other
indirect costs have been included
i~ the costs under production,
transmission, and genera~ plant
costs. ·
(b)
(c)
Ov.erhead Construction Costs Costs for engineering and. adminis-
tration.
Further subd·ivision within these groupings was made on the basis of the
various types of work 1flvalved, as typically shown tn the following
ex amp 1 e:
-Group:
-Account 332:
-Main Structure 332.3:
-.Element 332.31:
. -Work Item 332.311:
-Type of ~lark:
Production Plant
Reservoir, Dam, and vlaterways
Main Dam
Main Dam Structure
Excavation
Rock
The detailed schedule of account i,tems is presented in Appendix F.
Approach to Cost Estimating
The estimating process used generally included the following steps:
-Collection and assembly of detailed cost data for labor, material, and
equipment. as well as information on productivity, climatic conditions,
and other related items;
-Review of engineering drawings and technical information with regard to
construction methodology and feasibility;
-Production of detailed quantity takeoffs fr·om drawings in accordance
with the previously developed Code of Accounts and item listing;
-Determination of direct unit costs for each major type of work by devel-
·opment of labor, material, and equipment requirements; development of
other costs by use of estimating guides, quotations from vendors, and
other information as appropriate;
-Development of construction indirect costs by review of labor, mater1 a 1
equipment, supporting facilities, and overheads; and
-Development ·of construction camp size and support requirements from the
1 abor demand generated by the construction direct and indirect costs.
The above steps are discussed in detail in the following:
Cost Data 0
Cost information was obtained from standard estimating sources, from
sources in A 1 ask a, from q·uotes by major equipment supp 1i ers and vendors,
and from representat) ve recent hydroelectric projects. Labor and equipment
costs for 1982 were developed from a number of sources (1,2,3) and from an
analysis of costs for recent projects performed in the Alaska envi~onment.
It has been assumed that contractors wi 11 work an average of two 9-hour
shifts per day, 6 days per week, with an expected range as follows:
16 ... 2
-··
I ·..-· ...
••
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
·I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
(d)
Mechan1 cal/Electrical ·work
Formwork/Concrete Work
Excavation/Fi 11 Work
B~hoUr shifts
9 t..: .·· ·· h ~ rt . _ .. ,,our s ,." ·s
10-hour shifts
These assumptions provide-for high utilizatio.!l of construction equipment
and reasonable levels of overtime earnings to attract workers. The two-
shift basis generally achieves the most economical balance between lat )r
and camp costs.
Construction equipment costs were obtained from vendors on an FOBAnchorage
basis with an appropriate allowance included-for transportation to site. A
representative list of construction equipment required for the project was
assembled as a basis for the estimate. It has been assumed that most
equipment would be fully depreciated over the 1 i fe of the project. For
some activities such as construction of the Watana ma'in dam, an allowance
·tor major overhaul was included rather than fleet replacement. Equipment
operating costs were estimated from industry source data, with appropriate
modifications for the remote nature and extreme climatic environment·of the
site. Fuel and oil prices have also been included based upon FOB site
prices.
Information for permanent mechanical and electrical equipment was obtained
from vendors and manufacturers' who provided guideline costs o"n major power
plant equipment.
The costs of materi~ls required for site construction were estimated on the
basis of suppliers' quotations, adjusted for Alaskan conditionsc;;.
Seasonal Influences on Productivity
A review of climatic conditions, together with an analysis of experience in
Alaska and in Northern Canada on large construction projects was under-
taken to determine the average duration for various key activities. The
re.su 1 ts of this eva 1 uati on are presented in Tab 1 e 16.4 and these durati orfs
have been used to develop the construction cost estimates.
In general, it has been conservatively assumed for current study purposes
that a 10-month construction season is the maximum feasible duration and
that most work operations will cease during December and January because of
the extreme cold weather and the associated lower productivity. Productiv ..
ity is assumed to decrease by 30 percent during the November, February, and
March time periods. This productivity decrease results from a combination
of low temperatures, reduced daylight hours, precipitation, and soil condi-
tions.
Studies by others (4} have indicated a 60 percent or greater decrease in.
efficiency in earthwork operations under such adverse conditions. Typical
efficiency curves for the Fairbanks locat'ion, which were used as a guide,
are shown in Figure 16.1. Although this curve cannot be used directly for
the Susitna Hydroelectric Project, it illustrates the relative seasonal
effect on manual labor, hauling, and earth .excavation that is likely to be
experienced.
16-3
.... . . '
• ·: ..... J
. ,' . .
Studies performed as part of this work program indicate that,the general
construction productivity at the Susitna damsite during the·months of April
through September• would b~ comparable with that in the northern sections of
the Western United States.
Rainfall in the general region of the site .. , s moderate b.e.tween mid ... Apri1
and mid-October ranging from a low of·o.75 inches precipitation in April to
a high of 5.33 inches in August. This moderate amount of rainfall should
not create significant problems during fill placement activities because of
the good quality river run borrow material used in the haul road and dam
embankment. Temperatures in this period range from 33° to 66° for a
twenty-year average. In the five-month period from November through March, ·
the temperature r•anges from 9.4°F to 20.3°F with sno~tJfall of 10 inches per
month. Exc av at ion of grave 1 from· the river or be 1 ow ground \'later leve 1 and
p 1 acing the m~teri a 1 in a fi 11 wou 1 d have to be di S'cont i nued during these
months because of the snow and ice intrusion and the inability to obtain
satisfactory compaction with frozen material. However, other construction
activities could continue during this period (consideration being given to
the·cost of snow removal) with possibly the exception of the two coldest
months pf December (9.6°F) and January (9.4°F).
Productivity would decrease 10 percent to 30 percent from the normal or
base production rates during the periods of mid-October through November~
and February through mi d-Apri 1, and up to 50 percent during December and
January depending upon the activity and the exposure.
(e) Construction Methods
The construction methods assumed for development of the estimate and .
construction schedule, are gene.rally considered as "normal", tn line with
the available level of technical information. A conservative approach has
been taken in those areas where more detailed information will be developed
during subsequent investigation and engineering programs. For example,
normal drilling, blasting, and mucking methods have been assumed for all
underground excavation. Also conventional equipment has been considered
for major fill and concrete work. Various construction methods were
considered for several of the major work items to determine the most
economicallypractical method. For example, a comprehensive evaluation was
made of the means of exc av at i ng materia 1 from Borrow Area E and the
downstream river for the Watana dam shells. A comparison of excavation by
dragline, dredge, backhoe, and "sauerman'' (scraper) bucket methods was c
made, with consideration given to the quantity of material available,
distance from the dam, and location in the river or adjacent terraces.
(f) Quantity Takeoffs
Detailed quantity takeoffs were produced from the engineering drawings
using methods normal to the industry. The quantities deve 1 oped are those
1 i sted i ~ the detailed summary estimates in Appendix F.
(g) Indirect Construction Costs
·-
Indirect construction costs were estimated in detail for the civil con-
struction activities. A more general evaluation was used for the mechani-
cal an~ electrical work.
16-4
·····.\ . .
. . ;··-.·.: ... '
• J
.I··••·.
',. •'
I
I
I
I
I
I
I
I
I
I
••
I
I
I"
I
.I
I "'
I
I
'I
I
I
I
I
I
I
·.·1
I
I
I
I
I
I
'I "
. . ••
Indir·ect costs included the following:
-Mobilization;
-Techni:eaT and supervisory personnel above the level of trades foremen;
-All vehicle costs for .supervisory personnel;
-Fixed offices, mobile offices, workshops, storage facilities, and lay-
down areas, including all services;
-General transportation for wo.rkmen on site and off site;
-Yard cranes and floats;
Utilities including electrical power, heat, water, and compressed air;
Small tools;
-Safety program and equipment
-Financing;
-Bonds and securities;
Insurance;
-Taxes;
-Permits;
-Head office overhead;
-Contingency allowance; and
-Profit.
16.2 -!:!]tigation Costs
As discussed in previous sections, the project arrangement includes a number of
features designed to mitigate potential impacts on the natural environment and
on residents and communities in the vicinity of the project. In addition. a
number of measures are planned during construction of the project to mitigate
similar impacts caused by construction activities. The measures and facilities
represent additional costs to the project than would be normally required for
safe and efficient operation of a hydroelectric .development. A summary of these
mitigation costs is presented in Table 16.5. The costs include direct and
indirect costs, engineering, administration, and contingencies.
A number of mitigation costs are associated with facilities~ improvements or
other programs not directly related to the project or located outside the
project boundaries. These would include the following items:
-Caribou barriers;·
-Fish channels;
-Fish hatcheries;
-Stream improvements;
-Salt licks;
-Recreational facilities;
-Habitat management for moose;
-Fish stocking program in reservoirs; and
-Land acqui st ion cost for _recreation.
It is anticipated that some of these features or programs wi 11 not be required
during or after construction of the project. In this regard a probabi 1 i ty fac-
tor has been assigned to each of the above items, and the estimated cost of each
reduced accordingly. Th.e estimated cost of .these measur:es, based on this proce-
. dure, is approxi~ately $9 million. These costs have been assumed to be covered
by the construction contingency.
16-5
A number of stud.ies and programs wi 11 be required to,.monitor the impacts of the
project. on--the environment and to develop and record various data rluring projec.t
construction and operation1l These include the following:.
-Archaeological studies;
Fisheries and wfldl i fe studies;
.... Right-of-way studies; and
-Socioeconomic planning studies.
The costs for the above work have been estimated to be approximately $ -------and included in the owner•s costs under project overheads.
16.3 -Operation, Maintenance~ and Replacemeiit Cost.s
The f aci 1 i ties and procedures for operation and mai nten~nce of the project are
described in Section 15. Assumptions for the size and extent of these facili-
ties have been conservatively made on the basis of experience at large hydro-
electric developments in northern climates, noteably Canada. The annual costs
developed for operation, maintenance, and interim replacement for the Watana and
De vi 1 Canyon projects and the transmi ssi or~ faci 1 it i es are summarized in Tab 1es
16 ·~' 16 ._ and 16. .
16.4 -Engineering and Administration Costs
Engineering has been subdivided into the following accounts for the purposes of
the cost esttmat-es:
-Account 71
. Engineering and Pro3ect Management
. Construction t4anagement
. Procurement
-Account 76
. Owner • s Costs
Thrc total cost of engineering and administrative activities has been estimated
. at 12.5 percent of the total construction costs, including contingencies. This
is in general agreement with experience on projects similar in scope and com-
plexity. A detailed breakdown of these costs is dependent on the organizational
structure established to undertake design and management of the project~ as well
as more definitive data relating to the scope anti nature of the various project
components. However~ the main elements of cost included are as follows:
(a) Engineering and Project Management Costs
These costs include allowances for:
-Feasibility studies, including site surveys and investigations and
logistics support;
16-6
-·-·· •. ·-. '. '·-
1
I··
I-
I
I
:1
I-
I
I
I
I
I
I
I
I
I-
I"
•.
I;
-I
I --
I
I
I
I
I
I
I
I
I
I
I
I
·I
I
I
I
I
... Preparation of a license application to the FERC; -
-Techn·i cal and administrative input for other federal, state and 1oca 1
permit and license applicatlons;
-Ov~rall coordination and administration of engineering, construction ~
management, and procurement activities;_. ·-" :::~~="'"-:'c::;: ~~ >-·
-Overall p·lanning, coordination, and monitoring activities related to cost
and schedule of the project;
-Coordination with APA and reporting to APA regarding all aspects of the
project;
-Preliminary and detailed design;
Technic a 1 input to procurement of construction services, support
services, and equipment;
-Monitoring of construction to ensure conformance to design requirements;
-Preparation of start-up and acceptance test procedures; and
-Preparation of project operating and maintenance manuals.
(b) Co~struction Management Costs
-·
Construction management costs have been assumed to include:
-Initial planning and scheduling and establishment of project procedures
and organization;
-Coordination of onsite contractors and construction management
activities;
-Administration of onsite contractors to ensure harmony of trades,
comp 1 i ance with app 1 i cable regulations, and maintenance of adequate stte
security and safety requirements;
-Development, coordination, and monitoring of construction schedules;
-Construction cost control;
-Materia 1, equipment and drawing centro 1;
-Inspection of construction and survey control;
-Measurement for pa.J111ent;
-Start-up and acceptance test for· equipment and systems;
-Compilation of as ... constructed records; and
-Final acceptance. ,.
(c) Procurement Costs
-· Procurement·costs have been assumed to include:
-Estab 1i shment of project procurement procedures;
-Preparation of non-techni ca 1 procurement documents;
-Solicitation and review of bids for construction services, support
services, permanent equipment, and other items required to complete the
project;
-Cost administration and control for procurement contracts; and
-Quality assurance services during fabrication or manufacture of equipment
and other purchased items.
16-7
(d) Owner's Costs
Owner's costs have been assumed to include the fq1Towing:
... Administration and coordination of project management and engineering
organizations;
-Coordination with other state, local,_anrJ _te_deral agencies and groups
having jurisdiction or interest in the project;
-Coordination with interested public groups and individuals;
-Reporting to legislature. and the public on the progress of the project;
and
-Legal costs (Account 7?).
16.5 -Allowance for Funds Used During Construction
At current high leve 1 s of interest rates in the fi nanc i a 1 market-place, AFDC
wi 11 amount to a significant element of financing cost for the lengthy periods
required for construction of the Watana and Devil Canyon projects. However, in
economic eva 1 uati ons of the Susi tn a project, the 1 ow real rates of i ntere:st
assumed would have a much reduced impact on assumed project development costs.
Furthermore, as discussed in Section 18, direct state involvement in financing
of the Susitna project wi 11 also have a significant impact. on the amount, if
any, of AFDC. For purposes of the current feasibility ~tudy, therefore, the
convent1onal practice of calculating AFDC as a separate line item for inclusion
as part of project construction cost, has not been followed. Provisions for
AFDC at appropriate rates of interest are made in the economic and fi nanc i a 1
analyses described in Section 18.
16.6 -Escalation
As noted, all costs presented in thi~ Section are at January~ 1982 levels, and
consequently include no allowance for future cost escalation. Thus, these costs
waul~ not be truly representative of construction and procur~ment bid prices.
This is because provision must be made in such bids for continuing escalation of
costs, and the extent and variation of escalation \'Jhich might take place over
the l~ngthy construction periods involved. Economic and financial evaluations
discussed in Section 18 take full account of such escalation at appropriately
assumed rates.
16.7 -Cash Flow arid Manpower Loading Requirements
The cash flow requirements for construction of Watana and Devil Canyon are an
essential input to economic and financial planning studies discussed in Section
18. The basis for th·e cash flow are the construction cost estimates in January,
1982 dollars and the construction schedules presented in SectiDn 17, with no
provision being made as such for escalat1on. The results are presented in
Figures 16.2 and 16.3. Similarly, the corresponding manpow·er loading require-
ments are shown in Figures 16.4 and 1.6 .5. These curves were used as the basis
for camp loading and associated socioeconomic impact studies.
16-8
I
:1/
••
:I
I
I
:I
I
I'
I
I
I
I
I
·I
I
I
I
,I
I
I
I
I
I
I
I
I
'I
I
I
I
I
I
I
I
:I
I·
LIST OF REFERENCES
(1) Code of Federal Regulations, Title 18,. Conservation of Power and Water··
Resources, Parts 1 and 2, Washington, D. c .. , Government Pri-nting
· · --··· · ~-~--'~-·-{Jf-f-1-ee-; ·-l-9£'1;·-·c-"·· -· ------·· · --· --· ··
(2)
(3)
(4)
Handbook_ of Wages and Benefits for Construction Unions, January 1981_, U. s.
Department of Labor, Office of Construction Industry Services, 1981.
Caterpillar Performance Handbook, Caterpillar Tractor Co., Peoria,
Illinois, October 1981.
Roberts, WilliamS.~~ Regionalized Feasibility Study of Cold Weather
Earthwork, Cold Regions Research and Engineering Laboratory, July
1976, Special ~eport 76-2.
iABLE 16.1: SUMMARY OF COST ESTIMATE
Januer1 1982 Dollat·s $ X 106
Cate9ory Watana Devil Can~on Total
Production Plant $1,969 $ 766 $2,745
Transmission Plant 388 91 479
General Plant 5 .5 10
Indirect 449 222 671
Subtotal $2,811 $ 1,094 $3,905
Contingency 17 .. 5% 492 191 683
Total Construction $3,303 $ 1,285 $4,588
Overhead Construction 413 161 574
TOTAL PROJECT $3,716 $1,446 . $5,162
I
I
I
I
I
I
I
I
I
l"c.j
I
I
I
I
.~
I
I .,,
-· - - - - - - - - - - -... - - - ---
TABLE 16.2
ESTIMATE SUMMAR'( WATANA P57UOl .. Ul0
110 ALASKA POWER AUTHORHY TYPE OF ESTIMATE Preliminary CLIENT
PROJECT ____ s_US_.I_T_NA __ I1_YD_R_O_E~LE_C_T_RI_C __ PR_O_J_EC_T_. ____________ __ APPROVED BY ---'------------
JOB NUMBER ----:---------FILE NUMBER _P_5.,...70_m_ .• _m_'b __ _
SHEET 1 QW. __ 5 __
BY -------O#TJ":£ __ _
CHKD ~£
No. DESCRIPTION AMOUNT TOTALS REMARKS
PRODUCTION PLANf
330 L:and & Land Rights ...................................................... ,. $ 51
~
331 Powerplant Structures & Impro\lements ~··••••••••••••••••••••••••••••••• 74
332 Reser11oir, Dams & Waterways .............................................. . 1,519
333 W~terwheels, Turbines & Generators ..................................... . 65
334 Accesso£'y Electrical Equipment •••••••• ? •••.•••••.••••••••..•••••.••••.• 21
335 Miscellaneous Powerplant Equipment (Mechanical) ••••••••••••••••••••••• 14
336 Roads & Rail.roads ............... -••••••...••••...•••.•.••• -· • f:l •• • , ••••••••• 225
TOTAL PRODUCTION PLANT ............ , ................................... . $ 1,969
A~IR
No.
3~0
352
353
354
356
}59
I
TABLE 16.2
\~AT ANA ESTIMATE SUMMARY ,
CLIENT ALASKA POWER AUTHORirY TYPE OF ESTIMATE Preliminary
PROJECT ____ s_U_S_IT_N_A_H~Y_OO_O_E_L_E_CT_R_l_C_P_R_O_JE_C_T ____________ __ APPROVED BY ______________ __
DESCRIPTION
TOI AL BROUG~IT FORWARD ••••••• ~ •••••••••.•••.••••.•••• ,., :o ••••• • •• ·•• 61 ••••• ~ ••
TRANSMISSION PLANT
land & l .. a11d Rights ......................................... ~ . ,. •. ,. .•..•..
Substation & Switching Station Structures & lnprovellients ••••••8•••••••
Substation & Switching Station Equipment ••••••••••••••••••e•••••••••••
Steel Towers. & fixtures •.•....••.••..••.•.... ~ ..••••...••..•• " ....••..•
Overhead Conductors & De"' ices ........................................... " •
Roads & Trails ., .................................................. • ••.••••
fOTAL TRANSMISSION PLANT •••••••••••••••••••••••••••.•••••••••••••••••••
AMOUNT
$ 8
12
129
130
99
10
TOTALS
$ 1,969
$ 388
$ 2,357
JOB NUMBER P57Qttl..OO
FILE NUMBER P~FlU10 .. U6
-~--
SHEET 2 ~ 5 -----
BY ____ _ IDlTE ---
CHKO IDATE
- - --· .. -·-- -·"-- - - - - - - -
--
.. 1~111
No.
389
390
391
392
393
394
395
396
397
398
399
-----------
ESTIMATE SUMMARY WATANA
CLIENT ALASKA POWER AUTHORITY TYPE OF ESTIMATE __ P_re~l_i_m_i_n_ar_y_
PROJECT ____ s_us_I_T_N_A_H_Y_DR_O_E_LE_C_T_R_IC __ P_RO_J_E_c.T _______ · ______ __ APPROVED BY-----------
DESCRIPTION
TOTAL BROUGHT FORWARD
GENERAL PLANT
·•···········•···················•···•·•····•····
land & Land Rights ···················~···················••···~~t~·~~·
Structures & lrnpro'lements •...••••.••••••.•.• ·· ••• -. •..•.•..••••••.•.•••••.
Office Furniture/£quip..-ner1t ••••••••••••••••••..••••••••••••• ~ •••••••• ~ o.
Transportation Equipment •••..••••.•• * •••• ·o ••••••••••.•••.•• a ............. .
Stores Equipment ••••••••••••••••••••••••••••••••••••••••••••••••••••••
Tools Shop & Garage Equipment .......................................... .
Labot·ator~ Equipment ................................................... .
Power-Operated Equipment
Communications Equipment
Miscellaneous Equi~~nt
Olher Tangible Propert~
•.......•...............•......••.•..•........
••.•••••....••••..........•...................
·················~·····························
···············~·······························
TOTAL GENERAL PLANT •••••••••••••••••••••••••••••••••••••••••••••••••••
AMOUNT TOTALS
$ 2, 35"1
5
$ 5
$ 2,362
----
JOB NUMBER P57om.:no
FILE NUMBER P570Jll/ll6
SHEET 3 QfF 5
BY O..t.LW:£
CHKD OfA;T£
REMAR:~
Included under 3Jo·
Included under .331'
TABLE 16 .. 2
WATANA JOB-NUMBER P57rnin.IDO
FILE NUMBER P57f!liD..:06 ESTIMATE SUMMARY
CLIENT ALASKA POWER AUlHORUY TYPE OF ESTIMATE Preliminary SHEET 4 QiF 5
BY ~£ ----PROJECT ____ S~U_S_IT~N_A_~~lV_D_RO_E_L_E_CT_R_l_C_P_R_O_JE_C_T ____________ __ APPROVED BY ______________ __
CHKD
No. DESCRIPTION AMOUNT TOTALS REMAR!itS
--------+--------------------------------------------------~------------r-----------~-----------------------
61
. 62
63
64
65
66
69
T 0 I Al BROUGHT FORWA.RD ............................... c •••••••••••.•••••• ~ •
INDIRECT COSTS
Temporary Construction facilities ....... , •••••••••••••••••••••••• , ••••••
Construction EquiJ>illent .••.•.• -.. ~ ...... -. •••••.••.• e ••••••••••••• •. • ••••••••
Camp & Commi-ssary .................................................. C! ••• ·• ~ •
Labor Expense ••••••••••••••••••••••••••••••••••••••••••••••••••••••••••
Superintendence ................................................... , •••.••..
Insurance ·····························~································
r ees •• ~ ..................... 0 •••••••••••••••••••••• 0 • ·• ................... .
Note: Costs under accounts 61, 62, 64, 65, 66, and 69
are included in the appropriate direct costs
listed above.
TOTAL INDIRECT COSTS ••••••••••••••••••·•••••••••••••••••••••••••·······
$
449
--:-- - - ------- - -
$ 2,362
See Note
See Note
See Note
See Note
See Note
$ 449
$ 2,811
-------------------
TABLE 16.2
WATANA ESTIMATE SUMMARY JOB NUMBER P5713JD .. \OO
FILE NUMBER P57001.!.nb
CLIENT ALASKA POWER AUTiiORI TV TYPE OF ESTIMATE Preliminary SHEET 5 QJF 5 A~IU PROJECT _ _____;:S=U=S.::;.;lT:..:...N:.:....:A_;H=Y=OR..;..:..O=E=L=EC='T:..;.;R=IC=......:..P..;.;R=OJ=E=C;,.;_T ______ _ BY O'Am":.t:. ---APPROVED BY ___ "'---------
IT4mt. CHKD
No, DESCRIPTION
TOTAL BROUGHT FORWARD (Construction Costs) ...•.•.....•.••....•..•.•..•
Contingency 17. ·s~ .. ~ ...................................................... .
TOTAL CONSTRUCTION COSTS ••••••••••••••••••••-,•••••••••••••••••••&•••••
OVERHEAD CONSTRUCTION COSTS (PROJECT lNDlRECTS)
71 Engineering/ Administration ••••••••••••••• •.e................... ... . . . .. . $
72 -Legal Expenses •• ·• •••••••.••••.••••••••••.•••••••••••••••••.••••.•••••••••••
75 Taxes ...................................................... _ .•••••••••••••••
76. Administrative & General Expenses •••.••••••••• "........................ G
11 Interest ··~···········~······················~························
80 Earnings/Expenses IAJring Construction .................................. .
Total Overhead ••••••••••••••••••••••••4••••••••••••••••••••••••••••••·
TOTAL PROJECr COST •••••••••••••••••••••••••••• ~ •••••••••••••••••••••••
AMOUNT
413
-
-
-
-
-
TOTALS
$
$
"
$
2,811
492
3,303
413
3,716
REMAR;~
Included in 71
Not applicable
Included in 71
Not included
Not included
TABLE 16.3
ESTIMATE SUMMARY DEV ll CANYON
AIIU AlASKA POWER AUTHOR! fY CUE NT
PROJECT ____ s_u_S_lT_N_A_H_Y_D_RO_E_l_EC_T_R_IC __ P_RO_J_E_C_T ____________ __ APPROVED BY-------------..,...-
TYPE OF EST I MATE Preliminary
JOB NUMBER P5'7f.lt;t..l!ID
FILE NUMBER P57Uet ... !06
SHEET 1 Q,f= 5
BY 0./A.ll'E
CHKD fl/Al"E
No. DESCRIPTION AMOUNT TOTALS. R.EMAB.~'S
PRODUCTION PLANT
JJO -land & Land Rights ••••••••••••••oae••••••••••••e•••••••••••••••••••o•• $ 22
331 Powetplant Structures & liJllrovements •••••••e•••••••••••••••••••••••••• 72
J32 Reser~oir, Dams & Waterways ···~············••••••••••••••••••e••·••••• 571
JJJ \~alerwheels, Turbines & Generators •••••••••••••••••••• o ••••••••• • ••••• 42
334 Accessory Electrical Equipment •o•••••••••·•••••••••••••••••••••••••••• 14
335 Miscellaneous Powerplant Equipment (Mechanical) •••••••••6••••••••••••• 12
JJ6 noads & Railroads ·······•o••••·········~······························ 43
lOTAL PRODUCTION PlANT •• :II" •••••••••••••••••• " o •••••• ~ ................. , •• o ••• $ 776
--------~----------------~--------~------------------------------L-------------L-------------~~--------------------·---
·-----------------· -------''-~----... :
-·· -----'_ . -,. . : ':-. ---
--~---------------·-
350
352
35}
354
356
359
lABLE 16.,
ESTIMATE SUMMARY OEV ll CANYON
CLIENT ALASKA POWER AUTHORITY TYPE Of ESTIMATE
PROJECT __ s_u_sl_T_N_A_H_Y_D_RO_E_L_E_C_T_R_IC_P_R_OJ_E_C_T --------APPROVED BY __________ ~--
DESCRIPTION
TOTAL BROUGHT FORWARD •••••••••••••••••••••••••••••••••••••••••••••••••
TRANSMISSION PLANT
land & land "Rights· ............ ._ ........................................ e ••••••
Sllbstation & Switching Station Structures &. lnprovements ................. .
. SUbstation & Switching Stat ion Equipment ............................... .
-;' ~ -
Steel Towers & Fixtures ...• ._ ..•. il .............................. ~ ••.•• : ••••••.•
Overhead Conductors & Devices ......................................... .
Roads & Trails .••.••••..••••.••.•.• o •••.••••••.••••••• -....................... o
TOTAL TRANSM!SSION PLANT ···~···········~····••••••••••••••••••••••••••
AMOUNT
$
7
21
29
34
TOTALS
$ 776
$ 91
$ 867
JOB NUMBER P5700"..1liD
CHKO
REMARKS
Included in Watana. ~timat.e
Included in Watana, ftlS'ti·mate
TABLE: 16.3
JOB NUMBER -
ESTIMATE SUMMARY DEVIl~ CANYON P5;zum.oo
~~----------~· Fl LE NUMBER _P...,..5:_:mJ_n_._0_6 __ _,:_
CLIENT ALASKA POWER AUTHORITY TYPE OF ESTI MAlE Preliminar~ sHEET 3 J.O'F ----5 A~ Ill PROJECT SUSHNA HYDROELECTRiC PROJECT 8'1 ---------ID~TE __ _
!DATE APPROVED BY--------CHKD
No.
389
390
'
-; DESCRIPTION
TOTAL ·BROUGHT fORWARD. • ••••••••••••••••••••••• ~ ••.••••••••••••••••••••••
·GENERAL PLANT
land &: land Rigl1ts lr ., ....... ·t!• ••••••••••••••••••••••••••••••••••.•• o •••••••
St.ructures & lnpro"ements ••.•• o ..... _ •••••••••••••••••••••••••••••••••.•• ·•
391 Office Furniture/Equipment· ........ ~ ....................................... .
392 Transpnrtation Equipment ••••••••••••••• " ............................ • •••
393 Stores· Equipment ••••• o •••••• ~ •••••••••.•• ~. ·• •. • ............. ._ •••• ·• ••• • •••
394 Tools Shop & Garage Equipment ••••••••••••••••e••••••••••••••••••••••••
395 Laboratory Equiprnen·t •••• ·• ................................................. .
396 Po\oter (\>erated Equipment •••••••••••• ;, ................................... .
397 Communi·cat ions Equipment .................................................... .
398 Miscella1-.eous Equipment ................ ~ ................................ .
399 Other Tangible Property ........................................... , •••••
TOTAL GENERAL PLANT ······················~····························
---
AMOUNT TOTALS
$ 867
$
5
$ 5
$ 872
REM~-KS
Included under ~~
Included under-J;J51t
----
----
A~IU
61
62
63
64
65
66
69
No.
TABLE 16.3
ESTIMATE SUMMARY/ DEVIL CANYON
CLIENT .ALASKA POWER AUTH!lRI rY TYPE OF ESTIMATE __ Pr_e_l_im_i_n_a_,ry'--
PROJECT ____ S_U~SI_T_N_A'--H'--Y_DR_O_E_LE_C_T_R_Ic_·~p_RO~_J_E_CT ____________ ___ APPROVED BY ______________ _
DESCR!PTION
TOT.Al BROUGHT FORWARD ..................................................... .
INDIRECT COSTS
Temporary Construction Facilities ............................. ~ ••••••• •.•
Construction 'Equipment ..................................................... .
Caflll & Commissary ••• ·• ................................ _ •••••••••• 111: ......... .
labor Exp~~ae ••••••••••••••••••••••••.•••••••••••••••••••••••••••••••••
Superintendence •••••••••··~·o·~··••••••••••••~w•••••••••••••~·~••••••6
Insu~ance ····················~··~·····••••••••••••••••••••••••••••••••
fees ................................................. o ••• • •••• e •••• c.. •••••
Note: Costs under accounts 61, 62, 64, 65, 66, and 69
are included in the appropriate direct costs
listed abo11e.
TOTAL INDIRECT COSTS-············•·······•o••••••••••••••••••••••••••••
AMOUNT TOTALS
$ 872
$
222
$ 222
$ 1,094
-
JOB NUMBER P$71Jltl •. OO ----~~~------
FILE NUMBER _P_5_7,QID,;,.....;.. • .__o6 _____ ,....._
SHEET 4 {Q:ft= __ 5 _
BY -------llTJATE __ _
CHKD ~ATE
See. Note
See Note·
See Note
See Note
See Note
See Note
AIIR
71
72
75
76
. 77
BO
No,
TABLE 16.3
ESTIMATE SUMM~o\RY DEVIL CANYON
CLl~NT ALASKA POWER AUTHO.:.:.r<I::..;l:...'t: _________ _ TYPE OF ESTIMATE
·PROJECT __ __;;;;,S;:;.;US:;.:l:..:.l:.;.:NA~H..:..:VDR=O;.=:.EL::.:E::;:;C:..:.T::.:.;Rl=-=C:;....·.:...:PR~O:.::J;.=:.EC;;;..;T;...__ _____ _ APPROVED BY
DESCRIPTION -AMOUNT
TOTAL BROUGHT FORWARD (Construction Costs) ···-·······-···••4!1•••·,·······
Contingency 17. ·s-% •• _ ••••••••• -: ......................... .t: ••••.•••••••••• -a ••••••••
fOTAL CONSTRUCTION COSfS •••••••••••••••·····••••••••••••••••••••••••••
OVERHEAD CONSTRUCTION COSTS (PROJECT lNOlREClS)
:Engineering ....... ,.. ..•.•. , ....•.•••......•••.. o .................................. . $ 161
legal Ex·penses ..•.•.•.......•....•..••..••• ·· .••.• , •••••.•.•••.•.•.•••.•••• -
Taxes ·····~········•················~·······•·········~····•••••••••• -
Administrative & General Expenses ·•·•••••••••••"'•·········~·· .. ••••••• -.
Interest •••••••••···········~~··············•·•·•·····~·••••••••••••• -
Earnings/Expenses IAJring Construction ·············~·················· -
Total Overhead Costs •.•.•••• , ............. • ••••.••..•.•....••....•....•.•.••
fOTAL PROJECT COSl ••••••••••••••················•··••••••••••••••••••
---·",--·,-·· ~ .
0
Preliminary
TOTALS
$
$
1,094
191
1,285
161
1,446
JOB NUMaER P51UOl.!ii0
FiLE NUMBER P570m.J.!$
SHEET 5 O:!F
BY -DAIJE
CHKO QA"'i£
REMARJ§
Included in 71
Not Applicable
Included in 71
NOt lncluded
Not Included
.
5
"
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
'I
I
t¥JW···~ .,
iABlE 16.4: CONSTRUCTION SEASONS
DURAtiON ·OORAfldN
WORK ITEM. (MONTHS) (WORKING DAYS}
Granular Fill Placement 6.0 150
Impervious Fill Placement 5.0 100
Rock Fill Placement 1.0 170
Underground Work '10.0 250
Ctincrete ·?lacement a.o 200
(Generally)
Concrete Placement 10.0 250
(Devil Canyon Dam)
Other Abo"eground Work · 10.0 .250
SEASoN
START/FINISH
Apr 15-(lct 15
May 1-0ct 1
Apr 1-0ec 1
Feb 15-Dec 15
Apr 1-0ec 1
Feb 15-Dec 15
Feb 15-Dec 15
~~~.,..-·';'· -. ' '·
TAat..E 16 .. 5: MITI.tAtHlN MEASURES SlJ.1MARY Of COST$ INCORPORAT.ED
IN CONSTRtiCTlON COST .ESTIMATES --~---....;;.;.;:.....;;;.;;;.,;.;;;;;.;..;o;;.;;,.;.;;;.;;.;.;.....;;.;;.;;;,;....,;;:;.;;..;..;;.:..~;;.;;..-------........ ~ ........ · ... --
COSTS INCORPORATED IN CONSTRUCTION ESTIMAYES
Spillway·Valves in Dam
Main Dam at Devil Canyon
Service Spill~ay at Watana
Restoration of Barra~ Area D
Restoration of Bo~row Area F
Restoration of Cafll>
Restoration of Construction Sites
fencing around Caill's
Fencing around Garbage Disposal Area
Multileiel Intake Structure
Camp Facilities Associated with trying
to Keep Workers out of Local Communities
Restoration of Haul Roads:
SU8TOTAt
Contingency 17.5%
TOTAL CONSTRUCTION
Engineering 12.5%
TOTAL PROJECT
· .$ X 10 3
. WATANA
48;500
1,617
551
1,054
4,050
414
125
18,000
10,156
756.
85,221
14,2.1_4,
100,135
12,517
112,542
$X 103
D£YIL CANYON
14,510
700
2,016
305.
125
9,000
505
27,161
' 4, 753
31,914
3,989
35,903 148"555
.... •.
_'""·:~
1'1
~. . ~ .:
j
,, i
,~
I
1'.
.~.:
I
I
I
I
I
I
I
I
I
I
I
I
I
- -----·-.. : ····'----· ---
~--~-------------------~~-----·----~--------------------·.~~~· ------------------------------~-------.
-,..e 0 -(f)
1aJ -0 z
1aJ
0 -LL
(1.
1LI
·Z. -
LOCATlON-FAIRBANKS Ai<. LEGEND --·------MANUAL
--------EXCAVATION
----HAULING
80 ----------~--------~------~~~~----~----~~~~------~
so.~--------~----------~------~~---------+-----~--~~-------~
40
', ' ' ,,
', '
' ' ', ' ',
. "'
20 ~ ', ' '?-_ --
OL' ·-~~--~-,~~-_. ~----~--~~--~ ........ ........ _ .. ____ _
JAN MAR
FEB
MAY
APR JUN
MONTHS
JUL SEP NOV
AUG OCT DEC
EARTHWORK INEFFICIENCIES BASE.D ON MONTHLY TEMPERATURE ,LIGHTING AND 'PRECIPlTATlON
FfGURE IGJ
. ;
I
I
I
' ~· . '
1 t ·~ DEVELOPMENT. SCHEDtlLE ·
... · to. ·F.o.llow
' \' \l
'.
.;
,_ :
. 'I
.I
I
I
I
I
I
I
I
I
I
I
1
I
I
I
1
I
-~·""\
,, . ·,1··
·--,7
·a
,.' ~ ',1
···-.• :
·I,
I
I
I
•••
I
I
I
I
:-.,
I
il·. '
18' ..; ECONOMIC AND FINANCIAL EVALUATION ,
lS.l -EGonomic Evaluation
This· section provides a discussion of the kfJy economic parameters used in the
study and develops the net ~conomic benefits stemming from the Susitna Hydro-
electric Project. Section 18.l(a} deals with those economic principles relevant
to the analysis of _net economic benefits and develops inflation and discount
rates and the Alaskan opportunity values {shadow prices) of oil, natural gas and
coal. In particu1 ar the analysis -is focused on the longer term prospects for
coal markets and prices. This follows from the evaluation that in the absence_
of Susitna, the next best. thermal generation plan would rely on exploitation of
Alaska coal. ·The future coal price is therefore considered in detail to provide
rigorous estimates of prices in the most 1 ikely alternative markets-and hence
the market price of coal at the mine-head within the state, ·
Section l8.l(c) presents the net economic benefits of the proposed hydroelectric
power investments compared with this thermal alternative. These are measured. in
terms of-present valued diff~rences between benefi-ts and costs. Recognizing
that even the most careful estimates will be surrounded by a degree of uncer;-
tainty, the benefit-cost assessments are carried out in a probabilistic frame-
work. The analysis therefore ·provides both a most likely estimate of net eco-
nomic benefits _accruing to the state and a range of net economic benefits that
·can be expected with a 1ikelihood (confidence leve1) of 95 percent or more ..
{a) .§.conomic Principles and Parameters
(:i} _Economic Principles -Concept of Net Economic Benefits
The concept of net economic benefits has a profound importance to
the State of Alaska. A necessary condition for max.imizing the
. increase in state income and economic growth is the select i.on of
public or private investments with the highest present valued net
benefits to the state. In the context 0f Alaskan electric power
investments, the net benefits are defined as the difference between
the costs of optimal Susitna-inclusive and Susitna-exclusive (all-
thermal) generation plans.
The energy costs of po~er generation are initially measured in terms
of opportunity values or shadow prices which may differ from ·
accounting or market prices currently prevailing in the state. The
concept and use of opportunity values is fundamental to the optimal
allocation of scarce resources. Energy investment decisions should
not be made solely on the basis of accounting prices in the state if
the international value of traded energy commodities such as coal
and gas diverge from local market prices ..
The.choic,e of a time horizon is also cruci.al. If a too short-term
· planning period is se 1 ected, the investment rankings and cho ice.s
will differ markedly from those obtained through a more appropriate
long-term perspective. In other words, the benefit-cost analysis ,
would point to different generation expansion plans depending on the
selected planning period. A short-run optimization of state incomes
would, at best, allow only a moderate growth in fixed capital
18-1 .
formati.on, at worst, it would lead to underinvestment in not only
the energy sector but also in other i nfrastruct.ure faci1 ities such
as roads, airports, hospitals, schools., and communications.
It therefore follows that the Susitna project, as other Alaskan
~investments, should be apprais~d on the basis of long-run optimiza.-
tion, where the 1ong-run is defined as the expected economic li.fe of
the facility. For hydroelectric projects, this service life is
typi ca 11y 50 years or more. · The costs of a Susi tna-1 ncl us i ve gener..:
ation plan have therefore been compared with the costs of the next-
best a 1 ternat i ve which is the a 11-therma 1 generation plan and
assessed over a planning peri,od extending from 1982 to 2040, using
internally consistent sets of economic scenarios and appropriate
opportunity values of Alaskan energy.
Throughout the analysis, all costs and prices are expressed in real
{inflation-adjusted) terms using JQnuary 1982 dollars. Hence, the
results of the economic calculations are not sensitive to modified
assumptions concerning the rates of general price inflation. In
contrast, the financial and market analyses, conducted in nominal
(inflation-inclusive) terms, will be influenced by the rate of
general price inflation from 198.2 to 2040.
(ii) Price Inflation and Discount Rates
-General Price Inflation
Despite the fact that price 1 eve 1 s are generally higher in Alaska.
than in the Lo\~er 48, there is 1 ittle difference in the compara-
tive rates of price changes; i.e. price inflation. Between 1970
and 1978, for example, the u.s. and Anchorage consumer price
indexes rose at annua 1 rates of 6. 9 and 7.1 percent, respectively.
From 1977 to 1978, the differential was even smaller: the con-
sumer prices increased by 8.8 percent and 8.7 percent in the U.S.
and Anchorage (1).
Forecasts of Alaskan prices extend only to 1986 {2). These indi-
cate an average rate of increase of ~· 7 percent from 1980 to 1986.
For the lange~ period between 1986 and 2010, it is assumed that
Alaskan prices will escalate at the overall u.s. rate, or at 5 to
7 percent compounded annually.. The average annual rate of pr·i ce
inflation is therefore about 7 percent between 1982 and 2010~ As
this ""'is consistent with 1 ong-term forecasts of the CPI advanced by
leading economic consulting organizations, 7 percent has been
adopted as the study value (3~4).
-Discount Rates
I)
I
I
I
I
I
I,
,I
,I
I
I
I~
I
Discount rates are required to compare and aggregate cash f:l ows I
occurring in different time periods of the planning horizon. In
essence, the discount rate is a weighting factor reflecting that a .1. _
d9llar received tomorrow is h·orth less than a dollar received
18-2
I
I
' (l ... .> • . .., ... ' ·~ ~ • ~· .. . ~ • ... '.,,.~ -, • • ..r ~ ... • .~ 4 ,.. .. • , • • .. , ,p ~ i,J; • -• \: llo ;ae~~.~. ¥ , -G u
' • • I • • • "' ... ..-I' o "'' .. :· >v! -,•~ • ,"" •\Jd "\. .. '"• • p , • •'f 'f~"~ • ' ~' ~:1 • :. ~ "f#. fO • '!'!fr~ • ,.,.If; • • ,,
• I) . ' • "' .. :'.I '
I
I
I
I
I
I;
I:
I
I
I -
I
I
I
I
I
I
I
·. ~•
'-·
"J
,.
today.. This he 1 ds even .in an i nfl at ion-free economy as long · as
·the producti~ity of capital is positive. In other words, the
value of a dollar received in the future must be deflated to·
~eflect its earning power foregone by not receiving it today.. The
use of discount rates extends to both real dollar {e,conomic) and
escalated dollar (financial) evaluations, with corresponding
inflation-adjusted (real) and inflation-inclusive (nominal)
values ..
• Real Discount and Interest Rates
Severa'l approaches have been suggested for estimating the real
discount rate applicable to public projects (or to private pro-
jects from the public perspective). Three common alternatives
include:
•• the social opportunity cost (SOC) rate,
ee the social time preference (STP) rate, and
•• the government • s real borrowing rate or the real cost of
debt capital {5,6,7).
The SOC rate measures the rea 1 soci a1 return (before taxes and
subsidies) that capital funds could earn in alternative invest-
ments. If, for example, the marginal capital investment in
Alaska has an estimated social yield of X percent, the Susitna
Hydroelectric Project should be appraised using the X percent
measure of 11 foregone returns 11 or opportunity costs.. A short-
coming of this concept is the di ffi cul ty inherent in determi n-
ing the nature and yields of the foregone investments ..
The STP rate measures society's preferences for allocating
resource~ between investment and consumption. This approach is
also fraught with practical measurement difficulties since a
wide range of STP rates may be infer:"ed from market interest
rates and socially desirable rates of investment.
A sub-set of STP rates used in project evaluations is the
0\-Jner 1 s rea 1 cost of borrowing; that is, the rea 1 cost of debt
capital .. This industrial-or government borrowing rate may be
readily measured and provides a starting point for determining
project-specific discount rates. For example, long-term indus-
trial bond rates have averaged about 2 to 3 percent in the U.S.
in real (inflation-adjusted) terms (3,8). Forecasts of real
interest rates show average values of about 3 percent and 2
·percent in the periods of 1985 to 1990 and 1990 to 2000, res-
pectively. The u.s. Nuclear Regulatory Commission has also
analyzed the choice of discount rates for investment appraisal
in the electric utility industry and has recommended a 3 per-
cent real rate .(24). Therefore, a real rate of 3.0 percent has
been adopted as the base case discount and interest rat(~ for
the period 1982 to 2040.
18-3
,
,=-"=~,_
I, c
(iii)
Nominal Discount and Interest Rates
-
The nominal discount and interest ·rates are derived from the
real values ·and the anticipated rate of genera1 price infla-
tion. Given a 3-percent real discount rate and a 7-percent
rate of price inflation,_ the nominal discount rate is deter--
mined as 10.2 percent or about 10 percent*.
Oil and Gas Prices -
-Oil Prices
In the base period (January 1982), the Alaskan 1982 dollar price
of No. 2 fuel oil is estimated at $6.5u/MMBtu.
Long-term trends in oil prices will be influenced by events that
are economic, political and technologtcal in nature_, and are
therefore estimated within a probabilistic framework.
As shown in Table 18.1, the base case (most likely escalation.
rate) is estimated to be 2 percent (to 2000) and 1 percent from
2000 to 2040. To be consistent with Battelle forecasts, a
2-percent rate was used throughout the OGP p 1 ann ing period 1982 to
2010 and 0 p~rcent thereafter. In the low and high scenarios the
growth rates were estimated at 0 percent (1982-2051), and 4 per-
cent (to 2000); and 2 percent (beyond 2000), respectively. These
projections are also consistent with those recently advanced by
such organizations as ORI (9), World Bank (10), U.S. DOE (11)~
Canadian National Energy Board (12).
-Gas Prices
Alaskan gas prices have been forecast using both export opportun-
ity values (netting back CIF prices from tlapan to Cook Inlet) and
domestic market prices as 1 ikely to be faced in the future by
Alaskan electr"ic uti 1 it ies. The OGP analysis used market prices
as estimated by Batte 11 e, s i nee there are ind ic at ions that Cook
Inlet reserves may remain insufficient to serve new export
markets~
Domestic Market Prices
Table 18~2 depicts the low, medium and high aomestic market
prices used in the OGP analysis. In the medium (most likely)
case, prices escalate at real rates of 2.5 percent (1982 to
2000) J.nd 2 percent (beyond 2000). In the low case, there is
zero escalation and in the high case, gas prices grow at 4.
percent (to 2000) and 2 percent (beyond 2000).
* (1 +the nominal rate) = (1 +the real rate) x (1 +the inflation rate) = 1.03
x 1.07, or 1.102
18-4
I
I" ..
I
I
I.
I
I
I
I
I
,I
I
I
I
I
I
I
I
I
,I
I
I
I
I
I
I
I
I
I
l --·
I
I
I
I
I
I
I
I
a
II
.. Export Opportunity V~ues
Taule l8 .. 2 also shows the current and projected opportunity
value of Cook Inlet gas in a scenario where the Japanese export
market for LNG continues to be·, the alternative to domestic de-
. mand._ From a base period plant gate price of $4:.69 MMBtu (CIF
Japan), low_, medium and high price escalation rates have been
estimated for the interv a 1 s 1982 to 2000 and 2000 ' J 2040. The
cost of liquefaction and shipping (assumed to be constant in
real terms) was subtracted from the escalated CIF prices to
derive the Cook Inlet plant-gate prices and their growth rates.
These Alaskan opportunity values are projected to escalate at
2.7 percent and 1.2 percent in the medium (most likely) case.
Note that the export opportunity values consistently exceed the
domestic prices. In the year 2000, for example., the opportun-
ity value is nearly double the domestic price estimated by
Battelle.
(iv) Coal Prices
The shadow price or opportunity value of Beluga and Healy coal is
the delivered price in alternative markets less the cost of trans-
portation to those markets. The most 1 ike ly alternative demand for
thermal coal is the East Asian market, principally Japan, South
Korea, and Taiwan. The development of 60-year forecasts of coal
prices in these marke~s is conditional on the procurement policies
of the importing nations~ These factors, in turn, are influenced to
a large extent by the price movements of crude oil.
Historical Trends
Examination of historical coal price tre.nds reveals that FOB and
CIF prices have escalated at annual real rates of 1.5 percent to
6.3 percent as shown below:
. Coal prices {bituminous, export unit value, FOB U.S. ports) grew
at real annu:tl rates of 1.5 percent (195U to 1979) and 2 ... 3
percent (1972 ~~ 1979) (11) .
. In Alaska, the price of thermal coal sold to th·e GVEA utility
advanced at real rates of 2.2 percent (1965 to 1978) and 2.3
percent (1970 to 1978) .
. In Japan, the \1verage CIF prices of steam coal experienced real
escalation rates of 6.3 percent per year in the period 1977 to
1981 (20,21). This repre·sents an increase in the ;;verage price
from approximate'ly $35.22 per metric ton {mt) in 1977 to about
$76.63/mt in 1981.
18-5
As shown below, export. prices of coal are highly corre1 ated with
·on prices, and an analysi;s of production costs has not predicted
accur,ately the level of coal prices.. Even if the production: cost
forecast itself ·;s accurate, it will establish a minimum coal
prlce, rather than the market clearing price set by both supp-ly
and demand conditions. ···
. In real terms export prices of U.S. coal showed a 94-percent and
92-percent correlation with oil prices (195G to 1979 and 1972 to
1979J .. * . .
-Supply· function (production cost) analysis, has estimated
Canadian coal at a pr1ce of $23 .. 70 (1980 US$/ton) for S.E.
ijritish Columbia (B.C .. ) coking coal, FOB Roberts Bank, B.C.,
Canada (18,23).. In fact, Kaiser Resources (now B.C. Coal Ltd~)
has signed ag~"eements with Japan at an FOB Price of about $47 ~50
(1980 US$/ton (19). This is 100 percent more than the price
estimate based on production costs ..
.. The same comparison for Canadian B.C. thermal coal indicates
that the expected price of $55.00 (1981 Canadian~$) per metric
ton (2,200 pounds) or about $37.00 (1980 U~S. $) per ton would
be 60 percent above estimates founded on product ion costs (18,
19,23). 0
. In ltinger-term coal export contracts, there has been provision
for reviewing the base price (regardless of escalation clauses)
if significant developments occur in pricing or markets. That
is, prices may respond to market conditions even before the ex-
piration of the contract.**
. Energy-importing nations in Asia, especially Japan, have a
stated pol icy of diversified procurement for their coal sup-
plies. They will not buy only from the lowest-cost supplier {as
would be the case in a perfectly competitive model of coal
trade) but instead will pay a risk premium to ensure security of
supply (see Battelle 18,23).
-Survey of Forecasts
Uata Resources Inc. is projecting an average annual real growth
rate of ?.6 percent for U.S. coal prices in the period 1981 to
20UO (9). 7he World Bank has forecast that the real price of st.·:am
coal would advance at approximately the same rate as oil prices (3
percent/a) in the period 1980 to 1990 (10). Canadian Resourcecon
Ltd .. has recently forecast growth rates of 2 percent to 4 percent
(1980 to 2010) for subbituminous and bituminous steam coal (22).
*Analysis is based on data from the JAotld Bank.
**This clause forms part of the recently concluded agreement between Denison
.Mines and Ter:k Corporation and Japanese steel makers.
18-6
'I
.1'' . I
:_ ,.,.·
1,1
. ~ . l
~ ' ~ ~!
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
,I
I
I
I
"I ..
I
<
I
I
I
I
I
I
·I
I
I
I
I
I
I
-Opportunity Value of Alaskan Coal
Delivered_ Prices, CIF Japan
Based on these considerations, the shadow price of coal (CIF
price in Japan) was forecast using conditio:na1 probabilities
given low, medium and high oil price scenarios. lable 18 .. 3 de-
picts the estimated coal price growth rates and their associated
probabilities, given the three sets of oil prices. Combining
these probabilities with those attached to the oil price c~ses
yields the following coal price scenarios, CIF Japan.
Scenario
rvted i urn
(most likely)
Low
High
Probabi 1 ity
49 percent
24 percent
27 percent
Real Price Growth ·-
2 percent {1982-2000)
1 percent (2000-2040)
0 percent (1982-2040)
4 percent (1982-2000)
2 percent (2000-2040)
The 1982 base period price was initially estimated using the
data from the Battelle Beluga Market Study (18). Based an this
study., a sample of 1980 spot prices (averaging $1.66/MMB.tru) was
escalated to January 1982 to provide a stat4 ting value of
$1.95/MMBtu in January 1982 dollars.*
As more recent and more complete coal import price statistics
became available,· this extrapolation of the 19 sample was. found
to. .give a significar~t underestimate of actual CIF prices,., By
late 1981, Japan's a·1erage imJ:,Jrt price of steam coal reached
$2 .. 96/ MMBtu .. ** An important sensitivity case was the.refore
developed reflecting these updated actual CIF priceso The ;Up-
dated base period vaiue of $2.96 was reduced by 10 percent to
$2.66 to recognize the price discount dictated by quality dif-
ferentia 1 s bet\'.!een A 1 ask a co a 1 and other sources of Japaru:s'e
coal imports, as estimated by Battelle (18).
*The es.calation factor was 1.03 x 1.14, where 3 perce1t is the forecast real
growth in prices (mid-1980 to January, 1982) at an annual rate of 2 per·cent,
and 14 percent is the ld-month incre·ase if the CP1 is used to convert from
mia-1980 dollars to January, 1982 dollars.
** As reported by Coal Week International in October, 1981; the average Clf
value of steam coal was $75 .. 50 per metric ton. At an average heat value of
11,500 Btu/lb, this is equivalent to $2.96/MMBtu.
18-7
Opportunity Values in.Alaska
.• ·_Base Case-Battelle-Based CIF Prices,
-~ No ,Export Potential for rfea1y Coal· · ·
-
Transportation costs of $0.52/MMBtu were subtracted from the
init-ially est iroated CJF pr tee of $1.95 to determine the-op-
portunity val~~ of Beluga coal at Anchorage. In January
1982 dollars,. this base period net-back price is therefore
$1.43.. In subsequent years, the opportunity value is de-
rived as the difference between the escalated CIF price and
the transport cost (estimated to be constant in real terms)~
The real growth rate in these FOB prices is determined
residually from the forecast opportunity values.. In the
medium (most likely) case, the Beluga opportunity values as~
calate at annual rates of 2.6 percent and 1.2 percent during
the intervals 1982 to 2000 and 2000 to 2040, respB~tively~
For Healy coal, it was estimated that the base pe~iod price
of $1.75/MMBtu (at Healy) would also escalate at 2.6 percent
(to 2000) and 1.2 percent (2000 to 2040). Adding the escal-
ated cost of transportation from Healy to Nenana results in
a January 1982 price of $1.75/MMBtu.* In subsequent yParsr
the cost of transportation of which 30 percent is repre-
sented by fuel cost (which escalates at 2 percent) is added
to the Healy price resulting in Nenana prices-tha·l;-grow at
real rates of 2.3 percent (1982 to 2000) and 1.1 percent
(2000 to 2040). Table 18.3 summarizes the real escalation
rates app1 icable to Nenana and Beluga coal in the low,
medium and high price scenarios. . -
Sensitivity Case ... Update~ CIF Prices,
Export Potential fm .. -~~~ ~f.Y Coal
The updated CIF price of steam coal (2.66/MMBtu after ad-
justing for quality differentials) was reduced by shipping
costs from Healy and Bel_uga to Japan to yield Alaskan oppor-
tunity values. In January 1982, prices are $2.08 and $1.74
at Anchorag<a and Nenana, respectively. The differences be-
" tween esc a 1 a ted CIF prices and shipping costs result in FOB
prices that have real growth rates of 2.5 percent and 1.2
percent for Be~ uga coal and 2. 7 percent and 1. 2 percent for
Healy coal (at Nenana). Table 18.3 shows escalation rates
for the opportunity value of Alaskan coal in the low)c mediu!ll
and high price scenarios, using updated ,_ase period values.
* Transportation costs are based on Battelle (18,23).
18-8
I
I
·I
I
I
~-
1
I
I
I
I
I
I
I
I .,
I
'"'\\ .I
\\:
I ....
I
·~
I
I
I
I
I . ~·
I
I
I
I
I
I
I
I
I
I
I
I
I
(v) Generation PlanningpAnalysis ,.,. ltase Case Study Val~
Based on the considerations pr·esenter.: ":.:ct1ons (i) through (iv)
above, a consistent set of fue 1 price assembled for the base
ca·se probab i1 is tic generation planning · .1P) analysis, as shown in
Table 18.4~ The study values include pr·obabilities for the lowl
medium and high fuel price scenarios. The probabilities are common
for the three fuels (oil, gas and coal) within each !:iCenario in
order to keep. the number of generation planning runs to manageable
size. In the case of the natur-al gas prices, domestic market prices
were selected for the ·base case analysis with the export opportunity
values used in sensitivity runs. The base period value of $3 was
derived by deflating the 1996 Battelle prices to 1982 by 2.5 percent
per year. Coal prices were. also selected from the base case using
tla,7te11e's 1980 sample of prices as the starting point, with the
updated CIF prices of coal reserved for sensitivity runs. Oil
pr-·-ites have been escalated by 2 percent (1982 to 2040).
(b) Analysis of Net Economic Benefits
(i) Modeling Approach
Given the econom·ic parameters discussed in the previous section,~ the
alternatives for electrical energy generation in the Railbelt w~re
analyzed by comparing the production costs of electricity with and
without the Susitna project. The primary tool for the benefit cost
analysis was a generation planning model (OGP) which simulates pro-
duction costs over a planning period extending from 1982 t~ 2051.
The method of comparing the '1with 11 and 11 without 11 Susitna scenarios
is based on total system costs. The planning model determines the
total production costs of alternative plan? on a year-by-year basis.
These total costs for the period of modeling include all costs of
fuel and operation and maintenance (O&M) for all generating units
included as part of the system, as well as the annualized investment
costs of any production plants added during the period of study.
Facto~~·s which Cuntribute to the u·ltimate consumer cost of power
which are not included in this model are: all investment cost for
plants in service prior to 1993, costs of the transmission and dis-
tribution faciJ iti~s already in service, and administrative CO$tS of
utilities. These costs are common to all scenarios and therefore
have been omitted ftom the study, as they have no differential
impact on alternative generation plans.
In order to aggregate and compare costs, all annual costs frotn the
1993 to 2051 production simulations have been converted to a 1982
present worth (PW). These PWs are computed as the sum of two com-
ponents. The first is the 1982 PW of the first 18 years of model
study from 1993 to 2010. The second component is the est imatea PW .
of 1 ong-term systen costs from 2011 to 205 L
18-9
To model the system for an additional 40 years would r·equire the
developnent of further load fo-recasts and generatiNl alt-ernatives
that are beyond the rea 1m of any prudent pr:oJect ions. for-this
reason, the final study year (2010) production costs were assumed to
simply recur for an add it tonal 41 years, and aoded to the 18-year
PW, to establish the long-term cost differences between a1t.ernat ive
methods bf power generation.
(ii) Base Case Analysis
-Pattern of Investments With and Without Susitna
The base case comparison of the with and without Susitna plans is
based on production cost simulation for the period 19Y3 to 2051~
using mid-range va·lues for the load forecast, base period fuel
prices, fuel price escalation, base period capital costs and capi-
tal cost escalation. Load forecasts, fuel prices and construction-
costs are analyzed in Chapters 5, 18.l(b) and 16, respectively.
As discussed in Section 18.1(b), a real interest and discount rate
of 3 percent is used.
The with-Sus itn a p 1 an ca 11 s for" 680 MW of' capacity at Watana to be
avail able to the system 1n 1993. Although the project can come
on-line in stages durir.g that year, for modeling purposes full-
load generating cJpability is assumed to be avai1abie for the
whole year. The second stage of Susitna, the Devil Canyon pro-
ject~ is scheduled to come on-line in 2002. The running of ~he
project was tested for earlier and later dates and selection of
the year 2002 found to result in the iowest long-term cost. Devil
Canyon wi 11 have 600 MW of insta 11 ed capacity.
The without-Susitna plan is discussed in Section 6.7. It includes
3,200·-M~J coal-fired plants added in Beluga in 1993, 1994, and
2007. A 200-MW unit at Nenana -{s added in 1996. In addition,
970-MW gas-fired gas turbines (GTs) are added during tbe 1997 to
2009 period.
Base Case Net Economic Benefits
The economic comparison of these plans is shown in Table 18~5.
During the 1993 to 2010 study period, the 1982 !'W cost for the
Sus~tna plan is l3,119 million. The annual production cost is
$385.3 million in 2010. The present worth of this level cost -;:or
a period extending to the end of the 1 ife of the Uev il Canyon pr·o-
ject (2u51) is $3,943 million. The resulting total cost of tte
Susitna ... inclusive. plan is $7.06 bi1lion in real (1982) dollar~~~
pre.~ent ly v a 1 ued to 1982.
The non-Susitna plan modelled has a 19~2-PW cost of $3,213 mill ion
for the 1993 to 2010 period. With a 2010 annual cost of $491
million, the total long-term cost has a PW of $8.24 billion.
Therefore.~ the net economic benefit of adopting the Susitna plan
is ·l:.la bill ion.
18-10
I
·1·.· . . ,,
/.
I
I.
I
I
I
I ,,
I
1:
.I·:
. ' ·,
I
I'
I
I
I•.
I,
~.-
t
I
I
I
I
I
I
I
I
I
•••
I
I
I
I
o~ ••
t
:I
'.,·.·.···· .
. , .
··~
" !n other words, the pres~nt valued cost difference .between the
Susitna plan and the next ... best thermal expansion plan is $1 .. 2
billion in 1982 dollars~ The 1982 present valued cost advantage
of the Susitna plan {$1 .. 2 bill ion) is equivc1ent to a 1982 per
captta net economic benefit of $2;700-.in A1aska. Expressed in
1993 dollars (the on-line d;1te of Watana), this cost saving would
have a levelized value of $2.5 billion.* It is noted that the ·
magnitude of net economic benefits ($1.2 billion) is not sensitive
to alternative assumptions concerning the overall rate of price
inflation as measured by the CPI. The analysis nas been carried
out in real (inflation-adjusted) terms and therefor:e the present
valued cost savings-will remain at $1.2 billion regardless of CPI
movements, as long as the real (infl ation-adjustea) discount and
interest rates are maintained at 3 percent.
The Susitna projectts internal rate of return (fKKJ has also been
determined. This is the real (inflation-adjusted) discount rate
at which the with-Susitna plan has zero net economic benefits"
that is the discount rate at which the costs of the with Susitna
and the 11 alte.rnate 11 plans have equal costs. This IRR is about 4.1
percent in real terms, and 11.4 percent in nominal (inflation-
inclusive) terms. Therefore, the S~sitna investments would sig-
nificantly exceed the 5 percent nominal rate of return test pro-
posed by the State of Alaska in cases where state appropriations
may be involved.**
It is emphasized that these· net economic benefits and the rate of
return stemming from the Susitna project are inherently conserva-
tive estimates due to several assumptions used in the OGP analy-
sis .
. Zero Gro'?Jth in Far-Term Costs
From 2010 to 2051, the OGP analysis assumed constant annual pro-
duction costs in the Susitna and the non-Susitna plans. This
has the effect of excluding real escalation in fuel prices and
the (replacement) capital costs of thermal plants" and ther~by
underestimating the long-term costs of thermal generation
plants .
. Loss of Load Probabilities
The loss of load probability in the non-Susitna plan is cal-
culated at 0.09Y. This means that the system in 2010 is on the
verge of adding an additional plant~ and \vould do so in 2011.
These costs are however, not included in the anlaysis which is
cut off at 2011. On the other hand, the Susitna plan has a loss
of load probability of 0.025, .. anci may not require additional
capacity for several years beyond.2b10.
*$1.2 bill ion x 2.105, where 2.105 is the general price inflation index for the
period 1982 to 1993.
**See State of Alaska's SB 25, Section 44.83.470.
18-11
. I
I
~....,..---,--.,-,--.. -_-;;:--<--·. ~--· .~. ~-,--,--......,,...,. ··~---~-:,....,:-.. ~."'""":'.~ .• =';"-~,.,.-,·~·''"'7· ,., ..,.,.,. _. . ..,..,. ·'"""··=·-.-~"""'·"~. ~. __,....~,..........~-~~!!!!!!1
:-•.
Far...-Term En erg~ from Susitna
Some of the ~susitna energy output (about 344 GWh} is still not
used by 2010. This energy output would oe avail.abla to meet
future increases in projected demand in the summer months... No
benefit is attribut.ed to this energy in the: analysis.
Egua1.Environmental Costs
The OGP analysis has implicitly assumed. equal environmental
costs for both the Susitna and the non-Susitna plans. To the
extent that the thermal generation expansioo plan is expected to
carry greater environmental costs than the Susitna plan, the
economic cost savings from the Sus itna project are understated.
It is conc'eivable that these so-called negative exter~nalaties
from coal-fired electricity generation will have been mitigated
by 1993 and beyond, from the enactment of new environmental
legislation." However, such government action action would /
simply internalize the externality by forcing up the productiorr
market costs of therma 1 power. .
(iii) Sensitivity Analysi~
A sensitivity analysis has been carried out to identify the impact
of modified assumptions on net benefits. The analysis ~<~as directed
at the following variables:
-load forecast.
-real interest and discount rate
-construction period
period of analysis
capital costs
~ Sus itna
. Alternatives
~ O&i~ costs
~·base period coal price
-real escalation in capital and O&M costs and fuel prices
-system r e 1 i ab il it y
-Chackachamna included in non-Susitna plan.
Tables 18.6 to 18.13 depict the results of the sensitivity analysis.
In particular, Table 18.13 summarizes the net economic benefits of
the Susitna project associated with each sensitivity test. The net
benefits have been compared using indexes relative to the base case
value ($1.2 billion} which is set to 100.
The greatest variability in results occurs in sensitivity tests per-
taining to fuel escalation rates~ dis~ount rates, and base period
coal prices. For example, a scenario with high fuel price escala-
tion results in net benefits that have a value of 253 relative to
the base case. In other words, the high case provides 253 percent
of the base case net benefits. In general, the Susitna plan m~in
tains its positive net benefits over a reasonably wide range of
values assigned to the key-variables.
18-12
··I· ' . : ,,
II
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
1·· ..
I
I.
I
I
I
I
I
I
I
I
~I·
I
.I
I
-.~t
I
••
. I
~.· ...•. ~.·~.~ ..
.-~,'~ •,
Li£
JA multi~variate analysis 1n the form of probability trees is also
currently being undertaken lo test the joint effects Of varying sev-
eral assumptions in combinatiqn rather than individually. This ·
probabilistic analysi-s will provide a range of expected net economic
bene.fits and probability distributions that identify the chances of
exceeding particular values of net benefits at given levels of con-.
fidence. ·· · · ·
18.2 -Risk Analysis
A risk analysts was undertaken to provide a basis for determining the extent to
which perceived risks are likely to influence capital costs and schzdule. In
addition, because a mature Susitna project would represent a major portion of
the tott.l-generation system, a further risk analysis was made to assess the pro-
bability and consequences of a long-term outage of the proposed transmission
system. This section summarizes both risk analyses. A more detailed report is
included in the rroject documentation for Subtask 11.03, Risk Analysis.
(a) Approach
Any major construction effort is inevitably exposed to a large number of
risks. Low probabi'lity magnitude floods may occur at critical periods of
construction: accidents may happen: sub-surface investigations, no matter
how thorough, cannot always predict actual conditions uncovered when the
major excavations are undertaken: the normal estimating process impl ic ity
assumes a set of reasonably "normal 11 expectations as direct costs are
deve1oped, add·ing a contingency to the directly computed total on the
grounds that problems usually do occur even though their specific nature
may not be accurately foreseen at the outset.
The Susitna risk analysis took explicit account of 21 different risks~
apP.lying them, as appropriate, to each major construction activity. The
effort involved combining reasonably precise data (e.g. the probability
that a particular flood crest will occur in any given year can be deter-
.mined from analysis of hydrologic records) with numerous subjective judge-
ments (e.g. until a particular flood crest does occur, it cannot be known
with any degree of certainty what damage it will cause. The overall meth-
odology is illustrated in Figure 18.4-1.
(b) Elements of the Analysis
Figure 18.4-2 graphically depicts important questions which were addressed
at the start and relates them to elements of the analysis~ Each element is
further subdivided as follows:
( i) Configurations
Three primary configurations were considered:
-The Watana hydroelectric project (with trans;mission);
-The Devil Canydlf-hydroelectric project (with transmission); and
-The Sus itna transmission system alone .
18-13
(ii)
Separ,ate risk studies of these configurations permitted the produc"!!
t ion of data which can be aggregate;d in various ways to accommodate
alternative npo~ler"!"on-:line 11 dates which differ according to the
various demand ~forecasts·~
ConsisuratJon States
Two configuration states were considered:
-Construction Period --applicable to Watana and Devil Canyon
-Operation Period -applied only to the transmission system config-
uration.
(iii) Risks
Twenty-one risks were identified for consideration ip the analysis
and were grouped as fo 11 ows:
-Natural Risks
• flood
. ice
wind
seismic .
permafrost detflt' i oration
. geologic conditions
. low str"eamflow
-Design Controlled Risks
. seepage piping erosion
• ground water
-Construction Risks·
. equipment availability
. labor availability strikes
mater i ai avail ab i1 i ty
. equipment breakdown
matedrial deliveries
. weather
-Human Risks
. contractor capability
construction quality control
accidents
. sabotage .vandalism
-Sp·ecial Risks
. regulatory de·l ay
. estimating variance
18-14
·I';
"' .
.•. :···· .. ~ :. ~ !
.I.
: ,' .. ·<'
I
I
I
I
I
I
I
I
I
I
I
I
I.
I
I
I·
I
i'l
I
I
I
I
;''I
I
I
I
I
I
I
I
I
I
.:1
I
I
(iv1 Activities
For each configuration state involving constructi-on, up to 22 activ-
ities ·were considered, For Watana, for example, these included:
-main access
-site fac i 1 it ies
diversion tunnels
cofferdams
main dam excavation
-main dam fill initial portion
-main dam fill final portion
-relict channel protect~on
-chute spillway
-emergency spillway
-service spillway tunnels
-intake .
-penstock
-powerhouse
transformer gallery
-tailrace and surge chambers
-turbine-generators
-mechanical electrical equipment
... switchyard
-transmission
-impoundment
-test and commission.
(v) Damage Scenarios
Up to ten .different damage scenarios were associated with each logi-
cal risk-activity combination. While these varied significantly
from one risk-activity combination to another, they generally des-
cribed a range of possibilities which accounted for discrete i.ncre-
ments extending from "no damage 11 to •!catastrophic loss".
(vi) Criteria
The consequences of realizing particular risk magnitudes for each
activity were measured in terms of the following criteria:
-cost implications
-schedule implications
-manpower requirements
(vii) Boundary Conditions
The following assumptions and limitations were established t6 permit
. a • a reasanable and consistent analys1s of the problem:
18-15
..
-All cost -estimates were made in terms of January 19B2 dollars.
Thus, results are presented in this report in terms only of real
potential cost variations~ exclusive of inflati~n.
-The analysis was 1 imited only to the construction periods for
Watana and Devil Canyon since the greatest cpotent i al cost and
schedule variance would be possible during these periods. The
risk analysis for the operating ·period was associated solely with
the transmission system since that cor figuration represents the
most 1 ikely source of a major system outage during the project
operation.
-The risk analysis was accomplished concurrently with finalization
t ion of the total project cost estimate and was necessarily asso-
ciated with the feasibility level design. There is clearly some
potential foi design change as the project proceeds; a further
risk analysis should be undertaken coincident with completion of
final detailed design and prior to committment-to major ~onstruc
tion actiyities. Even so, the "estimating variance" risk takes
into account the fact that some design changes are likely to
appear as detailed design effort proceeds.
- A great deal of subjective judgement was necessarily involved in
assessing certain probab il it ies and in predicting pass ib 1 e damage
scenarios. This effort was accomplished initially by individual
qualified professionals in the variou~ disciplines and was sub-
jected tQ iterative group review and feedback efforts. To the ex-
tent that individual biases entered the analysis, their effects
were probably mutually offsetting. Even so,-sensitivity tests
were made for risk_s \-Jhich were important contributors to the final
results. ·
-The risk list do~s not include the important possibility of fund-
ing delays or of financing problems. These issues were dealt with
in a separate financial risk analysis as discussed in paragraph
18.5 below.
(c) Risk Assessments
For each of the risks identified in paragraph 18.2(b) {iii) above, the
assessment commenced with detailed definition of credible. events. Where
flood was identified as a risk, for example, the potential magnitudes and
associated probabilities of the floods was estimated. Data sources ranged
from reasonably accurate scientific data {particularly applicable to the
natural risk category), historical experience on water resources projectss
to subjective group judgements where data gaps existed.
I
•••• •••
. .
I
I
·I
I
I
I
I
I
I
I
I
••
I
In each case, the maximum cr--edible .. event was first established. This •...
choice set an upper limit on a scale of possible events starting at "no
damage" situation. Continuing with flood as an example, the m,aximum credi-
ble event was considered to be the probable maximum flood which had been
-,computed in ·the hydrologic studies (corresponding to a return period of I '
I
' '~
18-16
I.
I
I
I
I
I
I.
I
I
I
I
I
I
more than 10,000 years and an annual probabll ity of-occurrence of less tha.l'l---,"'_:c· ~..,,::
. 0001)'. -
Once risks were defined and logical risk~activity combinations were re-
viewed, the consequences of r_eal izing each selected risk magnitude were
considered (lf this risk magpitudejs realized, will a partially completed
structure· be damaged? Will it fatl? If it fails, is some other work in
progress disrupted?). Because of the uncertainties associated with these
projections, a range of damage scenarios and asso<;iated probabilities of
them occuring was established.
Even if a particular risk level is realized and a particular damage is suf ....
fered, the cost and schedule of restoring the activity are difficult to
precisely establish. Each of the risk analysts therefore provided three
values for each criterion:
-·a m~,.,imum value corresponding to the one time in twenty that the weather
is particularly good, materials are readily available, no accidents
occur, etc.
- a modal value associated with the most 1 ikely expectation of the analyst;
-a maximum value corresponding to the one time in twenty that everything
is more difficult than expected.
In the computerized calculation process, the three criterion values sup-
plied by the risk analyst were fitted to a triangular distribution, which
approximated the beta distribution illustrated at the bottom of Figure
18.2-3~
In effect, then, designation of the three conceptual criterion values led
to generation of a histogram with relatively narrow intervals and a nearly
continuous range of possible values over a relatively wide spectrum.
Figure 18.2-3 illustrates the structural relationship for handling risk-
activity combinations, damage scenar·~os, and criterion values.
While the procedure described above is generally applicable, some commen-
tary on particular aspects of its application and on certain unique risks
is appropriate~
(i) Tht1 terminology "damage scenario" has been used fo.r convenience
s;i; 'Ce most identified risks will normally be thought of as reasons
that the cost will be higher than had been estimated or that the
schedule will be exceeded. I~ fact, however, the proces~ does per-
mit consideration of what might be regarded as a "negative 11 damage
scenario. The geologic conditions risk is an excellent example.
The cost estimate was produced on the basis of estimates of require-
memnts for some concrete lining in the penstocks, extensive grout-
ing, a certain level of rock bolting, and the like. If geologic·
conditions are found to be better than currently assumed, the costs
could be less and the schedule might be accelerated.
18-17
.(ii)
{ i ; i)
The estimating __ variance risk was ·treated in a.sPHcial ,:11ay .because it
cannot easily be conceptualized i.n physical ter!ll~L --rc-·acc6-unts for
inevitable differences which do occur between ·eshhuates and actua1 ·
bids, and between bids and actual activity costs -even in the ab-
sence of an_y other identified risks .. Its probability of o~currence
and -associated range (fractions or multiples of the basic estimate)
were determined from historical data on water resources projects.
It includes, but is not necessarily limited to; such considerations
as:
~ the influence of competition and market pressures
-estimating discrepancies or errors in unit quantities on the parts
of both owner's estimator and bidder
-particular contract forms and the owner's ~cceptance non-
acceptance of certain risks;
-labor market conditions and the nature of project labor agree-
ments
-productivity and efficiency changes over time;
-the cost implications of variances between activity schedules and
actual activity durations;
-the potential for scope changes over time;
-extraordinary escalation of. construction costs above the underly-
ing inflation rate.
In addition to estimating variance, a second special risk is asso-
ciated with regulatory matters. Various legislated controls will
most certainly be applied to the Susitna project and it is a a rel a-
tively simple matter to compute the minimum time in which regulatory
requirements could be satisfied. It is a far more difficult task
indeed to estimate the precise nature and duration of poss·Ib1e
future regulatory delays .. It would also clearly be inappropriate to
attempt to app.ly regulatory risks at the activity level.
This risk was handled by developing a separate distribution for a
range of periods necessary for .satisfaction of important licensing
and permitting requirements~
Data used in arriving at a distribution were based on recent experi-
ences on other water resources projects as well as on distussions
with staff members of the Federal Energy Regulatory Commission.. The
effect of applying the regulatory risk is prl~arily one of shifting
the starting time for commencement of construction activitiess lead-
ing to corresponding change in the projected completion time... A
lesser effect of. the r~gulatory risk was to introduce delays dur\ng
construction. ·
Regulatory requirements have been an important influence during the
past decade on major construction costs and schedules, though it is
difficult to isolate their effects. In order to separately consider
estimating variance risks and regulatory risks, "estimating vari-
ance" probability determination relied heavily upon water resources
18-18
I
I
• ••
I
I
I
I
I
I
I
I
I-
I
I
I
I
I
-,,
I
••. ' . .
I
I
I
I
I
I
I
I
I
-• .. ,
)
:I
I
--·
I ,n
-----~~-.,~-. .,.,, ·~·"tonsttuct ic.\{1 -data ~dev~1oped for pr-ojects essentially completed prior
· to the pa'ssage of the National Environmental Pol icy Act_ (NEPA) .-As
noted .above~-regulatory risk probahn ity distributions \vere derived
from-more recent projects ..
. ( iv) Each of the various risk magnitude probabilities was originally
calculated as an annual value. On a risk-activity.by risk-activity
basis; these _annual values wer·e then converted by standard computa-
tional procedures to provide a probability of occurrence during the
duration of the acitivty.
(v) The concept of 11 response is particularly important in the formal
risk analysis process. As the terminology suggests, a 11 response 11
represents the action-to be taken if a particular event occurs ..
There are two kinds of "response11
• The first -and most often used
-is an expected react ion to the occurrence of a particular damage
level (i.e. if tt,l;is damage le.vel is incurred, then what action must
be taken to restore the activity to its pre-damage status? ·And what
cost, schedule, and manpower implications (consequences) will
result?). A second kind of response can a1 so be considered ~nd ·Jt
provides an important link between the design tean and the risk
analysis team. This latter type is the "preventive response" (i,e.
what changes might reasonably be made in the design and or construc-
tion procedures \vh i chwoul d permit us to avoid or reduce a particular
damage level? Is the cost and schedule change which might ensue
worthwhile when com::Ja~ed to the probability and magnitude of the
consequences which would otherwise be incurred?} A number .of pre-
ventive responses ·were identified.by risk analysts during the risk
study and several of these weri incorporated into the project design
and dedsi~1n criteria. There may be further opportunities for pre-
ventive re'sponse. Since none would be chosen unless it offered a
net benefit to cost and or schedule, it may reasonably be concluded
that as dHtailed design rJrocaeds and as subsequent risk analysis
updates :areaccomplished, a gradual reduction in the spread of
possible values can be expected.
(d) Interpretation of Results
( i) Presentation of Data
Minor variations in activity costs were generated by the estimating
team concurrent with development of the risk analysis. In addition,
account was taken of the expectation that construction costs will
escalate at a fa.ster rate than normal inflation -both in the eco-
nomic analyses and the risk analyses. To avoid confusion regarding
absolute cost values, the results of the risk analysis are presented
in this section as percentages of the estimated project cost or as
ratios between actual costs and estimated costs.
(ii) Watana Cost-Probability Distribution
Figure 18.2-5 illustrates the cumulative distribution of total
18-19
direct costs and their rel.ated non .... ectceedance 1Jrobabi 1 itfes as de-
termined~in the risk analysis$ Certain important points noted on
the figure ar-e inter.preted as follows:
... The project direct cost estimate, including contingencies, ,was
-presented in Chapter 16. Point 11 A11 ott Figure 18.2-5 corresponds
to this project esti.mate: the analysis suggests that the probabil-
ity of completing Watana for less than the project estimate which
includes a 17.5 percent contingency allowance.
-Point 11 811 corresponds to the ,.low" cost estimate which was teste.d
for sensitivity in the OGPS system cost analysis. The probability
that Watana will be completed for less than this cost estimate is
about 46 percent.
-Point "C 11 on Figure 18.2-5 corresponds to a cost equal to the
11 high 11 estimate tested in the OGPS analysis to determine the
effect of such a cost on total proje_ct economics. The risk
analysis suggests that there is a ~0 percent probability that this
cost will not be exceeded.
,
-As wi 11 be noted from Figure 18. 2 ... 5 t there remains a sma 11 but
measu;--able possibility that the project costs will exceed even the
11 high 11 ~stimate value at Point 11 C11
• It can be argued that the
degree of conservatism which was used in the analysis has over-
stated the possibility of extreme upper limits on total cost.
Paragraph (v) below addresses this issue, comparing these results
with historical data.
The expected value of the actual cost is 90.25 percent of the
project estimate.
(iii) Devil Canyon -Probability Distributions
Figure 18.2-6 provides the cumulative probability distribution for
Devi 1 Canyon costs. Points A, B, and C on the curve corresponj to
those discussed above for Watana and are associated with probabil-
ities of 74 percent, 47 percent, and 90 percent, respectively, for
actual percentages of the project estimate being less than indicated
values. Once again, a not insignificant long "tailu in the extreme
upper righthand portion of the distribution provides a measure of
the potential exposure to large overruns. The expected value of the
actual cost is 91.5 percent of the project estimate.
(iv) Total Project Distribution
Figure 18.2-7 combines the separate Watana and Devil Canyon proj-
ects, providing a cumulative distribution for the Susitna Hydro-
electric Project as a whole. Points A, B, and C now have associated
probabilities of non-exceedance of 73 percent, 47 percent and
18-20
I
I
.I
I.
I
I
I
I''
1.
I
• ••••• ••
. .
I
I·
I,
1:
1 ,~
'~
.:_1 ,,
I
I
·I
I
I
•••
·I
I
...
il
"I
··I
I
I
I
I
,.··I· . '
I .
I
tv)
90 percent, respectively, suggesting that a broad range of total
project cost ratios. are possible. In the 10 percent range and 90
percent prob·abil tty interva1, the cost ·range spans nearly three
billion dollars. If the project follows historical patterns, i.t may
be expected· that this ran_ge will narrow over time as detailed design
and constructi.on proceed .. Note that the cost distributions are in
every case based upon ·January 1982 do dollars and do not account for
the effects of inflation. Interest during construction or finance
cha-rges are not included. Only the potential for extraordinar·y con-
struction cost escalation (over and above inflation) has been taken
into account. It follows that if the project 1s completed in the
next several decades, the final 11 actua1" cost will have to be ad-
justed to equivalent 1982 dollars ·if it is to be compared with ri.sk
analysis results presented herein.
Comparison with Available Data r
During the assessment of the important "estimating variance 11 risk
(see paragr-aph 18~2(c} {ii) above), historical data for 49 federal
water resources projects completed prior to passage of NEPA were
considered. Figure 18.2-8 offers a cumulative probability histori-
cal program for various cost ratios. In each case!1 the cost ratio
reflects the actual pr~oject cost (after adjustment for-inflation)
divided by the 11 initia1 11 estimated cost. It will be seen that rela-
tively large overruns have occurred in the past, while there is also
evidence that a substantial number of water resources projects have
been accomplished for less than the originally estimated costs.
In order to compare this. information with the Susitna Risk Analysis
results, it is necessary to determine the meaning of 11 initial'1 .esti-
mate in terms of the historical data. In each case, the 11 initia1 11
estimate is the estimate presented to the Congress at the time that
a request vJas made for projtSct authorization. Thus, it would be in-
appropriate to regard the current Susitna estimate (as discussed in
Chapter 16) as an "initial" estimate in the federal sense .. Fortun-
ately, however, the Susitha project does have a long history of fed-
eral involvement. Indeed, the Corps of Engineers p~ovided a de-
tailed 11 initial" estimate in 1975 as the basis for seeking authori-
zation for important design activities. This uinitial 11 estimate was
further updated by a second uinitial" estimate in 1979 after some
additional exp1oratory~.work and further analysis were requested by
the Office of Management and Budget. Inclusive of contingencies and
excluding lands, the direct cost 11 initia1" Corps of Engineerst esti-
mate (from the 1979 report) in January 1982 dollars for the Watana
IJevi 1 Canyon (thin. arch dam) project was used as the denominator for
display of possible Susitna cost ratios.
Figure 18.2-9 overlays the results of the Susitna risk analysis on
the historical data. Note that the cost ratios differ on this
display from those on Figure 18.2-7 because of the necessity to use
the 11 initial'' estimate for comparison purposes.
~18-21
As may be s~en fronfF'l'gtire 18.2-9, the Susitrta risk analysis results
reflect a more ·pessimistic expectation at low cost leve1s than ~the
hi.storica1 data would appear to indicate is appropriate.
(vi) Schedule Risks
At the same time that minimum, rnoda1, and maximum cast val.ue·s were.
estimated for each damage scenario in each risk-activity set, esti-
mates were also made o.f similar values for potential schedule
changes. As a result, schedule probability distributions were gen-
erated for each major activity. HovJever, these individual distri-
butions could not be combined in the same manner in which the. cost
data were handled.
A critica-l path network was prepared for the entire set of activ-
ities for each configuration .. Individual probability distributions
for critical activities were then combined tr yield a distribution
for t~e total project schedule. . .
Several critical paths were identified in theprocess, since a long
delay on a non-cirtical activity can, of course, place that activity
on a new critical path. The 11 raw 11 schedule delay distribution .was
then considered in the context of a one-year schedule contingency
which had been suklt into the original estimate and in light of
regulatory delay risks. The resulting distributions are discussed
and interpretBd as follows:
-Figure 18 .. 2-12 provides a cumulative probability distribution in
months from the scheduled completion data for the Watana project ..
It reflects all risk contributions except those posed by regula-
.tory requirements. It is based upon a critical path through the
main dam construction and takes into account the one-year schedule
cont.ingency. The indicated probability of completing the project
ahead of schedule or on time is about 65 percent. There is only a
17-percent chance of completing the project a year early (i ,e. in
1992)~ . .
-Figure 18 .. 2-13 provides a similar distribution after regulatory
risks are accounted for. Two components are included: (1) prior
to the start of construction, a 1 icense must be issued by the
Federal Energy Regulatory Commission. There is a small chance
(estimated to be 25 percent) that the license will be issued a
year earlier than the current 30-month licensing schedule antici-
pates. The probabi 1 ity of meeting or improving upon the 30-month
estimate is about 72 percent and there is a 90-percent probability
that not more than 38 months wi11 be required; (2) during the con-
struction period, regulatory delays may be imposed as a result of
various permitting requirements, injunctions, etc. These delays
yield only increases in schedule and range from a 50-percent prob-
ability of delays of a month· or less to a 95 percent probabi 1 ity
that regulatory delays c!rrring construction will not exceed 12
months.
18-22
_I
I:
I~
~~ :-.• ' ·.· .. {l!
-~
1··--.'".'
.. : '~ :'
.-: i
I
I'
I
I
I
I
I
I-
I
I
I ,.
I
I
I
I
I
I
I
I
=•
I
I
I
I
I
I
I
I
I
I
I ,,
I
:~ ··'· '.
r . ~ .
t.' -.
1.-'
As may be seen from Figure 18 .. 2-13, the net effect of the re,gul ato"ry
risks is ,to broaden the range of possible values. At the 1ower end
of the distribution, it will be~noted that the chances of completing
at least a year early will have increased to near1y 40 percent -·
primarily because of· the chance of getting a 1 icense early and
therefore, starting early. ·No significant change appears .for the
probability of meeting or improving upon the schedule. A substan-
tial effect is evident in the upper portion of the curve where the
chances of long regulatory delays have pushed out the ~5 percent
confidence level to an expect at ion of no more than three months
attributable to risks other th&n regulatory, as may be seen on
Figure 18.2-12.
While similar distributions cna be plotted for Devil Canyon, they
are less meaningful since there is flexibility associated with its
starting date.
(vii) Transmission Line Risks
The separate risk analysis of the Susitna transmission system was
conducted to determine the probability of significant power supply
interruptions at the two major load centers in Anchorage. and
Fairbanks. The methodology was generally similar to that described
in preceding paragraphs. Recogn·; zing that the system is assumed to
be be in an operating mode, those risks which had applied only for
construction in the preceding analysis (e.g. contractor capability)
were eliminated from the risk list. Additions to the list were made
to account for the potential effects<>of lighting, aircraft, collis-
ions, and anchor-dragging in Knik Arm {.applicable to the submarine
cable segment). Account was taken of re.dundancies designed into the
system (e.g~ a 1oss of one line in the three-line system extending
south toward Anchorage can be tolerated with no loss of energy
delivery capability),
In addition, special attention was given to dependencies (e.g~ an
earthquake which causes the loss of two lines will very 1ike1y knock
out the third. On the other hand, vandalism which.causes an outage
on one line is only infrequently expectt.ri to extend to all lines).
Important assumptions included the availaiJility of well-trained re-
pair c·rews and equipment, and a reasonable supply of spare compon-
ents.
·The results of the analysis provide the cumulative probability of
not exceeding a given number of days of reducE:d energy delivery
capability. Figures 18.2-14 and 18.2-15 display this information
for Anchorage and Fairbanks, respectively. Interpretations are as
follows:
-In the particular case of Anchorage igure 18.2-14), it will first
be noted that the probability scale includes only the extreme
upper.range of non-exceedance probabilities. The intersection of
18-23
·:-~:'!·~
(.e)
--.> -.,._ ~ , ••• •,.·-"~ ' • ".-• .,-.;;
I
th·e d-istribution curves on the pr·obabil ity axis indicates that the :··· ·•.·
probability of no lost energy del ivety capah·ility in a given year ·
is 0.958 and of not having 50 percent reduction of 0.955. Beyond
these points the curves rise sharply, indicating that outages be----~-~.---·,'-,
yond five days are extremely unlikely. The "expected" annual . _. .
value of 0 .. 06961 da days for a total delivery loss may be compared·
with the "loss. of load probabil ity11 of 0.1 (one day in ten years) . -whi~h had ~een used in the gener:ation pla~ning e~forts in the eco-·I•
nom1c stud1es. In short, tne r1sk analys1s conf1rms that the ·
reliability of the transmission system for energy delivery to
Anchorage is consistent with the requirements of_ the overall Rail-1.:
belt generation system. The 11 expected" annual value of 0.09171 · .·
days for a 50-percent reduction in energy delivery a appears to
be similarly acceptable when compared to assumed loss of load 1.
probability.
0
The cumulative probability distribution for-Fairbanks (Figure
18.2-15) has a slightly different intercept on the probability
axis and its shape is also slightly different from those for
Anchorage. These differences stan from the fact that delivery to
Fairbanks requires no suomer'ged crossing and certain other risks
(e.g. flood, temperature extremes) would be expected to have dif-
ferent probab i1 ities for northern and southern segments of the
system. In spite of the absolute differences, it may be s~n from
the display that 11 expected 11 annual value of .08116 does not exceed
the loss of load probability criterion of 0.1 day per year.. No
50-percent loss for Fairbanks is shown since the loss of one of
two lines causes no reduction in delivery capability. Two 1 ines
lost is, of course, a 100-percent loss.
(viii) Emergency Response
I
I
I
I
I
I
I
I
I .:,
In spite of the apparent reliability of the transmission ~ystem, it
is nonetheless true that a small but finite chance of relativ·ely
long-term outages does exist. It is also unfortunately true that
certain extreme risk magnitudes (e.g. combination of extreme loss
temperature 3 wind, and ice) which could lead to an outage also tend
to coincide with high demands by users on the generating system.
The 11 response" in this case is extremely important. The final re-
port for Subtask 11.03, Risk Analysis, provides such a response in
the form of a preliminary emergency plan which includes such meas-
ures as shedding non-essential loads, putting reserve capacity on
1 ine, and energy transfers from military generation systems. Prior
to the time that the Susitna Hydroelectric Project begins operation~
this plan should be updated and occasional tests should be made~
Conclusions
Based upon the risk analysis, it ·is concluded that:
( i) The probabilities that actual costs will not exceed values subjected
to sensitivity tests in the economic anlaysis are are as follows:
'li
18-24
I
I
I
>
:...,•
I
I
'I
I
I
I
I
I
I
I
I
I
··I
I
·,·
I
Value
Project Estimate
Low Capital Cost Tested in the
Economic Analysis
High Capital Cost Tested in the
Economic Analysis
-..
Probability That Value
Will Not Be Exceeded
73 %
47%
90% . .
(i i) Exposure to potential_tosts above the project estimates does exist
and there is about a one 1 percent chance that an overrun of 40
percent or more (in 1982 dollars) will occur.
(iii) The annual probability that no interrupt ion in energy delivery to
major load centers will occur as a result of transmission line
failures is in excess of 95 percent.
Expected values of energy delivery interruptions are less· than one
day in ten years and are consistent with loss of load probabilities.
asstiTled in the generation planning efforts.
( iv) There is a 65-percent probability that the ~~atana rpoject will be
completed pri.or to the scheduled time "in 1993. Exposure to schedule
delays is heavily influenced byregulatoryrequirements and there is
a 10-percent probability that the Watana project will not be
camp 1 eted unt i 1 1995 or later. ·
18.3 -Marketing
This section presents an assessment of the market in the Railbe1t Region for the
energy and capacity of the Susitna development. A t"ange of rates at which this
power caul d be priced is presented together with a proposed basis for contract-
ing for the supply of Susitna energy.
(a) The Railbelt Power System
Susitna capacity and energy will be delivered to the "Railbelt Region
Interconnected System" which will result from the 1 ink age of the Anchor-age
and Fairbanks systems by an intertie to be completed in the mid-1980's.
The Railbelt Region covers the Anchorage-Cook Inlet area, the Fairbanks-
Tenana Valley area, and the Glennallen-Valdez area (Figure 18.14). The
utilities, military installations and universities within this area \\*hich
own electric generating facilities are 1 is ted in Table 18.14. The service
area of these utilities is shown i'n Figure ltL15 and the generating plants
serving the region are 1 isted in Table 18.15.
The Railbelt Region is currently served by nine maJor utility systems; five
are rural electric cooperatives, three are municipally owned and operated,
and one is a federal wholesaler. The relative mix of electric generating
technologies and types of fuel used by the Kailbelt utilities in 1980 are
summarized in Figure 18.16.
18-25
In 1980, the Anchorage ... cook Inlet area had Bl percent, the Fairbank·.)-Tenana
Valley area 17 percent, and the Glennallen-Valdez area 2 percent of the
total energy sales ic the Railbelt Region~
If the recommend at ions of the May 1981 Gi.lbert/Gommonwealth Report are im ...
· plemented" toe Anchorage. and Fairb_anks_Jlmier .. systems will be intertied be-
fore the Susitna project·comes into operation. The propo.sed intertie will
allow a capaci.ty transfer of up to 70 MW in either direction. The proposed
plan of interconnect ion envisages initial operation at 138 kV with subse-
quent uprating ·to 345 kV allowing the line to be integrated into the
Susitna transmission facilities. ..
(b) Regional Electric Power Demand and Supply
A review of the socioeconomic scenarios upon which forecasts of.electric
power demand were oased is presented in Section 5 of this report. The
forecasts adopted here. are the mid-range levels presented by Battelle
Northwest in December 1981.. Subsequent forecasts which introduce price
demand considerations have not been considered at this stage. The results
of studies presented in Section 5 call for Watana to come into operation in
1993 and to deliver a full year's energy generation in 1994. Uevil Canyon
wi 11 come into operation in 2002 and de 1 iver a full year • s energy in 2003.
Energy demand in the Rail belt Region and the deliveries 'from Susitna are
shown in Figure 18.17.
(c) Market and Price for Watana Output in 1994
It has been assumed that Watana energy will be supplied at a single whole-
sale rate on a free market bas is. This requires in effect that Susitna
energy be priced so that it is attractive even to utilities with a low cost
alternative source of energy. On this basis it is estimated that for the
3315 GWh of energy generated by Watana in 1994 to be attractive, a price of
140 mils per kWh in 1994 dollars is required. Justification for this price
is illustrated in Figure 18.18. Note that the assumption is made that the
only capital costs which would be avoided in the early 90s would be due to
new coal-fired generating plants (i.e. the 2 x 200 MW coal-fired Beluga
station).
The financing considerations under which it would be appropriate for Watana
energy to be sold at approximately 14~ mils kWh price are considered in
Section 18.4 of this report; however, it should be noted that some of the
energy which would be displaced by Watana' s 3315 GvJ~ would have been gen ...
erated at a·lower costthan 145 mils, and utilities might wish to delay
accepting it at this price until the escalating cost of natural gas or
other fuels made it more attractive. A number of approdches to the resolu-
tion of this problem can be postulated.
(d) Market Price for Watana Output 1995-2001
After its initial entry into the market in 1994, the price and market for
the 3387 MWh of Watana output is· consistently upheld over the years to 2001
by the projected 2U-percent iricrease in total demand over this period.
18-26
I,
I
I
I
I
I
I
I
I
I
I
I
I
'-1
·•I;
I
I
I
I
I
I
--I
I
'I
I
I
I
,: •.
I
nl .. ..
..
0
•
. . . There would,· as a result, be-a 70~percent increase in cost savtngs compared
wi-th the best thermal altern:ative: these savings per unit ·of -output are_
illu$trated i·n Figure-18.19 ..
(e) Market ·and-Price f.or Watana and Devi 1 -Canyon. Output in __ 2003
A diagramatic analysts of the~ total co·st savings which the combined Watana
and Devil Canyon output wi 11 . confer on the ·system compared with the present
thermal option in the year 2003 is shown in Figure 18.20. Dividing these;
total savings by the energy contributed by Susitna indicates a price of 250
mils per kWh would be the maximum price which .can be charged for Susitna
output. Here again, the problem of competing with lower cost combined
cycle, gas turbines, etc., wil1 have to be addressed; however, this problem
is likely to b.e short term in nature, as by this time period these thermal
power ·faci 1 it ies wi 11 be approaching retirement.
Only about 85 percent of the total Susitna output will be absorbed by the
system in 2002, the balance of the, output being progressively absorbed over
the following decade. This will provide increasing total savings to the
system from Susitna, with no associated increase in costs.
(f) Potentia 1 Impact of S~ate Appropriations
In the preceding paragraphs the maximum price at which Susitna energy could·
be sold has been identified. Sale of the energy at these prices wjll
depend upon the magnitude of any proposed state appropriation designed to
reduce the cost of Susitna energy in the earlier years, At significantly
lower prices it is likely that the total system demand will be higher than
assumed. This, combined with a state appropriation to reduce the energy
cost of Watana energy, would make it correspondingly easier to mar~et the
output from the Susitna development;. however, as the preceding analysis
shows, a viable and strengthening market exists for the energy from the
development even when the output is priced up to the cost of the best
therma 1 a1 tern at i ve. ,
18.4 -Financial Evaluation
(a) Forecast Financial Parameters
The financial, economic, and engineering estimates used in the financial
analysis are summarized in Table 18.16.. The interest rates and far-ec.ast
rates of inflation ·(in the CPR} are of especial importance. They-have been
based on the forecast inflation rates in CPR and forecast interest ~ates on
industrial bonds, as given by Data Resources Incorporated, and conform to a
range of other authoritative forecasts. To allow for the factors which
have brought about a narrow·ing of the differential between tax-exempt and
non-tax-exempt securities, it ·has been assumed that any tax-exempt finan-
cing would be at a rate of 80 percent rather than the historical 75 percent
or so of the non-exemr;:»t interest rate. This identifies the forecast inter-
est rates in the fina:ncing periods from 1985 in successive five year per-
iods as 8.6 percent, ·;.a percent, and7 .. 4 percent .. The accompanying rate
lB-27
{b)
(c)
of. inflation· is around 7 percent.. In view of the uncertainty attaching to
such forecasts and in the interest of consE!rvati.sm, the following financial
projections have been based upon the assumption of a 10-percent rate in-
terest-for tax~exempt bonds r-t;~d an· ongoing inflation t•ate of 7 percent ..
The Inflationary Financing Deficit
··:·.. . .. . . !2'. ~
The basic financing problem of Susitna is the magnitude .of its .. inflation-
ary deficit." Under inflationary conditions these deficits (early year
losses) are an inherent characteristic of almost al 1 debt financed long
1 i-fe, capita] intensive projects (see Figure 18.21). As such, -they are
entirely compatible (as in the Susitna case) with project showing a good
economic. rate of return. Although only a financing characteristic brought
about by the project being h-eavily financed by debt under infl ai:ionary con-
ditions, this characteristic makes it possible for the project to proce~ed
without unacceptable burden of early year costs on consumers.
The Has ic Financial Upt ions
A range of financing options compatible with the conditions laid down in
Senate Bill 25 have been considered as means of meeting the inflationary
financing deficit. The -financial parameters used in these plans are as
given in subsection (a) above .. The options basically consist of a range of
preappropr iations (in 1982} by the State of Alaska with the balance of the
project financing made up by a combination of G.O. bonds and 35 year· Rev-
enue bonds, with G.O. bonds refinanced into Revenue bonds at the earliest
opportunity.,
(i) State ~f Alaska Legislative Appropriation of
100 Percent of Cost ($4.5 billion)
This conforms to the possible outcome legislated by Senate Bill 25
and represents the simplest financing option. It could take the
form of the state meeting capital costs as incurred over the 15 year
schedule.. Alternatively, it could take the fonn of 11 preappropri-
at ion 11 where such a sum would be appropriated in; say 1982, as tak-
ing into account interest accumulated, would totally finance the
.pro-ject. For simplicity of interpretation of the options involving
state appropriation all are assumed to take the "preappropriation11
form.
A preappropriation of $4.5 biilion in 1982 would wholly complete
Susitna (on the basis of centr.al estimates). On the basis of the
present wholesale energy rate sett~ng requirement incorporated in
Senate Bill 25, the APA would, h~wever, not be able to charge more
than the actual costs incurred. Given that in this case the only
costs would be the very small year-to-year operation costs, this op-
t ion waul d invo 1 ve the output from Sus itna being supplied only at a
fraction of the price of electricity from the best thermal option.
18-28
·.' ·.-.··"'1!'·.-·.·j ... ·. < '·_.)~~~~
; . . -. ·~"<'
.• ,1
I~
I
I
I
:.~-·. 1 .. ·~·
I
I
I~
I
I
I'
I
I
I.
I
.~
I ·. I< _)
1·-
1>·
. .-· ~:: \".
-.(. , • .:..:-o;r.
I
·I
·':1:
. -·-·
I
I
I
I
I
~.1
I
~I
,I
••
··.1
I
I
·J
I
:-I
(d)
(ij) 50 Percent State Pr~~4;ppropriqtion {$2~5 bill ion)
wit~· Residual Bon~ F1nancimr
-Tbe;Foutcome fo.r this opt ion-is summarized tn terms of figure 18-22 ..
It is· seen that it would still enab1e Susitrra energy to be produced
at a price 37 percent less than that of the best thermal option.. It
would also enable the project to be completed with on·ly $1 billion
(in l982·dollars) of G.O~ bonds (see below) over the period 1990-93.
The Devil Canyon stage could then be completed with a further $2.4
billion (in 1982 dollars) of Revenue bonds over_the period 1994 to
2002 ..
This level of appropriation would enable Susitna energy prices to be
held virtually constant at their initial levels for nearly a decade.
A temporary step-up in price to the cost of the electricity from the
best thermal opt ian would be required when Devil Canyon was com-
pleted an the basis of its lQQ..;.percent Revenue bond financing ..
Thereafter, however, the cost of the Sus itna energy would again
stabil jze and give ever increasing savings compare.,d with cost of the
best thermal opt ion.
(iii) 11Minimum .. State Preappropriation {$2 bill ion)
with Residual Bond Financing
The "minimum" State appropriation is taken as the minimum amount re-
quired to meet debt service cover of L.25 an the residual debt ser-·
v-ice cover of 1. 25 on the residual debt financing by Revenue bonds
and make Susitna's wholesale· energy price competitive with the best
thermal option in its first normal cost year (1994). This level of
appropriation would require $1 .. 8 billion (in 1982 dollars) of bond
financing 1990-93 and a further $2.2 billion (in 1982 dollars) over
the period 1994 to 2002 to camp 1 ete Oev i 1 Canyon. (Figure 18.,23)-
These levels of State appropriation would all therefore eliminate
Susitna's 11 inflationary financing deficit."
. .
Issues Arising fromthe Basic -Financing Options
( i) Tax-Exempt Bond Financing
In the $2 bill ion State apprapri at ion case interest costs an the
basis of tax-exempt financing accounts for 88 percent of the unit
price of Susitna output in 1994. Failure to obtain tax-exempt bond
financing would increase these interest costs by approximately one
third. Ensuring tax-exempt status far the Susitna bond issues is
therefore of fundamental importance to the economics of the project
under these apt ions.
Difficulties could arise in obtaining tax-exempt bond financing if
the financing entailed (as would probably be the case with Revenue
bond financing at the precompletion stage) contracts of the take-
or-pay or take--Jnd-pay type.. This is because the bulk of the
18-29
~
Sus itna output wou1 <:1 be taken by non-tax ... exempt uti 1 ities and
contracts of this type_ with non ... tax-exempt entities would, under
certain general ·conditions laid down in Section 103 o.f the IRS code~
lead to the ·band issues being classified as i"ndustrial development
funds and foregoing their tax:-~x:empt status" lt is also
questionable whether contracts with the Railbelt ut·nities as
current1y financed(·wou1d canst itute adequate security in the eyes of
bond hold.ers. Both these considerati.ons indicate the need for soma
fornr of ·;nd·ep-endent financing guarantee reducing dependence on the
contractual relationship with the utilities., ·
This might take the form of the initial financing being G~O. bonds
or by a State guarantee being given to the Revenue bonds. Given
that either represents the same .effective burden to the State, it is
cone 1 uded . that G. 0. bonds are to be preferred on grounds of flex i-
bil ity and administrative simplicity.
(ii) Refinancing Watana and the Financing of bevil Canyon
,. Earl-y refinancing of any G.O. bonds used to finance Watana and the
financing of Devil Canyon by Revenue bonds is taken to be an impor-
tant financing objective. The main factors determining the date at
which such refinancing will be possible ts the magnitude of the ini-
tial state appropriation.. This is dealt with in detail terms of the
risk analysis in 18 .. 5 below.
The basic conclusion from this analysis is that with a state appro-
priation of $2.5 billion there is a very high degree of certainty
that refinancing into Revenue bonds waul d occur by 1994 and that the
r·ema·inder of the project could be financed by Revenue bonds.
(iii) Importance of Adequate Preappropriation
Funding to Subsequent Financing
The principal effect of preappropriations significantly less than
$2 bill ion would be a possible delay in refinancing of the G.O.
bonds issued to finance Watana and possibly the need for additional
G .. O .. bon.d financing for Devil Canyon. This is because the impact of
such lesser preappropriation would (as illustrated in Figure 18 .. 24)
give r·ise to inadequate earnings covered in the early years of
Watana and subsequently Devil Canyon so that the raising of Revenue
.Jonds requiring such cover had to be delayed. In addition, such
inadequate funding waul d force the Susitna price to 11 track11 the cost
of energy from the best thermal opt ion until adequate revenue had
been obtained for such refinancing.
(iv} Impas.t on State Credit Rating of Susitna G.O .. Bond Financing
The impact on state credit rafing of G,O. bond financing of the
order of $1..8 bill iorl (in 1982 dollars) in the $2 bill ion state ap-
propriation case has been assessed by the APAs financial. advisors•,
First Boston Corporation and First Southwest Corporation. They have
I~
I
I
I
I
I
I
I
I
••
I
I
I
I
')
I
I
.I"
I
1··.
. ;";
I
I
'I
I
I
I
I
I
cl
~~, .•
I
I
I
I
. rJ/ (,
. .
••
II
'I
i~ :·· i ;:;. ' . ~~~
concurred (as fully stated in sub-s'ection (c), (i'f',) of the main
report) in:· the statement that uon the assumption that the State of
JUaska•s bond· rating at that time is unchanged from today's level
and that norma_lity prevails in the bond market, the effect on the
credit rating of the Statr; of Alaska would not be perceptible."
(e) Conclusion
. The principal conclusion of the financial evaluation is that with a state
appropriation of not less than $2 billion and consent for G.O~ bond
financing of $1.8 billion (in 1982 dollars), Susitna would be financially
viable. It would also be able to market its output at an initial price
competitive with the most efficient thermal options and produce very
substantial long-term savings compared with this option.
The evaluation, however, stressed the importance of establishing the pro~
ject on a strong financial basis that wou1d enable it to secure conversion
of the :G.O. bonds to Revenue bonds in 1994 and obtain a highly competitive
rate of interest without jeopardizing the tax-exempt status of the bond
issues. These objec.tives (together with the marketing of the Watana output
in 1994 and price some 37 percent bel ow that of· the most efficient thermal
option), .cou1d be secured by state appropriation of $2.5 bill i-on-~---
Methods by which the state appropriation could be recovered have not been
considered since recovery is not required by existing legislation. It
should be noted, however, that the cost benefit analysis shows that full
recovery long-term would be possible with a better than 10-percent rate of
return. Meeting the Susitna inflationary financing deficit can therefore
be considered as a separate issue from subsidization of electricity prices
by foregoing recovery of all or part of the State appropriation designed to
meet this deficit. ·
18.5 -Financial Risk
The financial risks considered are those arising to the State of Alaska and
to Alaskan consl.lllers. The analysis of these risks is restricted to the
period up to 2001 covering the completion of Watana and its first nine
years of operation.
(a) Precompletion Risk
The major precompletion risk is the risk that the project will not be com-
pleted. The possibility of this arising owing to natural hazard is dealt
with in Section , and on the basis of this analysis this possibility
is assigned a vanishingly small probability.
The risk of non-completion owing to capital overrun is also assessed to
have negligible probability. This is on the grounds that the project only
·involves well established techno_logy, has been extensively assessed and
surveyed, and has been assessed independently by estimators and formal
probabi1 ity analysis as having only a 27-to 20-percent probaoil ity of any
real capital overrun.
18-31
',.,;,,:
'...,.,_.,.../
I
J I
I
(b) Post..:comp1e~ion R is~~.·
( i} The l1ener·ation of Post ... comp-letion Risk~
A probabilistic financial model was developed taking into account
the probability distri_butions of the major engineering and financial
variables on which the financial outcome for Susitna depends. This
mode1 was then used to consider in detail critical specific risks
and the aggregative risk posed by the project.
(ii) Specific Risks
-Specific Risk I; Risk of G.O. Bond
Financing Overrun (Figure 18.25)
Extensive analysis was undertaken to assess the probability that
the G.O. bond financing requirements would overrun the forecast
values as a result of capital costs, inflation, interest rates~
etc., being less favorable than forecast. In the $2.5 billion
state appropriation case it was found that the probabi1 ity of the
G.OQ bond requirement exceeding the forecast of $1 bill ion {in
1982 dollars) by more than 50 percent was only 0.15. This implies
that there is less than one in six chance of the G.O .. bond overrun
exceeding $1.5 billion. The probability of its exceeding $2 bil-
l ion was only 0.03. There is also a significant probability that
the bond financing requirements will b~ less than forecast.
-Specific Risk II; Delayed Conversion of G.O. Bonds
Minimization of the magnitude and duration of G.O. bond require-
ments is taken as an important financial objective. Evaluation of
this specific risk in the $2.5 bill ion appropriation case indi-
cates that:
. The probability of any delay compared with the forecast date of
1994 for the conversion from G.O. bonds to Revenue bonds is
0.05.
. The 1 atest date at which complete conversion to Revenue bonds
occurs in any outcome is 1996 (3 years after completjon of
Watana}.
.. Specific Risk III: Early Year Non-viability {Figurel8 .. 26)
The measure of financi a1 non-vi ab i1 ity in the early years is taken
as the ratio of Wat an a • s unit cost to the costs of the best ther-
mal option in Watana•s third year (1996). (For comparability debt
service excess cover was excluded). This· analysis indicates that
there is only a 0.15 chance of the Susitna costs exceeding their
forecast value (30 percent of the best thermal) by more than 15
percentage points. ·
18-32
I
I
I
I
I
I
••
I
I
I
I
I
I
I
I
I
I··
••••• ' ':
el
I
I
I
I
'I
I
I
I
I
I
I
I
I
I
I
..
• ••••
I
-I
f iii)--I_he ·Aggregate __ Risk
While specific risks of the type considered above are of importance
basic concern must center on the aggregate risk. In long-term eco-
nomics this is measured by the risk attaching to the rate of return.
· For the purpose of the financial risk, however, it is taken as rep-
resented by accumul at'lve net operating earnings at the end of the
first nine years of operation of Watana. Since this statistic is.
net of interest and debt repa}ment, if effectively subsumes all the·
risks involved in capital expenditure, inf1 at ion, interest rates,
revenue, etc .. , deviating from their forecast values. This statistic
was also adjusted to allow the pricing up of Watana energy to the
cost of the best thermal option so that statistic affects the
11 UpSideH risk as well. as the 11 d0W0Side.11
On this basis the statistic (see Figure 18~27) was found to have
only a 0.15 chance.of being below forecast level of $1.35 billion
(in 1982 dollars) by more than $.35 billion. There is also a 0~34
probability of the statistic exceeding $1.5 billion and thus
creating greater savings for the Alaskan consu.mer.
(c) Conclusions
The analysis shows the exposure of the project either to critical s-pecific
risks or to aggregative risk is relatively limited. The qualifi.cation
attaching to this analysis is that the estimates and probabilities used are
free from any systematic biases~ The structure of the plan of the overall
plan of study for Susitna and analysis of its alternatives has however been
specifically desi9.·.~d to take every reasonable precaution against this pos-
sibility by seeking extensive independent verification of the key variables
by Batelle and Ebasco operating wholly as independent consultants.
18-33
. LIST OF. REFERENCES
-
(1)·· U.s. Department of Labor, Mont~lY, Labor ~ReviJW, various issues .•
(2) Alaska Department .of Commerce and Economic Development, The Alaska Economic
Information and Reporting System, Ju 1y 1980.
(3} Data Resources Inc., U.S~ Long-Term Review, "Fall .1980, Lexington, ,Mass.,
1980.
(4) Wharton Econometric Forecasting Associates, Fall 1981, Philadelphia, Pa.
(reported in Economic Council of Canada CANDIDE Model 2-0 Run, dated
· December 18, 1981.}
(5) Baumel, W.J., "On the Social Rate of Discount", American Economic Review,
Vol. s·a, September 1968.
(6) Mishan, E.J., Cost-Benefit Analysis, George Allen and Unwin, London, 1975.
(7) Prest, A.R. and R. Turvey, 11 Cost-Benefit Analysts: A Survey 11
, Economic
Journal, Vo 1. 75, 1965. ·
( 8) U.S. Department of Commerce, Survey of Curre!lt Business, various issues.
(9) Data Resources, Inc., personal comnunication, .November 1981.
(10) World Bank, personal communication, January 1981.
(11) u.s. Department of Energy, Energy Information Administration, Annual Report
to Cong~ess, Washington, D~C.
(12) National Energy Board of Canada, Ottawa, Canada, personal communication;,
October 1.981.
{13) Noroil, 11 Natural Gas and International LNG Trade 11
, Vol. 9, October 1981.
(14) Segal, J. 11 Slower Growth for the 1980•su, Petroleum Economists 80 December
1980.
(15) Segal, J. and F. Niering, 11 Special Report on World Natural Gas Pricinga•,
Petroleum Economist, September 1980.
(16) SRI International, personal communication, October 1981.
(17) World Bank, CoiTUllodity Trade and Price Trends, Washington 1980.
(18) Battelle Pacific Northwest Laboratories, Beluga Coal Market Study, Final
Report, Richland, Washington, 1980.
{19) B.C. Business, August 1981.
(20) Coal Week International, various. issues.
I
I
~--~,~-----···.,_
I
••
I
I
I
I
I
I
: •...
I
I
I'.
I
I
:··\'
I
.,· . '
I
I
I
. ';::•
I
I
I
"1.
I
I
I
I
I
I
,I
.I
~.~
,~.: :1
.· .
·t
·LIST. 0~ REFERENCES (Continued)
{21) Japanese Ministry of International.Trade and Industry, persona;1
communication, January 1982. ·
(22) Canadian Resourcecon Limited, Industrial Thermal Coal Use in Canada, 1980
to 2010, May 1980.
{23) Battelle Pacific Northwest Laboratories, Alaska Coal Future Availability
(24)
and Price Forecast~ May 1981. · ·
Roberts, J.Q. et al, .Ireatment of Inflation in the Development of Discount
Rates and Levelized Costs in NEPA Analyses for the Electric Utility
Industry, u.s. Nuclear Regulatory Commission, Washington, D. c., ·
January 1980. ·
: ''l . . ..
TABLE. 18.1: REAL (lNF'LATION-AOJUSTE.D) ANNUAL
.. JlROWUf·lN OIL PRICES
Low Case
Hsdi.un (most likely case)
High Casa
Base Period
(Jam~ary 1982)
Growth Rates (Percent)
1982-2000 --
0
2.0
4.0
2000-2040
0
1.0
2.0
Pric:e of No. 2 Fuel Oil -$6 .. 50/MMBtu.
. Probability
O.J
0.5
0.2
;
,, ···,,.~ -~· ;· ·,. ~~~ .. :~
. -:::
I .,
I
I
I
I
I
I
I
••
I
I
I
I
..•. ·.<
( . -~
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I.
I
TABLE t810Z: DOMESTIC MARKET PRiCES AND EXPORT
OPPORTUNITY VALUES:"OfcNil.lURAL.GAS
Domestic Market Price1 Exeort Oe§ortunit~ Value
Cow Rea1um Hl.9F\ Cow Me 10m igh ·
P-robability of
Occurr&nce N.A. N .. A. N.A. 2.7%
Base Period Value $3.00/MHBtu
Real Escalat~on ClF
Price, Japan
'1982 -2000 N.A. ~
2000 -2040 i)' G%
Real Escalation
Alaska Price
1982 -2000 0% 2.5% 5.0% 0%
2000 -2040 0% 2.0% 2.0% 0%
1 OGP analysis used domestic market prices with zero
escalation beyond 2010. (Source: Battelle)
46%
$4.65/MMBtu2
2%
1%
2.7%
1.2~
2 Based on CIF ptce in Japan ($6.75) iess estimated cost
of liqllefaction and shipping ($2.10). (Source: 13, 14, 15).
3 Source: (9), (16).
4 Alaska opportunity value escalates more rapidly than .
CIF prices as liquefaction and shipping costs are estimated
to remain constant in real terms.
27%
4%
2%
5&2%
2.2%
<1-
tABlE' 1S.Jt._ SUMMARY OF .COAL OPPORTUNlTY_VALUES
-"
Base :Perio~, Annual Real Grwoth Rate
(Jan.-. 19.82}
Value 1980 -2000 ' (%) ' 2000 .. 2040
( $/f+tBt u) . . (%)
Base Case
Battelle Base
Period ·crF Price
Medium Scem .. rio
-CIF Japan 1.95 2.0 1 .. 0
-fOB Beluga 1.43 2.6 1.2
-Nenana 1 .. 75 2.3 1.1
Low Scenario
-Clf Japan 1.95 0 0
· -FOB Beluga 1 .. 43 0 0
-Nenana 1.75 0.1 0.1
High Scenario
-CIF Jspan 1.95 4.0 2.0
-fOB Beluga 1.43 5.0 2.2
-Nenana 1.75, 4.5 1. 9
Sensitivit~ Case
Updated Base ·
Period CIF Price1
Medium Scenario
-CIF Japan 2.66 2.0 1.0
... FOB Beluga 2.08 2.5 1.2
-FOB Nenana 1. 74 2.7 1.2
Low Scenario
-CIF Japan 2.66 0 0
-FOB Beluga 2.08 0 0
-FOB Nenana 1. 74 -0.2 -0.1
High Scenario
_, Clf' Japan 2.66 4.0 2.0
. -FOB Beluga 2.08 4.8 2.2
-FOB Nenana 1.74 5.3 2.3
1 Assuming a 10 percent discount for Alaskan coal due to
quality differentials, and export potential for Heal coal.
0
Probability
of
Occurrence
49
49
49
24
24
24
27
2.7
27
49
49
49
24
24
24
27
27
27
' '
I
I
I
' ' · ..•... _
I
I
I
I
I
I
I
••
I
I
I
I
-1· '_a
. '
:12 _· ..
"'
I
I
I
I
I
I
I
••
I
I
.I
I
I
I
I
I
I
I
TABLE 1S.4: . Sl.ttMARV Of F'UEL PRICES USED IN THE ,
OGP PROBABILITY TREE ANAL VSIS a
Fuel ~rice Scenario
Probability of occurrence
Base period Jan1,1ary '1982 prices
(1982$/MMBtu)
fuel Oil
Natural Gas
Coal
-Beluga
-Nenana
Real esca.lationrates per year
(percent)
Fuel Oil
-1982 -2000
-2000 -2040
Natural Gas
-1982 -2000
-2000 -2040
Beluga Coal
-1982 -2000
-2000 -2040
Nenana Coal
-1982 ... 2000
-2000 -2040
Low
25%
6.50
).00
1.43
1.75
0
0
0
0
0
0
-0.1
0.1
1 Beyond 2010, the OGP analysis has.used zero
real escalation in all cases.
,_
Medium
50%
6.50
3.00
1.43
1.75
2.0
2.0
2.5
.2.0
2. 6
1.2
2. 3
1.1
25%
6.50
3.00
1.43
1. 75
4.0
2.0
5.0
2.0
s.o
2.2
4.5
1.9
.-"'--·
.. .
J;f
..
TABtf 18,.5:· tCONOHIC ANALYSIS
Plan 10
Non Susitna A
Susit~'l c·
Net Economic Benefit
of Susit.na Plan
_ SUSllNA PROJECr -BASE PLAN
1982 Present Worth gr .System Costs
. $ X 10 · .
199:3-Estimated 1993-
Components 2010 2010 2011.;.2051 2051 -
600 MW Coal-Beluga 3,213 491 5,025 8,238
200 MW Coal-Nenana
630 MW GT
680 MW Watana 3,119 385 3,943 7,062
600 MW Devil Canyon
180 MW GT
1,716
•
I -·" ·-'
"·
1•···.· .. · ... ' .
.,
···:···!; .. . f: .. ·.
; . . .
I
I
I
I
I
I
I
I
I
-··· .·_ ..
I
:1-..
. .
.. 1· .
. _,. \_:_
'·
I
:t":':'?'' ~~T·--•,.,. ,.
-.•...• ·-,
'•
: .. 1·----c:· . ..
. ·· '·
I
I
I
I
I
I
-·
I ;'" ,,
I
I
I
I
I
'I ,,
~1i\: <·: -
.< ~ .. ·. ;:: :
'..!.. .
MW
1990 892
2000 1,084
2010 1,537
---
TABLE 18.6: SUMMARY OF LOAD FORECASTS
USED fOR SENSITIVITY ANALYSIS
Medium Low High
GWh MW GWh l~W -
4,456 802 3,999 1,098
5,469 921 4,641 1,439
7,791 1,245 6,303 2,165
-::-
GWh
5,703
7,457
11,435
TABLE 18.7: LOAD fORECAST SENSUIVUY ANALYSIS
1982 Present Worth of S~stem Costs ($ x 106)
Net
1993-Estimated 1993-Economic
Plan 10 Components 2010 2010 2011-2051 2051 Benefit
Non-Susitna K 1. 1400 MW Coal-Beluga 2,640 404 4,238 6,878
Low forecast 200 HW Coal-Nenana
560 MW GT
Susitna K2 680 MW Watana (1995) 2,882 360 3,768 6,650 228
Low forecast 600 MW Devil Canyon (2004)
Non-Susitna J1 800 HW Coal-Beluga 4,176 700 6,683 10,8591 1
High forecast 200 HW Coal-Nenana
700 MW GT
430 MW Pre-1993
Susitna J2 680 HW Watana (1993) 3,867 564 5,380 9,2471 1 1,612
High forecast 600 MW De~il Canyon (1997)
350 MW GT
430 HW Pre-1993
1 from 1993 to 2040
!,._--;-.·--~'-~ ..... -.;.:. _ _,' ·~ ,,.. ~ ......... __ ._ ~~-~-.. ""' -·-•• -· -·-··-. -J -· -
-------------------
TABLE 18.8: DISCOUNT RATE SENSITIVITY ANALYSIS
1982 Pr-esent Worth.of S~stem Costs($ Xc106)
Real Net
Discount Rate 1993-Estimated 1993-Economic
Plan ID (Percent) 2010 2010 2011-2051 2051 Benefit
Non-Susitna Q1 2 3, 701 465 7,766 11,167
Susitna Qz 2 3,156 32J 5,394 8,550 2,617
Non-Susitna A 3 3,213 491 5,025 0·,328
Susitna c 3 3,119 385 3,943 7,062 1,176
Non-Susitna 51 4 2,791 517 3,444 6,235
Susitna s2 4 3,080 457 3,046 6,126 109
Non-Susitna p1 5 2~468 550 2,478 -!·~946
Susitna Pz 5 3,032 539 2,426 5,459 (513)
TABLE 18.9: CAPITAL COST SENSITIVITY ANALYSIS
1982 Present_Worth of S~stem Fasts-$ x
1993-. Estimated 1993-
Plan ID 2010 2010 2011 ... 2051 2051
Non-Susitna. Capital
Costs Ue 20 Percent
Non-Susitna G 3,460 528 5,398 8,858
Susitna c1 3~119 385 3,943 7,062
Non-Susitna Capital
Costs Down 10 Percent
Non-Susitna G 3,084 472 4,831 7,915
Susitna c1 3,119 385 3,943 7,062
Susitna Capital Costs
Less Contingenc~
Non-Susitna A 3,213 491 5,025 8,238
Susitna x2 2,710 336 3,441 6,151
Susitna Capital Costs
_ Plus __ Doubled Contingen~
Non-Susitna A 3,213 491 5,025 8,238
Susitna v2 3,529 434 4,445 7,974
1 An adjustment calculation was made regarding the + capital costs of
the 3GT 6nits added in 2007-2010 since the difference ~as less than
$10 x 10 • Beyond 2010, this effect was not included.
---------~___,....,.,~
106
Net
Economic
Benefit
1,976
853
2,087
264
-a_--__
--
I
I , ...•
I
••
I
I
I
I
I
I
I
I
I
••
I --'
I
.•. -. ., --
-; .>:'
,:.::/
I
I
~·
I
I
I
I
I
I
I
I
I
I
I
I
•
~=I
'I,
~,-J . ' .... · ' ·-""" . ' ' ' <'
Base Case
Sensitivity
TABLE 18.10: SENSITIVITY ANALYSIS -UPDATED BASE PLAN
(JANUARY 1982) COAL PRICES
Base
Period Beluga
Coal Price
(1982 $/MMBtu)
1.43
PW in 1982 ($ X 106) ------
Costs of
Non-Susitna
Plan
8,238
Costs of
Susitna
Plan
7,062
Net
Economic
Benefits
1,176
(Updated) Case 2.08
0
TABLE 18.12: S~NSITlVITY. ANALYSIS -REAL COST ESCALATION
1982 Present. Worth gf System Coat-s··
· ($ X 10 )
·1993-Est1mated 1993-Net
Plan ID 2010 2010 20.11-2051 2051 Benent --
Zero-Escalation in
Caeitsl ~nd O&M Costs
• Non,...Sus1tna 01 2,838 422 4,319 7,157
• Susitna o2 2,525 299 3,060 5,585 1,572
Double Escalation
Capit~ and O&M Costs
. Non-Susitna p1 3,650 602 6,161 9,811
• Susitna Pz 3,881 503 5,148 9,029 782
Zero-Escalation
in Fuel Prices
• Non-Susitna v, 2,233 335 3,427 5,660
• Susitna v2 3,002 365 3,736 6,738 (1,078)
High Escalation
ln Fuel Prices
• Nan-Susitna w.t 4,063 643 6,574 10,367
• Su~Htna w2 3,267 403 4,121 7,)88 2,979
TABLE 18. 12 {a): SENSITIVITY ANALYSIS -NON-SUSITNA
PLAN WITH CHACKACHAMNA
Plan
• Non-Susitna with
Chackachamna
·• Susitna
1993
1982 Present Worth of System CQsts
($ X 10b)
Estimated 1993-Net
ID Components 2010 2010 2011-2051 2051 Benefit
c
--
B 330 MW Chackachamna 2,038 475
400 MW Coal-Beluga
200 MW Coal-Nenana
440 MW GT
C 680 MW Watana 3,119 385
600 MW Dev1l Canyon
180 MW GT
4,861 7,899
3,943 7,062 837
I
, •..
. ',::,_,
·-.;
••• ••
I • •
'I
I
I
I
I
I
I
I
I
I
I
I
I'
I
I
I
.I
I
I
••
I
I
I
I
I
I
I
•••••••••
'
TABLE 18.13: SUMMARY OF SENSHIVITY ANALYSIS INDEXES
Of NET ECONOMIC BENEt: ITS . BASt CASE .
($1, 176 MILLION} = 100
Fuel Escalation
-High
-Low
Discount Rates
-High-High (5%)
-High (4%)
-Low (2%)
Susitna Cap1tal Cost
-High
-Low
Load Forecast
-High
-Low
Non-Sus~tna (Thermal)
Capital Costs
-High
-Low
Capital and O&M
Cost Es~alat ion
-H1gh
-Low
Chackachamna (included 1n
Non-Susitna Plan)
Updated Base Coal Price
-44
9
223
23
178
137
19
168
73
67
134
71
0
t High fuel escalation case provides net benef1ts equal to 253 percent of the
base value, 2.53 .x 1, 176, or 2,975.
2 Low fuel escalat1on case provides minus 92 pe·rcent of the base case net
benefits, -.92 x 1, 176, .or -1,082.
'I·
---·-
.
Generating Purchases Utility Annu.Ait
Capacity 1981 Predominmt Tax Status Wholesale .Provides Energy Dem~Uldf
MWat0°F Type of Re: IRS Electrical Wholesale 1.980
UTILITY Rating Generation Section 103 Energy Supply GWh
-·'
'
IN ANCHORAGE-COOK INLET AREA
Anchorage Municipal Light and Power 221.6 SCCT Exempt • -585.8 '·
i
Chugach Electric Association 395.1 SCCT t~on-Exempt • • 941.3
Matanuska Electric Association 0.9 Die~tl Non-Exempt • -268.0 j'
Homer Electric Association 2.6 Diesel Non-Exempt • 284.8 i -\
Seward Electric System 5.5 Diesel Non-Exempt • 26.4 1 -r
Alaska Power Administration 30.0 Hydro Non-Exempt -• -l
National Defense 58.8 ST Non-Exempt I ---k
Industrial -Kenai 25.0 SCCT Non-Exempt ---!
i
• IN FAIRBANKS-TANANA VALLEY ' ~
Fairbanks Municipal Utility Syatem1 68.6 ST/DieseJ Exempt -· -116.'1
Golden Valley Electric Aasociatcon 1 221.6 SCCT/Diesel Non-Exempt --316.7
University of Alaska 18.6 ST Hon-Exompt --·-;
National ~ef&nsa1 46.5 ST Non-Exempt ---.
IN GLENALLENNALDEZ AREA
Copper Villey Electric Association 19.6 SCCT N6n:-Exempt --37.4
TOTAL 1114.3 2577.1
1 Pooling Arrangements in Force
TABLE 18.14 _ RAlLBE~T UTILITIES PROVIDif~G MARKET POTENTIAL I ~~~m I · ..... •'
..
I
I
,I
I
I
I
I
'I
••
I
I
I
I
I·
I·
I
P~ANT
No.
2
3
6
7
10
2?
~·
32
34
35
36
37
38
47
55
58
59
75
80
81
82
83
84
PLANT LIST
TYPE OF
NAME OF PLANT UTILITY OWNERSHIP
Anchorage .No. 1 Anchorage Municipal Light and Power Mun3cipal
Anchorage Anchorage Municipal Light and Power Munici.,af
Eklutna Alaska Power Administration. Federal
Chen a Fairbanks Municipal Utilities System Municipal
KnikArm Chugach Electric Association, Inc. Cooperative
Elm•ndorf-West United States Air Force Federal
Fairbanks Golden Valley Electric Association, Jne. Cooperative
Cooper a..ke Chugach Electric Association, Inc. Cooperat:ve
Eimendorf-East United States Air Force Fideraf
Ft Richardson United States Arrny Federal
Ft. Wainright United States Air force Federal
Eielson United States Air Force ·Fedtral
Ft. Gnttlty United States Army Federal
Bemice Lake Chugach Electric Association, Inc. Cooperative
International Station Chugach Electric Association, Inc. Cooperative
Healy Golden Valley Electric Aaociation, Inc.. Coopemive
Beluga Chugach Electric Association, Inc. Cooperative
Cle~rAFB United States Air Force Federal
Collitr·Kenai Collier-Kenai Municipal .
Eyak Cordova Public Utilities Municipal
North Pole Golden Valley Electric Association, .Inc. Coopenmve
Valdez Golden Valley Electric Association, Inc. Cooperative
Glennallen Golden Valley Electric Association, Inc. Cooperativ-e
TA81.E 18.15 -LIST OF GENERATING Pl.ANT SUPPL!NG RAI LBE1.T REGION ~~~~ l
I
I
I ., ....
. '
I
I
I
I
I
·I
I
•••
I
I
I
I
I
I
TABLE 18.16: ESTIMATED FINANCIAL PARAMETERS
Project Complet~on -Year
Energy Level -1993
.. 2002
-2010
Costs ~n January 1982 Dollars
Capital Costs
Operating Costs -per
annun
Prov~s~on for Capital
Renewals -per ar.num
(0.3 percent of Cap1tal Costs)
Operat1ng Working Capital
Wstana -
1993
$ 3.647
bl.llion
$10..0
million
$10.94
million
-1'5 percent of Operating Costs
-10 percent of Revenue
Reserved Contlngency Fund
Devil·
Canyon
2002
$1.470
bilhon
$5.42
million
$4.41
million
-100 percent of Operat~ng Costs
-.100 percent of Provision for Capital Renewals
Interest Rate -10 percent per annum
Debt Repayment Period -35 years
Inflation Rate - 7 percent per annum
Real Increase in Operat1ng Costs
-1982 to 1987 -1.7 percent per annum
-1988 on -2.G percent per annum
Real Increase in Capital Costs
-1982 to 1985 ... 1.1 percent per annum
-1986 to 1992 . -1.0 percent per annum
-1993 on -2.0 percent per annum
Total
J :387 GW·h
5 721 If
6 616 "
$ 5.117
bllhon
$15.42
million
$15.35
million
• - ----..... -J - --; - - --· ... :a a &
I.
REVIEW
_ BASE COST
.--.... _...,..AND
~TART
--
SCHEDULE
ESTIMATE
i
i
I
SUMMARY
NOTES
.____, I
I _,
n.
RISK
LlST
DEVELOPMENT
RISK
LISTS
I
I
Ill. VI.
.,.. METHODOLOGY _;_-._ SOFTWARE
-REVIEW
--
v.
TRANSFORMATION
ASSESSMENTS
IV.
RISK
ASSESSMENTS
I
'
ASSESSMENT
DOCUMENTS
-REVISIONS
VII.
CONSEQUENCE
/RESPONSE
CRITERION
ASSESSMENTS
I
I
DOCUMENTATION
.x.
INITIAL
COMPUTATION
AND
INTERPRETATION
VUI.
REVIEW
AND
REVISE
RISK ANALYSIS STUDY METHODOLOGY
..
)(.
TRANSMISSION
SYSTEM
AND
EMERGENCY
GENERATION
X f.
UPDATE
...... AND -FEEDBACK
XII.
FINAL
COMPUTATION
AND
INTERPRETATION ;
~
RISK
ANALYSIS
REPORT
FIGURE 18.1
I
1" ·-~ ~
I
I
'1
I
I
I
I
'.
~I
I
:I
I
1.
I
I
I
I
QUESTION: -'
WHAT MAJOR CONSTRUCTION
PROJECTS ARE INVOLVED?
WHAT KIND OF WORK IS GOING ON
FOR A GIVEN . CONFIGURATION?
WHAT ARE THE POSSIBLE INITlATING
MECHANISMS WHICH COULD -INFLUENCE
ESTIMATED COSTS OR COMPLETION TIMES'?
.
WHAT MAJOR PORTIONS OF ANY GIVEN
CONFIGURATION ARE SUBJECT TO
RISK REALIZATION?
IF A PARTICULAR RISK MAGNITUDE IS
REALIZED, WHAT POSSIBLE
CONSEQUENCES CAN OCCUR ?
HOW CAN THESE CONSEQUENCES
BE MEASURED ?
WHAT IMPORTANT ASSUMPTIONS AND
.Ll MITATIONS MUST BE ESTABLISHED
TO PERMIT A REASONABLE ANALYSIS
AND TO DRAW IMPORTANT
CONCLUSIONS 7
ELEMENTS OF THE RISK ANALYSIS
-~----.... -.--:-~-~-
RISKS
FIGURE 18~2
•• ........... ~.,.
el.
I
·····;:
t
;~~-
1·
I
I
I
••
I.
I.
I
I
I.
I
I
~,
. ·;:
;
PROBABtLlTV
OF A,
PARTICULAR
RISK
MAGNITUDE
t ®
PROBABl LITY
OF A
PARTICULAR
DAMAGE LEVEL
IF A PARTICULAR
RISK
MAGNITUDE
lS REALIZED
t
PROBABILITY
OFA
PARTICULAR
CRlTERION
RISK
NONE LIGHT MODERATE MAJOR CATASTROPHIC
MI.N MODE
;)lb
INCREASING CRITERION
VALVE
MAX
CD A SERlE$ OF OlSCREJE .. .
RISK PROBASlLlTY LEVELS
EX1STS FOR EACH ~lSI( ..
ACTJ VlfY . COMSlNATJON~--
THE ANNUAL PROSABJLlTY.·
O_F. EACH IS OETERMIN.EO.
@IF A RlSK EVENT OCCURS,
· IT CAN CAUSE A NUMBER
OF POSSIBLE DAMAGE .
LEVELS, EACH WITH A
PARTICULAR PROBABl LITY
OF OCCURENCE. lF
RlSK MAGNITUDE @
c;>CCURS, THE PROBABILITY
IT WILL CAUSE MODERATE
DAMAGE lS THE VALUE OF ® ON THE DIAGRAM.
@FOR ANY GIVEN DAMAGE
LEVEL, THREE CRITERION
VALVE:S ARE ESTlMATED
AND FIT TO A MODlfTED
BETA OISTRl BUTtON~
STRUCTU9AL RELATIONSHIP FOR HANDLING
-RISK ACTIVITY COMBINATIONS, DAMAGE SCENARiOS
' \. .r.lll.· , \ · . . FIGURE 18.3 liiiJ I ;'·....... . &.............._........_. ________ ..._ ______ .......___......_ ........... ...___ .....
AND CRITERION VALUES
. .. . -. {~
-----_· .. ~-{-· -__ .· -.-' ---·-···· ___ · _: __ .t· ..... · -.. --. . i .• . . .I -- - ----~ --~-'i
>-t-
-m
<( m
0 n: n.
.8
.7
.6
I
I
I
I
I
I
l
I
I
;
oT-._----~----------------c
CR I TERlON VALUE
(!) CUMULATIVE DISTRIBUTION
ANY POINT ON 1"'HE CURVE
INDICATES THE 'PROBABlllTY
I P) THAT THE CRITERION·
VALUE (C) WILL NOT BE
EXCEEDED.
.5
.4
.3
. >-
t-
:i.2 -m
<( en o I
Q! ·n..
p
I
0 ~--+--------------c
CRITERION VALUE
@ DENSITY FORM
ANY POINT ON THE
CURVE INDICATES THE
-PftOBABILITY (?)THAT
A PARTICULAR CRITERION
VALUE ( Cl WILL BE
INCURRED.
ALTERNATIVE FORMATS FOR
PRESENTING THE ANALYTICAL RESULTS
---m
ct m
0
0: .n.
1.01
.3
.2
p
.I
CRITERION VAtJ.IJE
@ REVERSE CUMULATIVE
ANY POINT ON THE CURVE
INDICATES THE PROBABll..lTY
( P) THAT THE CRTtER lON
VALUE (Cl WILL BE
EXCEEDED.
FIGURE 18.4
1.0
0 w .9 LL.W
0~
ww ~8
(!)t-~0 .7 zZ w_.
u...J .s o::_ w:= a. .5 t-w-
<ti-LIJ :r: <( .::> .4 ...... ~_.
-
--~ >-~::> . 3 t-w
--0 _., liJ _.....,
.2 mo·!;t
<two m-,· oo-• t 0::0::~ Q..Q.._
0
.
lt
'. l
. ··-
I
EXPECTED VALUE I ~~~~ESTIMATE 90.25°.4-~ I
I ~ -it ~ROJECT ESTIMATE) I _l
~L·
L ~
t
foe= "LOW" ESTIMATE
.
/
. I
I
J . v I
~ I
/ I
I
-•
I
I
70 80 90 100 110 120 130
PERCENTAGE OF FINAL DIRECT COST ESTIMATE WITH CONTINGENCIES
CUMULATIVE PROBABILITY DIST,RIBUTION
FOR WATANA PROJECT COST
'
•
FIGURE 18.5
. -I
~
!
I
--+ ~·~
!
f
1
!
140
·[iJ
_J 1.0 _. -~·.
.9 w
1-(!) (/)~ .a Oz ow
0 .7 _JO::
<[W
::>·0... ~6 ~ oo
<(W
.I-! .5 ·~<(
<to :t:o
t-z .4 ->-o !::w .3
_.w -o .2 mx
<(l..LJ m ot--. I
o:O o..z
/
.L' ~ v
, ..
60
I -l --~~ EXPECTED VALUE~ ! ~C= •• HIGH .. ESTIM 91.5°k '~l ~ ·~~
ATE
I ~~PROJECT ESTIMATE) I
I
. v _z i
I
. [_)!'B ~ 11LOW"
1
ESTIMATE
~
/ l
t v I
L I
I
I
I
I .
70 80 90 100
PERCENTAGE OF PROJECT ESTIMATE
CUMULATIVE OiSTRlBUTION
OF DEV.IL CJ,\JNYON COSTS
I
110 120 130
FIGURE 18.6
,,
fill
-~) ' -
1.0
.9
(!) z .a
a
IJJ .7 w
0 w X w => .6 ..J I-~· 0 .5 z
0
lL w
0 .... 4 <(
>-0
1--.3 0
..J z -m .2 <( m
0 .I a=
0..
0
-··} ----., -.. --" -._,, -·-
I :
I
. -c--~ I
EXPECTEO VALUE~ ~C:: ''HIGH11ESTIMATE ' 9.06% -I
~· /
I 1V .
~· :/ A•(PROJECT ESTIMATE) '
-
/ 1 -
I ,fiJB '· ., B •''LOW
1
" ESTIMATE
/ I
I v I . I
""' I
/ I
I
' / -:t
I
Y! it
70 eo 90 100 110 120 130
PERCENTAGE OF FINAL DIRECT COST ESTIMATE WITH CONTINGENCIES
CUMULATIVE PROBABILITY DISTRIBUTlON
FOR SUSITNA HYOROELFtr:TRIC PROJECT
-
£
' ~
' ' t
~
\
-· .' . -
;
.)
i ;
'.~~"
1
< ~
1
rl
!I
' i
~
I
I
'
'
r
140
FIGURE 18.7
-·
-. . "
0.9 l.t t.3 1.5 1.7 1.9 2.1 2.3 2.5 2.7 2.9 3.1
RATlO OF ACTUAL COST TO 11 fNITIAL" ESTIMATE'
HISTORICAL WATER R.ESOURCES -
• PROJECT COST PERFORMANCE {48 PROJECTS l
FIGURE 18.8.
----~---
11.., .eo ....----+-----t----+-~-·~----
0
.70 ........ ---+----+-----1
.20
.10
0.9 1.1 1.3 1.5 1.7 1.9 2.1 2.3
RATIO OF ACTUAL COST TO "INITIAL .. ESTIMATE
COMPAaiSON OF SUSITNA RISK RESULTS WITH
HISTORICAL WATER RESOURCES PROJECT .
COST PER~RMANCE (48 PROJECT}
2.5
f-··.:·· ? .
DATA FOR 48
WATER RESOURCES
PROJECTS
2.7 2,9
FIGURE 18.9
3.1
-. . . -····
~~--------------------~----------------------------~~~ .. -·-------------~--------~---
1.0
.9
C)
z 8 0 .
laJ g LIJ .7
::l ~ ..J .• 6
0.""""" z~
LLQ ow
.5
>-~ .4
t-o
:iS
m~ .3
'((
al
~ .2
a.
.I
0
~-----
I
.. ,__,
·~ ~
·--· -/ / ~ ----/
)ifh = SCHEDULE ESTIMATE
_,£ -----INCLUDING A ONE YEAR
/ CONTINGENCY
/ 0
/ ..
/
-./
v
B= SCHEDULE ESTIMATE 7 :::::::1 WITHOUT CONTINGEI~CY . . I
-15 -10 -5 0 5 10
MONTHS FROM SCHEDULED COMPLETION
WATANA s.~CHEDULE OISTRIBUTIO~
EXCLUSIVE OF REGULATORY RISKS
~
~-·"-
'I
'· ,.
i
[
?
l
I
!
'
' ~ ' ;>
'
!
~
'"
' j
I
1
15
FIGURE 18.10 J sia I ,,"<_:l(.._,.~--..... -------~ ...... ---...... ---..... --..... -----...-.----llllillli.--..:~......-----......-....... ~~;;;;;;;;.a ,..-i.~
,; ,.,
1.0
.9
.8
(!)
z -. 7 0
i1J w
0 .6 X Ww
::l .5 ..... ..J ~~
lL 0 ~4
ow
>-ti "2 1-0 ·'-~
:::io
ffi z .2 <t-
m
~ .I
D...
0
;f.
'
·~---, .
~ ~
/ lo-"
....(o~ SCHEDULE ESTIMATE -/ INCLUDING A ONE YEAR
CONTINGENCY
~ ~ .
/ B= SCHEDULE ESTIMATE
WITHOUT CONTINGENCY
~ / .
7
__.,., v
-30
.
" .
-20 -10 0 10
MONTHS. FR0~1 SCHEDULED COMPLETION
WATANA SCHEDUL:E DISTRIBUTION
INCLUDIN.G THE EFFECT OF REGULATORY RISKS
\, ,,·
20
~.··
' ~ 1 ,
1
' I
' !
(
i1
If
. .
1 .
I r
t
I
l
i
l
' 1
l
' )
;
!
J
I
'
I
' i
" '
I l
I .
30
FIGURE 18.ll ill
1.0
(!) z· -.99 0 wo:
W<( ow
X>-w
a: .98 sw 0.. z
l.L(/)
0~ .97 >-0
!::a
-'w m~ .96 <( ,~ m-o-a:::· a a..~
•
T -~o-----~~ ~ ....... ss .,.. ~
,,P "',o~ · .r ~~v
'\0 / ~XPECTED VALUES: / o\-
./ ';:)0 TOTAL LOSS 0.06961 /,jv 50°/o REDUCTION o.o917t
I
I
I
ll_ v r
0 r
..
2 3 4 5
DAYS PER YEAR OF REDUCED ENERGY· DELIV~RY
CUMULATIVE PROBABILITY
DISTRIBUTION FOR DAYS OF REDUCED
ENERGY DELIVERY TO ANCHORAGE
1i
jl .. ~ n ~~ ~ i1 n
u n li p
·{ t H
fi
1! ~ I;
I ANCHORAGE I i
'
6 7
(j
FIGURE 18.12 ~~~~~J
1.0
(!) z
c
~~
o<t xw
. 99
W>
.,_a:: ow
zO... .98
LLcn
0~ <(
>-0
1-.97 :Jo _w
ml-~<( mo o-0::0 .96 .a.~ T
r
0
/
~
v
EXPECTED VALUE: .08116 .
. --~· ---
2 4 5
DAYS PER YEAR WITH NO ENERGY DELIVERY
CUMULATIVE . PROBABILITY DISTRIBUTION
FOR DAYS PER YEAR WITH NO SUSITNA
ENERGY DELIVERY TO FAiRBANKS
[
. 1
l . I '
1
J
I
!
I
I .
I FAIRBANKS I
6 7
FIGURE 18J3 [iJ
1-
)'
'
'
,1·· __ _
' #
o ••• ~
,,._,_---._---__
. -.~
i
I ~-...
1--
'
I
I
I
I
I '.
I
'~;
I
I
I
t
' I
I
RAILBELT .REGION
GENERATING AND TRANSMISS.ION FAClLlTJES
0 65 .130 lC1\.0._.£TERS
I :1 --
1!0 MI\.ES !Ail 0 65
FIGURE 18.14 · ·
::.·~
,·:
,.
' 1·:.
.. ·
I
I
... < .. _,.
I
""'
,,
LOCATION MAP
J..EG£NO
\f PROPOSED
OAM SlTES
----,ftOPOs.al 13$ ICV UNE I EXISTING LINES
I
• ......:
' I
I
•• •.·
I
I
I '"l./
t
;.1
c~~~
FIGUR.E JiiJ
.I
I
I
I
·II ..,.
a ·-·
I
I
I
I
:I ""'' .,
' ...
I
I ..,
' i: ·~
I . ;
~·,.· t '::'"
. l : r ,
I
Oil
2%
-·~-
Rural Electric Cooper1tives
.10%
MunicSp1l Sy,ternr..
23%
U.S. Gcv•rnment
Alatka Power
Adnaini!tration -Eklutna
1. Don Not Include Sttf Supplied Energy from
Mili".ary tnstallatians and The UnivlfSity oi
A Ink a
A ENERGY SUPPLY
(Based on Net G~neration 1~0)
Gas 76%
1. Does Not Include GeMration by Military
lnstt!latiQns and Thi UnivGnity .of Alaska
C NET GENERATION SV TYPES OF FUEL
(Eased on .Net Generation 1980)
Rural Electric Cooperativoes
58%
Municipal Systtrns
27% ' .
Urliversity of ...J
Afuka Administration -Eklutna
B GENERATING FACILITIES
(Based on Namepla~e Genera~h\l Capacit'/1980}
Dinll
(60.6 MW -6%)
Combined Cycle
Combustion Turbine
(139 MW-14%)
A-v-n•rative
Cycle
Combustion
'Turbine
~~t111MW-
Simple Cycle
Combustion Turb~n•
(520 MW-55%)
12%l
0 RELATIVE MIX OF ELECTRICAL GENERATiNG
TECHNOLOGY~· RAILBELT UTILiTIES --1980
FIGURE 18,1(1.
t ..
t 10,000•
I
I 9,000
I ~·..,_~~
s.ooo
I '
"'~·
I :'
7,000·
t 6,000·
li
I
I 4,000
I 3,000
I -.,.
1992
..
/
Entfgy Oe!lverits
From Susitna
~/
-
1•--------Watana Alone -----~--"14------Wauna And Devil Canyon------
I I 1 ' I I I ., I ' ' • I
1995 2000 2005 2010
v ......
FIGURE 1lt17 _ ENERGY DEMAND AND DEUVERIES FROM SUSITNA r~~~r~ 1 .
CL 9 2 9'
250
200
i ,.JIC
~
:i
t; 160 n u
~ ..,
c w
100
50
0
N§
lEGEND
·-~---·
•••••••••
•
-
Area Under This line It Annual Cost of Best Thermal Option r. (Including Investment Costs)
I WATANA ONLY IN 19941
.
Energy Coat of Otat Thermal 0 ptlon
Energy Coat of Sualtna Option
Operating Coats of 'fttermal Pla nt fn U111
in 1993 btended to 1994
Shaded Ana Represtntc Plant Operating
In 1992 ililpllleed by Watene
~I
H:~ -!~:~.~~mt·········· .•........• ., •..... ; ... ~ .....•. , ... J ... :~~Undcr This Line Is Annual Cost of Susitna Option
: Are1 Under This lines is Annual
:;;n~~11:: ::::: :::::::::
::::;
.:::;:
;;;;; .•:;:;
·~ :::::: •:•:
:•:~:
1,000
.----w-·.-Operaiifig Cost of Extsting ~apacity 1993/4
(Avoided Costs of Fuel and O&M Only)
Ar~a Represents Annt~al Operating Costs
from Existing Generating Plant
.-Common to Both Susitna and
, Thermal Options '" ..... , .. , ...
-!I!IJ:::: .. :::::~~f:lfii!'I:~::~::::::;~~~;W::;t¥.1.~;
~:;:_:: :;~ ~:::;~§:;] ~
'::: ::·· ~
II
::::: Medium Growth System Energy
:::~::: .//f.h, ,/f./.;.
::::::~:~:~imu ~·-"~:~ , · •
3.000 4.000 5,000
Annual En~'DY Output GWh
FIGURE 18.18-ENERGY PRICtNG COMPARISONS-1994
-
--
Rw. 1
380
360
340
320
300 -.z:
~ ~
280
·-~ 260. -M u u ·-.. a. 240 ,
c .,
::
M 220 8
>-Cit .. u 200 c: w
180
160
140
120
100
~-~-----------~~~---------
SYSTEM COST SAVINGS PRODUCED BY SUSITNA COMPARED
WITH BEST THERMAL OPTION IN MILLS· PER UNIT OF
SUSITNA OUTPUT IN CURRENT DOLLARS
~' # I
'I ·' I
•• COST SAVINGS FROM SUSITNA INCRr!ASING
# OVER WHOLE LIFE OF PROJECT •• •• ,. ., ....
I
!ncreaing Thermal Fuel / ~
Costs Avoided s ...
. . ~ . ••
~~·· ·---·-····# ~----:;'
•
/-Avolds Cost of 1 Further 200 MW Coal Find Generating Unit
• • • • .,_.L. Avoids Cost of 2 x 2f.l1' ~i!W CaJI Fired Gener~ting Units
Watana on Stream in 1993 Devil Canyon on Stra1m in 2002
9~4----5-----6--~-7----8--~9---~--00---0-t----~-~y-,-.~,~~4--~0~5~~06~~0~7--~0~8---0~9~~2~0~10~~,~.1---.1~2~FI_G_U~t:~E-1-8-.1-9 [i~~mJ
I
-
400 I""
300
200
100
0
r-l
' t-l
I I Area Under This LIM Is Annual Cost of Bast Thermal Option
H \ • (lncluC::~!J lnvestmen.t Cost) '
r-l: '------------. t I
r-~~~i~: I ~~~~~l:.l ~ ·····~·· -
I WATANA & DEVIL CANYON ~~:~03 J
LEGEND
•••• . Energy Cost af Best The.rmaJ CbJ:Ith>n
11::111111 -
~ ~
Energy_ Cost of Susitna Optio.ru
Operating Com of. Thermal' P.taottJn Usa
in 199~ Ext1nded to 1994
Shaded Portion Rtprtstnts Plard(lQ~rating
tn 2001 Obplac:ed by Susitna
r-~:i:l:i:l:"'·=·:·l!l!l!:·:·:~·:·:·"'=·:·:!l!l!·:·:·.. I
I ~mi11fm1l~lllllll~ ·--[111.. . r Are.• Under Thislinels Annual Cost of Sus~tn~ Option r-:;:;:::::;:;:;:;:;:;:;::::~::::;:; · 1 ~ (Including Investment Costs of Watana and Devd Canyon) 1~4:~~~~~~i~~ ~llllllltlllllllllllllllllllllltlttlllllltlllltlllllllllllallllllltB\11111ililtiJIIIIIIIIIIItii11111111111UIIIIDIIIIIIIIIIiiiiiiiiiJ :::::::::::~:::::::::::::::;:;:::: I . ; •
r-~:;:::::::::::::::::::::;:::::::: .. : : ~=l=~=~:l:l:l:l:l:l:l:l:l:i:l:l:l: • : : ::::::::;:::~::::::::::::::::::::: l••••••••••alliilmt••-<•••~r.t••u•m•••••••••••a..s "l~IIRIIII.III ! -=
,...·.······················································································ I E
• I
' I
I
' I
' I %.\;W}fi1\®~l11Wlt€i.J.@!f§i.~{m!~ Energy 0ll1put Wallna --j Watana and Devil Canyon-~
:rlllllllllltl l i ~~=t~:~~~~~4E~~1v
r-i~1~~f~~~;;~~~l~l11i1~~~11i~l~i~I~l~i~I~l;I~~~~;1~1~1~~;~lm;!~l~~~ll~~i~i!lll~I~lill~~~lllli~::;~:~~;~~~ i ~~4.
N~~;~!~~~l~ijlji~l~f~lj~j1jl1ll~~l~;jijl~lj!jljljljlilillfi~jijljljil1lll1ll~lli~lllli~~i:::li~~ 1 ; i 1 1 ~Other Hydro
1,000 2,000 3,000 4,000 s.ooo 6,000 [iU
Ann'Ual EHergy Output GWh ~PO[~
FIGURE 18.20-ENEr1GV PRICiNG COMPARISONS HUU[d
-•~J· J.J· •••• --
Rev, 1
r---------------------~------------------·-------------------------------------------------------------------------------------------------=----------------------------------~------------.
:~o
360
340
320
300 -.c
~ 280 -~ ._,
~ 260 -ait
3 ·a: c. 240 "0 c • :1
" 220 0 u
> r
' GJ 200 c w
180
160
140
120
100
94
~STATE APPROPR!ATION.· SCENA.RIO L___ 100% DEBT FINANCING
Susitna Mill Rata Cost With
7% hdlatlon" 10% I ntt'J'elt
COST SAVINGS
5 6 '7 8 2000 01 02 03· 04
Yean
COST SAVINGS GROWING OVER
WHOLI? OF SUSITt"A Ll FE
Mill Rate Cost
Best Thermal Option
7% Inflation, 10% Interest
05 06 07
Mill Rate Cost
Beat Thermal Option
0% Inflation. 3% Interest
08 09 2010 11 12 13
FIGURE 18:21-ENERGY COST COMPARiSON 100%'DEBT FINANCiNG 0 AND 7% INFLATION
'---------------------------------------------------------------------------------------------------------~--------------------------------------~------~--------------~------.-m•------------a. .. .-.......... _._. .. _.._._._._._ .. nJ
-.c
~ .:r: -s
:E -M
CP
.!:! ..
0..
"C c:
ftJ
"' ...
1111
0
0
> 0) ..
Q) c: w
"·
380
360
340
32Q
300
280
260
240
~20
200
180
160
140
120
--
SOOA, STATE APPROPRIATION SCENARIO
($2.5 BILLION) 7% INFLATION AND 10% INTEREST
Watana Completed with $2.1 billion ($1.0 bn 1982) of
GO Bonds 1991 -93; Cover of 1.25 at 89 Mills/kWh
and Allows Rtvenue Bond ReflnL,cing ln.1994
Susitna wholesale energy price {alJb.lts
anorgy increase= to 2009 and ris~
slowly thereafter
.Devil Canyon Completed with $8.0 billion ($2.4 bn 1982) of
Revenue Bonds1994-2002
100 Ll/._,..~-':__;s~u:si:tn:a~W~h:_:o:te:••;Je:.E;n;e;.:rgy~P-rliB~·--~-~-~ ..
94 5 6 8 9 2000 01 02 03 .04 05. 06 07 08 09 2010 11 12 13
Yean
FIGURE 18.22 _,ENERGY COST COMPARISON 50% STATE APPROPRIATION SCENARIO
Rev •. 1
--L!!l -.. -· -[8J! -------
380
360 ~
340
320
300 -.c a: u. 280 -"" -·-:E -260 ....
4) u ·.::
Q..
"C 240 c
tV
Ill .... ...
0 220 u
>-en ..
'CU c 200 w
'i80
160.
140
120
100
-
MINIMUM STATE APPROPRIATION SCENARIO
($2.0 BILLION) 7% INFLATION AND 100,.{, INTEREST
5
Mill Rate Colt
Btst Thermal Option
Susitna Pricing Reatdcted to
Maximum of Best Thermal Cost
Susitna Wholesale EnergY Price
Watana Complated with $3.4 billion ($1.8 bn 1982) of
GO Bonds 1990-93; Cover of 1.25 at 142 Mills/kWh
and Allows Revenue Bond Refinil~cing 1994
6 7 8 9 2000 01 02 03 04 05
Years
COST SAVINGS GROWING OVER.
WHOLE OF SUSITNA LIFE
Susitna wholesale energy price:hlls as
energy increases to 2010 and t~
~lowly thereaftElr
Devil Canyon Completed with $7.6 billion ($2.2 bn 191~2) of
Revenue Bonds1994-2002
06 07 08 09 2010 11 12 13
FIGURE 18.23 ·-ENERGY COST COMPARISON MINIMUM STATE APPROPRIATION
-
-
Rev. 1
-
380
360
340
320
300
-.c
~ 280
'"iit -~ 260 .,. _,
u ·;::
0.. 240
"'D c:
11:1 :s
C). 220 o·
~
&I 200 c w
180
160
140
120
100
94
------
$1.6 BILLION STATE APPROPRIATION SCENARIO
7% INFLATION AND 10% INTEREST
Mill Rate Cost
· Bast Thermal Option
'
--' .. ----
-· ,-
I • I I • I COST SAVINGS GROW.ING OV.ER I WHOLE OF SUSITNA LIFE
## ,#
~-~ .tit'
~,..----~~~--------.. ~~.
~ ....... «
\
........ ---··· __ .... ·-. * .~. •• •* I
I '---------1 • I
I
Watana Completed with $4.6 billion.($2.4 bn 1982)Qf
GO Bonds 1990 -93. Inadequate Covef for Revenue
Bond Refinancing Until 1995
7 8 9 2000 01 02 03
Susitna Price Tracks Coat of Best Thermal Option
Until Complete Conversion of GO Into Revenue Bonds
pevil Canyon Complet~ with $7.5 billion ($2.3 bn in 1982)
of GO Bends Converting to Revenue Bonds 2002 -2004
05 06 07 08 09 2010 11 1.2 13 6
Year~· FIGURE 18.24 -ENERGY COST COMPARISON WITH FINANCING flESTt:UCTED 94/95 AND 03/04
I
I
I
I
I
I
I·
I
I
·I
I
I
I
I
I
I
I
I <.
SPECIFIC FUSfS 1: RISK OF· GO BOND FINANCING OVERRUN
0
Probability
.5
.2
Foracast ($ bn) .1
~
0.5 . 1.0
Probability of
GO Bond
Requirement
Exceeding
$1.5.bn • .15
1.5
GO Bond Financing Requirement
Probability of
Exceeding $2 bn
2.0
F=GURE 18.25 -GO BOND REQUIREMENTS IN 1982 DOLLARS BILLION
SPECIFIC FINANCING RISK II: EARLY YEAR NONVIAB!LITY
Probability
.5
.4
\
.3
0 10 20 30
Probability of Watana
Unit Costs Exceeding 45% of
Best Thermal Option
60 70
Watana Unit Cost as% Best Thermal
FIGURE 18.26-WATANA UNIT COSTS AS PERCENT OF BEST THERMAL OPTION IN 1996
[ AGGREGATE RISK I
.Probability of Cumulative
Net Operating Earnings
Falling Below $1 bn
0 0.5
Probability
.5
.4
.3
.2
Forecast ~ .1
$1.35 bn
1.0 1.5
Cumulative Net Operatinq Earnings $ bn
2.0 2.5
FIGURE 18.27 -CUMULATIVE NET OPERATING EARNINGS BY .2001