16728 DECEMBER 1981 EY2

MEDIUM HEAD HYDRO POWER HOUSE

CONCRETE VOLUMES
By James L. Gordon,1 M. ASCE

f SUMMARY

Equations are developed for rapid estimation of medium head power house
concrete volumes for conventional surface power houses containing vertical
'fis, F rancis, Kaplan, or fixed-blade propellor turbines with steel spiral casings.

•i TROOUCTION

During the design of a hydropower project, there are various methods whereby
an engineer can determine whether optimum use has been made of the materials
for each component, in order to obtain the most economic project layout. For
example, the design of a dam can be optimized on the basis of attaining the
required factors of safety for stability . Similarly a penstock can be optimized
based on the estimated installed cost and the capitalized hydraulic losses.
However, for a power house substructure, as in Fig. 1, there are no guidelines
available to determine whether maximum usage has been made of the concrete
substructure. There are recognized criteria for stability and flotation, but
adherence to these criteria will not ensure the most economic layout since a
different layout may result in less use of concrete, and still satisfy stability
requirements.
If a simple measure of the required substructure concrete volume could be
developed, then at least an engineer will have a yardstick against which he
can compare his own or alternate designs. Furthermore, such a measure would
be useful in feasibility and prefeasibility studies, where much time is spent
on calculating power house quantities.

DEVELOPMENT OF FORMULA

For many years, the author has used the following formula to extrapolate
from one power house size to another:
1
Vice Pres. and Hydro., Montreal Engrg. Co., Limited, P.O. Box 6088, Station A,
,r ontreal, Canada H3C 328.
· Note.-Discussion open until May I, 1982. To extend the closing date one month,
:. written request must be filed with the Manager of Technical and Professional Publications,
· ASCE. Manuscript was submitted for review for possible publication on November 7,
1980. This paper is part of the Journal of the Energy Division, Proceedings of the American
Society of Civil Engineers, ©ASCE, Vol. 107, No. EY2, December, 1981. ISSN 0190-
8294/ 81 / 0002-0237/$01.()0.

237

I
238 DECEMBER 1981 EY2
2 5
V. =K d · . (1)

The exponent, at a value of 2.5, was rationalized as half way between 2,

which would represent a hollow cube with constant wall thickness, and 3, which
would represent a solid cube. The constant, K, was obtained by substituting
the substructure volume and turbine throat diameter in the existing power house.
T he volume for the new power house was then obtained by calculating the
new turbine throat diameter, (Fig. !(a)) and substituting into the formula. Over
the years, this formula worked well and was expanded to include the number
of units and the effect of an erection bay, when it was found that concrete
required for an erection bay generally represented about half the concrete required
for one unit. This resulted in the following formula: ·
V, = K(N + 0.5) d 2
' •• •• •• • ••. • •• • ••••••• • . •••• •• • (2)
However, the author has never been fully satisfied with the rather arbitrary
selection of 2.5 for the exponent; thus, it was decided to undertake a mo;
detailed analysis, to try to include other factors which affect concrete volurr,
in the power house substructure.

(a) ( bl

FIG. 1.- Typical Sections through Pow er House: (a) with Head below Approx 100
m; (bl with Head above Approx 100 m

Eq. 2 takes into account the physical size of the turbine, the number of
units, and, an allowance for the erection bay. Other factors which affect the
concrete volume are:

l. Size of the generator casing.

2. Runner submergence below tail water.
3. Tail water level at flood.
4. Foundation geology and topography.
5. Size of erection bay .
6. Layout of equipment.
7. Requirement for a penstock valve.

To try to develop a formula which takes all the foregoing factors into account
would be time consuming, but could probably be accomplished, provided sufficient
data were available. Unfortunately, a search of published literature has revealed
little useful data, thus, some of the factors could not be included in the evaluation.
 EY2 CONCRETE VOLUMES 239
These factors, and their effect on concrete volumes are discussed in the following
paragraphs.
Size of Generator.-An article published by Taylor (9) has indicated that unit
spacing will be dictated by the generator, when unit speed is 428 rpm or more,
and by the turbine, when unit speed is 375 rpm or less. Due to the advances
which have been made in equipment design, since the article was published
in 1969, and due to the different criteria used by the various designers for
equipment layout, a different answer has been obtained from an analysis of
existing power house layouts. It has been found that the generator begins to
affect the unit spacing when the head exceeds about 100 m. This will be discussed
in more detail later. For an initial appraisal, units where head exceeds 100
m have been designated in Table l.
Runner Submergence Below Tail Water.-Most turbine units are now set with
center line of the spiral casing at, or a few meters below, low tail water level;
lhus, it was reasoned that within this narrow range of settings, unit submergence
jvm have but a minor effect on overall power house concrete volume. Therefore,
tlnit setting was not considered further.
Tail Water Level at Flood.-As tail water levels increase, the downstream
wall of the power house must be reinforced to resist the higher water pressure,
and mass has to be added to the substructure to counteract any tendency towards
flotation. Since most of the data available to the author were for power houses
with a normal range of tail water levels from about spiral case center line
level up to about the elevation of the top of the generator, it was decided
to exclude this factor from the analysis.
Foundation Geology and Topograpby.- Most power houses are located on
a bedrock foundation, and since the foundation loading imposed by a power
house is relatively small, it was reasoned that fo r competent foundations the
foundation geology will have but a minor effect on power house concrete volume.
Therefore, this effect was not considered further. However, it can have a marked
effect where adverse topography, rock levels, or geologic conditions are encoun-
tered.
Size of Erection Bay.-The length and number of erection bays has a distinct
effect on the power house concrete. Since previous experience has indicated
that erection bays contain about half the concrete per meter length as contained
per meter length in a turbine unit, it was decided to measure the total length
of erection bays in a power house, and obtain the "equivalent unit" by dividing
the erection bay iength by the unit width, and by a factor of 2. Thus, an
erection bay having a length equal to one unit width, would have an "equivalent
unit" value of 0.5. Table 2 shows the results of this analysis. (In all tables,
the projects are listed in alphabetical order.) For multiple unit plants, the unit
,i width was taken as the unit spacing. For single unit plants, the drawings were
{ examined to determine what the unit spacing would have been if a second
l ·, unit had been installed and the repair bay length was then calculated as the
·J
i total power house length minus the multiple of number of units times unit
width. It is interesting to note the wide variation in repair bay equivalent unit
ratio from a minimum of 0.18 at Brazeau, where the generators are serviced
in place (4) and the turbine is removed through the generator rotor, minimizing

I
repair bay area, to a maximum of l .62 at the Grand Coulee third power house,
where there are two repair bays, one of which will be required for the power

f
240 DECEMBER 1981 EY2

TABLE 1.-Power House Substructure Concrete St atistics

Concrete, in
Turbine Num- volu m e
D 3, in ber times
cubic of N+ cubic VJ
Project name Valve meters units R/2S meters (N + R/1S)
(1) (2) (3) (4) (5) (6) (7)
Aviemore No 125 4 4.75 116,000 24,420
Bayano No 115 4 5.07 52,800' 10,410
Bay D'Espoir 1~• Yes 14 6 7.35 22,800 3,100
Bay D' Espoir 7 • No 41 l 1.50 5, 350 3,570
Bighorn No 34 2 2.65 8,350 3, 150
Boysen Yes ll 2 2.57 9,800 3,810
Brazeau 1- 2' No 78 ' 2 2.18 9,900 4,540
Charlot River No 5.6 2 2.54 1,420 560
Chute Georges' Yes 8.4 2 2.46 2,800 850
Chute Willson Yes 27 l 1.50 2,780 1,850
Coteau Creek No 95 3 3 .46 33,000 9,540
Dadin Kowa No 25 2 2.54 5,000 1,970
Davis No 112 5 6.74 82,300 12,200
a
Glen Canyon No 62 8 8 .40 152,260 18,130
Grand Coulee' No 774 6 7.62 413, 165 54,220
Hart Jaune Yes 18 3 3.25 4,860 1,500
Hinds Lake' No 9.9 I l.50 3,350 2,230
Horsechops Yes 1.9 I l.64 400 250
Jebba No 358 6 6.70 101 , 800 15,190
Kingston Mills No u l l.00 !JO !10
Kootenay Canal No i ]6 4 4.97 38,230 7,690
La Grande 3 No 178 !2 13.20 116, 000 8,79 0
La Grande 4' No 171 9 I0. 10 80,500 7,970
Mayo 2 Yes 1.3 1 1.00 210 210
Outardes 2 Yes 114 3 3.74 44,800 J 1,980
Peace Canyon No 512 4 4.80 107,040 22,300
Pelton No 40 3 3.46 12,480 3,610
Pocatena Yes 9.7 1 1.26 900 720
Rattling Brook • Yes 1.2 2 2.62 630 240
Sandy Brook. No 4.3 I 1.44 420 290
Seven Mile No 221 4 4.97 67,740 13,630
Smelter No 16.2 l l.50 3,160 2, 110
Snare Falls No 15.1 l l.15 l.310 1, 140
Spray 2' Yes 3. 8 l 1.00 410 4 10
Taltson No 32 I !.36 3,060 2,250
Waterloo No 15. l I 1.20 l.l60 970
Wells No 127 2 2.43 16,390 6,740
Wbatshan• Yes l0.5 I 1.52 3,820 2,513
Whitehorse l-2 No l1 2 2.07 2,020 980
Whitehorse 3 No 22 I l.00 1,530 1,530
Yellowtail' No 17.4 4 4.46 17,400 3,900
Yellowtail' No 17.4 4 6.78° 17,400 2,570
' Unit where bead exceeds (328 ft) JOO m .
b Has dismantling sections for closure by bulkhead.
' Average of throat diameters 4.07 and 4.47 in.
• Includes concrete for future units 3 and 4.
' Including equivalent repair bay length.
Note: 1 m 3 = 1.31 cubic yards.
 EY2 CONCRETE VOLUMES 241

TABLE 2. -Dete rmination of Erection Bay Factor

Erection bay
Project Unit width, S, in meters length, R, in meters R/2S
(1) (2) (3) (4)

Aviemore 23 34 0.74

) Bayano
Bay D'Espoir 1-6
Bay D'Espoir 7
20.1
12.8
18.3
43.1 •
34.5
18.3
1.07
1.35
0. 50
Bighorn 12.2 15.9 0.65
Boysen 12.2 14.0 0.57
Brazeau 1-2 . 15 .9 5.8 0. 18
Charlot River 7.9 8.5 0 .54
Chute Georges 9.8 9.0 0.46
, Chute Willson 13.4 13.4 0.50
} Coteau Creek 19.8 18.3 0.46
' Dadin Kowa 12.5 13.4 0. 54
Davis 22.0 76.5 1.74
Glen Canyo n 19.8 15.7 0.40
Grand Coulee 36.3 117.6 l. 62
Hart Jaune 11.6 5. 8 0.25
Hinds Lake 13.4 13.4 0.50
Ho rsechops 5.0 6.4 0.64
Jebba 27 .9 39.0 0.70
b
Kingston Mills 5 .8 - 0 .0
Kootenay Canal 22.0 42.7 0.97
La Grande 3 21.6 51.8 l.20
La Grande 4 21.3 46.6 I.JO
b
Mayo 2 7.9 - 0.00
Outardes 2 20.1 29.6 0.74
Peace Canyon 30.5 48.8 0.80
Pelton 16.6 15 .2 0.46
Pocaterra 8.8 4.6 0.26
Rattling Brook 5.8 7.2 0 .62
Sandy Brook 7.2 6.3 0.44
Seven Mile 25.6 49.7 0 .9i
Smelter 9.8 9.8 0.50
Snare Falls 10.4 3. 1 0.15
b
Spray 2 9.7 - 0.0
Taltson 12.8 9 .1 0.36
Waterloo 9.6 3.8 0.20
Wells 20.7 18.0 0.43
Whatshan 14.0 14.6 0.52
Whitehorse 1- 2 10.7 1.4c 0.07
Whitehorse 3 13.2 - b
0.0

) Yellowtail
Yellowtail •
11.6
I J.6
10.7
32.3
0.46
2.78
•including equivalent repair bay length.
"Power house extension, no repair bay required.
<Repair space available between units .
Note: I m = 3 .28 ft.
242 DECEMBER 1981 EY2
house extension. Also, some explanation is required for the two dimensions
given for Yellowtail, where there are four units separated by 11.6 m and 12.5
m at a construction joint. At Yellowtail, there are two hollow jet valves in
an outlet structure at the east end of the powerhouse. The service bay has
a nominal length of 11.6 m, and the power house crane rails extend through
the service bay into the outlet structure. Total length of power house and service
bay is 78.7 m, for which an " equivalent repair bay" length has been calculated
by subtracting four unit widths at 11.6 m from the total power house length.
Layout of Equipment.-It was decided that it would be impossible to quantify
equipment layout in a rational manner, so that it could be included in a formula .
Accordingly, it was discarded from further analysis.
Requirement for a Penstock Valve.-A control valve on the penstock upstream
from the turbine will increase power house concrete, thus, it was decided to
include this factor by assigning a different designation to those power houses
containing valves, when plotting the data.

UN1T SPACING Ser av TURBINE

)
Based on inclusion of the repair bay length, Eq. 2 can now be changed to
the following:

V, = K(N + ~) dx .. . . . . . . . . ... . . ... . . . . . . . . ... (3)

in which K and x are to be determined. Data for 40 power plants was then
assembled and is reproduced in Table J, and the concrete volume per unit
has been plotted on log-log paper against the cube of the turbine throat diameter
as in Fig. 2.
An examination of Fig. 2 indicates that, as would be expected, there is a
wide variation in unit concrete, which can be almost 4 : I as indicated by the
difference between Waterloo and Bay D'Espoir units l to 6 which both have
turbines with almost identical throat diameters, but operate under different heads,
which are only 21 mat Waterloo compared with 173 mat Bay D'Espoir. Most
interesting is a line can be drawn which represents a minimum volume of concrete
per unit. The equation of this line is
Vu = 130 d 24 . . . • • • • . • • . • • • . • • • • . . • _ •.•....•• . (4)
A total of 18 power plants fall on or near this line, not counting the three
small power plants which have less unit concrete. An examination of the features
common to these power plants indicates that:

1. All are located on competent rock foundations with the rock level near
or above repair bay floor level.
2. None have tail water levels at flood above the level of the top of the
generator.
3. With one exception, at Spray 2, all have turbine rated heads less than
120 m.
4. Three of the power plants have valves, which indicates that the addition
of a valve need not significantly affect unit concrete volume.
5. All have turbines with spiral cases set at or near low tail water level.
 EY2 CONCRETE VOLUMES 243
6. Four of the units have generators with extra inertia. The additional inertia
is 185% at Charlot, 105% at Whitehorse, 136% at Taltson, and 70% at Dadin
Kowa (6). However, these four units operate at heads of less than 30 m. Even
with additional inertia in the generator, the turbine spiral case still determines
unit spacing and concrete requirements in the powerhouse substructure.

I
r
Based on the foregoing, it is evident that powerhouse layouts for units which
fall within the above constraints, can be developed to limit the volume of concrete

't ,o,,---------~------------------~ GRANO COULEE

l
@

t
i :j t
I "'e
..
0
"'::,
J (,)

! ;., @ GLl:N CA.N'rON

..0
li

I .....
C

l'
a:
...
:t

f "";;;z
a:
...
::, W HAfSHAN
L E GENO

• \JNJT SPACING DICTATE0

OV TUABIMf CASING

®
15
~~ ~e :£..,;~~;RD
~~~i!~0
A UNITS W ITH VALVES

POWERHOUSE UNIT CONC RETE VOLUME m 3 ___..,..

FIG. 2.- Refationship between Power House Concrete Volume and Turbine Throat
Diameter

I to that indicated by the following equation:

l V, = 130 ( N + : ) d2
4
• • • • • • • • • • . (5)

'
2
. Turning now · to some of the units which do not fall on the minimum line,
f )1e reasons for less concrete being used at four of the power plants are:

I
1. Kingston is a small, 500-kW unit with a vertical steel cone draft tube.
There is only sufficient concrete to surround the unit as described by Newbury
and Hutt (8).
2. Sand Brook, Waterloo, and Pocaterra are small single unit plants where
bedrock was fo und at repair bay floor level, requiring only a thin covering

I
244 DECEMBER 1981 EY2
of concrete. At these plants, rock excavation was just sufficient to accommodate
the unit, thus, minimizing concrete requirements.

As for the units which exceed the minimum line:

1. At Rattling Brook, Hinds Lake, Bay D'Espoirunits l-6 and unit 7, Whatshan,
Grande Coulee, Glen Canyon, and Yellowtail, all have generator sizes which
affect unit spacing.
2. Smelter includes a substantial concrete superstructure, and additional
concrete in the substructure required to bridge a cavity in the rock foundation.
3. Bighorn includes additional concrete in the substructure along with a
post-tensioned anchorage system, to overcome a tendency towards sliding on
a buried coal seam, as described by Thicke and Bakar (10).
4. At Glen Canyon, where approx 12-15 m of mass concrete was required
below the powerhouse, Aviemore, and Boysen, foundation rock is located several
meters below the bottom of the draft tube.
5. Davies has a conservative design, two pier draft tubes producing a wide
unit spacing, an outdoor type design with a concrete floor several meters above
the generator, and substantial concrete in the repair bay areas-all of which
contribute to a large unit concrete value.
6. Coteau Creek power house rests on a soft expansive shale foundation,
and contains large water holding tanks to balance load on the foundation and
keep the units vertical (1).
7. Bayano was built on a soft tufaceous siltstone, having an unconfined
compressive strength ranging between 100 and 140 kg/cm 2 , which required
additional concrete and keying of the substructure concrete into the rock
foundation.

Based on these few examples, it is evident that the adverse foundation conditions
can double, or even quintuple, the volume of concrete in a power house.

UNrr SPACING Ser sv GENERATOR

Turning to the problem of estimating concrete volumes for power plants where
the unit spacing is dictated by generator space requirements, it was decided
to determine, first, whether there is a simple parameter used to distinguish
between power plants where unit spacing will be established by the generator,
from those where spacing will be determined by the turbine. A comparison
of unit spacing to turbine throat diameter was made for the multi-unit power
plants, lying closest to the minimum unit concrete volume line in Fig. 2. This
is shown in Table 3 which indicates that unit spacing is about 3.8 times throat
diameter for power plants where the turbine throat diameter is more than approx
3 m, and between 4.4 and 4.8 times throat diameter where this is less than IJ
approx 3 m. ,
The Brazeau power plant has a unit spacing 3.9 times the throat diameter
of the first unit, and any increase in size of the generator casing would have
required an increase in unit spacing (5). If unit spacing of about 3. 9 dis accepted
as a practical minimum, then the ratio of generator casing size to the throat
diameter at Brazeau of 3.0 can be regarded as a practical maximum, beyond
 EY2 CONCRETE VOLUMES 245
which the generator will affect unit spacing. (It is interesting to note that the
same conclusions can be reached on the La Grande 4 power house layout.)
Accordingly, the ratio of generator casing size to turbine throat diameter was
determined for those power plants where unit spacing was established by the

TABLE 3.- Effect of Turbine on Unit Spacing

Project Unit spacing, S, in meters Turbine, d, in meters S/d

(1) (2) (3) (4)
Brazeau I 15.9 4.07 3.9
Charlot 7.9 1.77 4.5
Chute Georges 9.8 2.03 4.8
Dadin Kowa 12.5 3.21 3.9
Ha rt Jaune I l.6 2.62 4 .4
Jebba 27.9 7. 10 3.9
) La Grande 3 21.6 5 63 3.8
La Grande 4 21.3 5.55 3.8
Peace Canyon 30.5 8.00 3.8
Wells 20.7 5.03 4.l
Whitehorse 1-2 I0.7 2.23 4.8
Note: l m = 3.28 ft.

TABLE 4.-Effect of Generator on Unit Spacing

Unit Diameter
Speed, in Head, Generator, Turbine,
revolutions h, in G, in d, in
Project per minute mete rs meters meters G /d
( 1) (2) (3) {4} (5) {6)
Bay D' Espoir 1--6 300 173 9 .1 2.41 3 .8
Bay D'Espoir 7 225 173 !0.8 3.45 3.1
Brazeau l64 118 12.2 4.07 3.0
Glen Canyon 150 174 13.3 3.96 3.4
0
Grand Coulee 85.7 105 28 .9 9. 18 3.1
Hinds Lake 360 2!4 9.5 2. 15 4.4
La Grand 4 128.6 !!7 16 .5 5.55 3.0
Rattling Brook 514 IOI 3.6 !.06 3.4
Spray 2 450 273 7.1 1.56 4.6
Whatshan 327.3 168 8.8 2 .19 4.0
1.(i Yellowtail 225 147 8.7 2.59 3.4
·. •/
h • Allis Chalmers unit speed, maximum head .
l
I h} = N=o==t=e==:=l ==m====3==·=28= ft==.= = = = = = = = = = = = = = = = = = = = = = =
i
t..··
{1
generator. This is shown in Table 4 where the genera tor casing size, G, (Fig.
l(b)) has been defined as the outside diameter of the circular steel casing or
the distance across outside flat concrete or steel surfaces in a rectangular casing.
f) Taylor (9) undertook a similar analysis, using certain assumptions, one of
which was that the generator diameter was equal to the rotor diameter plus

I;.\
',!
a
",j
246 DECEMBER 1981 EY2
a constant of 4.17 m, based on data by Walker (15). If instead it is assumed
that the generator diameter is some function of the rotor diameter, then Eqs.
2 and 12 in Taylor's article can be combined to show that the G / d ratio is
a function of head divided by speed, with unit size cancelling out. Table 4
lists both head and speed, and an examination of Table 4 will indicate that
in all cases the G / d ratio exceeds 3.0 and reaches a maximum of 4.6. No
relationship was found between G / d and h / rpm; however, it can be seen from
the tabulation that the G / d ratio tends to increase with the increasing head,
as indicated in Fig. 5, as would be expected.
10'~- - - - - - - - - - - - - - - - - - - ~
/
/
/
GRANO COULEI;

Vu=lOG~
/
/
t
'-.
"--... j
,0'1 - - - - - - - - + - - - - - - - - " t - t --+--,~ ------1

,o' 1 - - - - - - - - - -~~,,___________,
tm•os iAKY e1,.Y o·es1>01~ , - o

SPRAY 2 II!
I \ \_ "f 'Ei.LO~TAB.
\_ w1-u,rstt,+

/
/
,o's---- - --+---,£..-- - l - - - - - - - -+-- - ------l
/
,/. ATTLJNG BROOK

/
'{/
I

3
POWEflHOUSE UNIT CON CRE TE VOLUME m ~

FIG. 3.-Relationship between Power House Concrete Volume and Generator Casing
Size

Based on the foregoing, power plants with a G / d ratio below approx 3.0
will have a unit spacing established by the turbine, and power plants with G / d
ratio over approx 3.0 will have a unit spacing established by the generator.
Also, power plants likely to have unit spacing established by the generator fl
:I
are those where the head exceeds approx 100-120 m, depending on design of
the generator, for generators with standard inertia, and at lower heads for power
plants with generators having extra inertia, depending on whether the additional
inertia has been obtained by increasing the rotor diameter, or adding to the
rotor weight.
 EY2 CONCRETE VOLUMES 24 7
Since it was found that power plant concrete volume was a function of turbine
throat diameter in Fig. 2, it was decided to plot the unit concrete volume against
the cube of the generator casing size as shown in Fig. 3.
From Fig. 3, it is apparent, based on the examples listed in Table 5, that
!
! power house concrete volume is a function of generator casing size. The exceptions
can be explained as follows:
,. 1. Spray 2, is a small single unit addition to an existing power house with
,f,
competent rock at generator level, where controlled blasting was used during
{l
t construction, both to minimize overbreak and prevent damage to the adjacent
l operating unit. A measure of the care used in excavation can be obtained from
l the estimate for concrete, which was 695 m 3, where only 405 m 3 was poured.
j 2. Glen Canyon includes a large volume of mass concrete within the power
house foundation.
'I 3. Yellowtail includes an outlet structure; however, when the unit concrete
_1olume is reduced to allow for the outlet structure concrete, a more satisfactory
result is obtained.

TABLE 5.-Power House Concrete and Generator Casi ng Comparison

Unit, in Casing Unit concrete volume.

Project name megawatts G' in cubic meters
( 1) (2) (3) (4 )
Bay D'Espoir 1-6 75 750 3, 100
Bay D' Espoir 7 150 1,260 3,570
Brazeau l -2 153 1,820 4,540
Glen Canyon I 12.5 2,350 18,130
Grande Coulee 650 24, 140 54,220
Hinds Lake 75 860 2,230
La Grande 4 300 4,490 7,970
Rattling Brook 6.4 47 240
Spray 2 47 360 4IO
Whatshan 50 668 2,513
Yellowtail 6i.s 658 2,570
Yellowtail 62.5 658 3,900
Note: Im' = 1.31 cu yd.

r
i
i

From Fig. 3 it will be observed that two equations can be derived. For a minimum
concrete volume the following equation should be used:
J V" = 10 G 2 4 • . • •... .... .... .. ...•• . •. ... . . .. • . (6)
'I from this, the total minimum concrete volume in the power plant can be obtained
ffrom the following equation:
2
V, = 10(N + :) G ·• •• • • ••••••••• •• . .•• • • • •. • ..• (7)
2
A more conservative (higher) estimate of the concrete volume can be obtained
from
248 DECEMBER 1981 EY2
V" = 12 G 2 ·5 . (8)
From which a conservative estimate of the total concrete volume in the power
plant can be derived from the following equation:

V, = 12 ( N + : ) G
25
• . • . • •• •• • • •.• . • .... . . • •• .• . (9)
2
It should be noted that Eqs. 7 and 9 will include the effect of additional inertia
in the generator rotor since power house concrete volume is derived as a function
of generator casing diameter, which, in turn, is a function of generator rating,
speed, and rotor inertia. For example, the unit at Hinds Lake contains 145%
additional inertia (6), yet still plots on Fig. 4 in the minimum area. Furthermore,
it is interesting to note that Eqs. 5 and 7 are identical when G = 2.9 d. In
other words, for the power plants investigated, the generator size begins to_
affect unit spacing when the generator casing size exceeds 2.9 times the turbin ) _,
throat diameter. ·e I

PRELIMINARY ESTIMATES

Since the formulas developed so far are relatively simple, it would appear
worthwhile to continue the investigation further, in order to substitute other
dimensions for G and d since both of these dimensions require a fair amount
of effort in their determination. For power plants operating under a head of
less than 100-120 m, the concrete volume is a function of turbine throat diameter.
If an average throat velocity of 9.0 m/s is assumed, then throat diameter can
be converted into flow, which in turn can be converted into a function of
power and head. In this manner, Eq. 5 can be converted into the following
formula, assuming a turbine efficiency of 90%, and adding an extra 15% to
allow for the fact that Eq. 5 gives a minimum concrete volume

V, = 1.05 ( N +
R )(kW),.,
-h- . . . . . . . . . . . . . . . . . . . . . . (JO)
28

In order to deter mine whether this equation is accurate, the power as expressed
in kilowatts per meter head was calculated for the developments shown in Table
6 and is plotted against the unit concrete volume in Fig. 4 , from which it will
be noted that Eq . JO gives a conservative value for the unit substructure concrete
volume . The five developments which have substantially more concrete in the /
s ubstructure than indicated by the formula, are those at Smelter, Bayano, Coteau
Creek, Davis, and Boysen.
Eq. JO can also be used to obtain an appreciation of the advances in power ) '
house substructure and turbine design which have been made over the pas,n ·;. .
40 yr, by comparison with similar data developed by Creager and Justin (31 . ·
based on 84 power houses, 38 of which have heads between 20 and 100 m,
and were built in the United States between 1921 and 1943 . The curve shown
in Ref. 3 which shows power house substructure concrete in cubic yards per
horsepower per unit plotted against head in feet can be converted to the following
formula:
 EY2 CONCRETE VOLU M ES 249

kW
V, = 4.2-;;.-2 (11)
h·

for the range of heads between 20 and 100 m. Eqs. IO and 11 can be compared

TABLE 6.-Power per Meter Head and Substructure Concrete Volume

Unit
Capacity per concrete
Turbine head, in volume,
I' Unit capacity, head, in kilowatts i n cubic

r Project name
(1)
in megawatts
(2)
meters
(3)
per meter
(4)
meters
(5)

I
i
\ )
Bayano
Bighorn
~/ ! Boysen
75 .0
55.0
7.5
50.0
74.7
29.3
1,500
740
260
10,410
3, 150
3,810

I Brazeau
Charlot River
153.0
5.5
117.7
28 .0
l,300
200
4,540
560

i
j
Chute Georges
Chute Wilson
Coteau Creek
25.5
60.0
55.0
l03.0
78.9
52.7
250
760
1,040
845
1,850
9,540
Dadin Kowa 17.0 29.4 580 1,970
Davis 45.0 36.6 1,230 12,200
Hart Jaune 16.0 37.5 430 1,500
Horsechops 7.3 84.l 90 250
Jebba 88.0 27.7 3,180 15, 190
Kingston Mills 0.5 12.2 40 110
Kootenay Canal 132.0 74.7 1,770 7,690
La Grande 3 192.0 79.3 2,420 8.790
La Grande 4 300.0 116.7 2,570 7,970
Mayo2 2.5 33.5 80 210
Outardes 2 151.0 82.9 1,830 11,980
Peace Canyon 175.0 39.6 4,420 22,300
Pelton 40.0 46. 3 860 3,610
Pocaterra 13.4 56.4 240 720
Sandy Brook 6.0 33.5 180 290
Seven Mile 175.0 57.9 3,020 13,630

I Smelter
Snare Falls
31.0
6.7
86.2
19.2
360
350
2,110
1,140
}_ Talston
Waterloo
18.5
8.0
29.9
21.3
620
380
2,250
970
t
Wells 100.0 62.2 1,610 6,740
5.5
i. Whitehorse 1-2
Whitehorse 3 8.0
18.6
17. I
300
470
980
1,530

I = 1.31 cu yd.
3
, · Note: m
°'_
h ll= = = = = = = = = = = = = = = = = = = = = = = = = =
: JI
I and it will be seen that Eq. 10 gives lower substructure volumes for units
smaller than about 100 MW, with the difference increasing with increasing head.
At a head of 57 m and a generator capacity of 25 MW, the average head
and capacity of the 36 plants included in the Creager data, about half of the
difference in the substructure volumes can be accounted for by the higher turbine
250 DECEMBER 1981 EY2
throat velocities and efficiencies in modern turbines, thus, reducing turbine
size and, therefore, substructure concrete volume, and the other half of the
difference is, therefore, due to the more efficient use of substructure concrete
in current power house designs.
From Ref. 3 it is interesting to note that the power house substructure volume

TABLE 7.-Generator Casing Size-Turbine Head Relation

Project G/2.9d Turbine head, in meters

( 1) (2) (3)
Bay D'Espoir 1-6 1.30 173
Bay D'Espoir 7 1.08 173
Brazeau 1.03 118
Glen Canyon 1.16 174
Grand Coulee 1.09 108
Hinds Lake 1.52 2!4 )
f
La Grande 4 J.03 117
Rattling Brook 1.17 !01
Spray 2 1.57 273
Whatshan l.39 168
Yellowtail 1.16 147
Note: l m = 3.28 ft.

0
~
:,;

...,.. ••'
0:

:l

..:5
.,...
...<
~ Vu • 1.05 ( K; )'·'
~ ,o'
"

30
,o' 10•

UN I T SUBSTRUCTURE CONCRETE m 3 ____..

FIG. 4.-Relationship between Power, Head, and Substructure Concrete

is independent of head, for heads in excess of 110 m, thus, confirming the

previous finding that the turbine ceases to influence substructure concrete
volumes, at heads in excess of about I00-120 m . For these higher head power
plants, conversion of Eq. 7 into a function of head and power is somewhat
more imprecise, but can be accomplished. Since it has been determined that
the generator begins to affect concrete volume when G is greater than 2.9
 EY2 CONCRETE VOLU M ES 251
d, a relationship between generator size and head can be obtained by comparing
h with the ratio G /2.9d, in Table 7 and as plotted in Fig. 5, from which the
following formula has been derived:
G
- - = 0.10 h 0 · 5 . . (12)
2 .9d
The value of G obtained in Eq. 12 can then be substituted into Eq. 9, and,
using the same relationship between turbine throat diameter, d to power and
head as outlined previously, it will be found that head cancels out, producing

350 - - - - - - -- - - - - - --

3001----1-------4---1---4--1

1
E
0
"':,;
w
w
z 1501--~---.-1----1-------1---1
iii
IIC
,..
:::, YELL7TA1L

9
9:AZEAU
e L.A GRAN DE 4

e GA ANO COULEE
RAT T LING BROOK
1001----'--,i__ _ _-4-_ _ _ _4-....j

90 '-----'--..4----'--'---'-....L.-'--'-....J--'--'-'-'
,9 1.5

G/2.9 d

FIG. 5.-Relationship between Generator Casing Size, Turbine Throa t Diameter, and
Head

the following formula which is a function of capacity only:

V, = 17(N + ~) MW1. 2
s . . . . . . • . . . . . . . . . • . . . . . • . . (13)

II tn order to determine whether this formula was accurate, it was decided to

,plot unit capacity as indicated in Table 5, against unit concrete volume as in
Fig. 6, from which the following minimum volume formula has been derived:

l
V, = 46( N + ~ ) MW091 (14)

Il
. •• . • . • .... • . . . .•.• . ••• . .

Equation 13 is also plotted on F ig. 6, from which it will be seen that Eq.

l
l
l'
I
l 252 DECEMBER 1981 EY2
13 produces a maximum power house substructure concrete volume.
Equation 14 can be compared with the Creager data (3) which indicates that
for heads in excess of I IO m, the substructure unit concrete volume is about
0.05 cu yd/hp which can be expressed as:
V. = 51 MW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (15)
Equation 15 is also shown in Fig. 6, from which substructure concrete volumes
based on Eq. 14 will be about 27% lower at 10 MW capacity, increasing to
about 40% lower at 100 MW capacity when compared with the Creager data.
It must be emphasized that Eqs. 10 and 13 or 14 will only give a preliminary
estimate of the power house concrete volume, and their use should, therefore,
be confined to prefeasibility work. They eliminate dimensions G and din favor

t•
:i;
10'

'lu• 17Mw 1, 25
e GLEH C,t;NYON

> SPflAY 1 •
>-
u
,.
.,
C.

0
;-
%
::,
,o

/ . .·-·-·· · ,,, I
/'- I
// Vu•51M w
/ I

3
POWERHOUSE UNIT CONCREiE VOLUME m .-.,..

FIG. 6.-Relationship between Power House Concrete Volume and Generator Ca•
pacity

of power and head, and, since the turbine throat velocity can vary between
about 7.5 and 9.8 m/s, substantial errors can be introduced. Moreover, these
formulas do not take into account the effect of extra generator inertia on casing
diameter, which can be significant.
Finally, the formulas can be used to determine the "power house layout
efficiency," a measure of the effective use of concrete in the substructure.' )
Assuming that there are no adverse foundation conditions present, the layout
efficiency can be determined by dividing the theoretical minimum concrete volume
as given by Eqs. 5 or 7, by the estimated unit concrete volume. For example,
the layout efficiency of the La Grande 3 power house can be calculated from
Eq. 5 as 8,208 divided by 8,790 or 93 %. Bearing in mind that Eq. 5 gives
EY2 CONCRETE VOLUMES 253
a minimum volume, and requirements for storage space, access to equipment,
etc., a layout efficiency in excess of about 80% can be regarded as the best
attainable.

CONCLUSIONS

An estimate of 1rummum power house concrete volumes can be obtained

from Eq. 5 when the ratio of generator casing size, G, to turbine throat diameter,
d, is less than about 2.9 . For larger values of this ratio, use Eqs. 7 or 9.
Adverse geological and foundation conditions may have a significant effect on
concrete volumes, increasing these by a factor of up to 4 or 5. For prefeasibility
estimates, an approximation of power house concrete volume can be obtained
from Eqs. 10 or 13 or 14. Equation 10 should be used where the turbine head
is less than 110 m, and Eq. 13 or 14 where the turbine bead is over 110 m.

·: 4.CKNOWlEDGMENTS

The author wishes to thank the British Columbia Hydro and Power Authority
for data on the Peace Canyon, Seven Mile, Kootenay Canal, and Whatshan
power houses; the U.S. Department of the Interior for data on Grand Coulee,
Glen Canyon, and Yellowtail power houses; Rousseau Sauve Warren, Inc. for
data on the La Grande 4 power house; Ontario Hydro for data on the Wells
power house; and Hydro Quebec for data on the Outardes 2 power house.

APPENDIX !.-REFERENCES

I. Allas, L. , and Smith, E. R., "The Coteau Creek Project," Transactions, Canadian
Electrical Association, Part 2, Paper 66-H- 67, 1966.
2. Arthur, H. G., "The 820,000 Horsepower Turbines for Grand Coulee," Water Power,
Vol. 25, No. 2, Dec.
3. Creager, W. P., and Justin, J. D., "Hydroelectric Handbook," 2nd ed., John Wiley
and Sons, Inc. , New York, N.Y., 1950, p. 776.
4. Duck, D. J ., "Construction of the Grand Coulee Third Power Plant," Journal of
the Construction Division, Vol. 101, No. CO4, Proc. Paper 11774, Dec., 1975, p.
869-881.
5. Gordon, J. L., "Installing Hydro Units in Restricted Spaces,'' International Waler
and Power and Dam Construction, Vol. 29, No. 12, Dec. , 1977, p . 57.
6. Gordon, J. L. , "Estimating Hydro Power House Crane Capacity," lnternalional Warer
Power and Dam Construction, Vol. 30, No. II, Nov., 1978, p. 25.
7. New Zealand Electricity Dept., "Aviemore Power Station," Water Power, Vol. 20,
No. 9, Sept., 1968, p. 365.
8. Newbury, F . E., and Hutt, R ., "Small Hydro Still Economic," Transactions, Canadian
Electrical Association, Vol. 18, Part 2, 1979.
9. Taylor, E. G., "Unit Spacing of Hydroelectric Machines,'' Water Po wer, Vol. 21,
No. 10, Oct., 1969, p. 377 .
. JO. Thicke, R. H., and Bakar, L. J., "Plant Layout and Turbine Design for the Bighorn
' Power Project, " Transactions, Canadian Electrical Association, Vol. 11, Part 2, 1972.
11. USBR Specification, Glen Canyon Dam and Powerplant, Vol. I , No. DC-4825.
12. USBR Technical Record of Design and Construction, Davis Dam and Powerplant.
13. USBR Technical Record Design and Construction, Boysen Dam and Powerplant.
14. Walsh, 0 . E., "The Pelton Hydroelectric Development," Portland General Electric
Company, Portland, Oreg.
15. Walker, J. H., "Project Data for Hydroelectric Generators," Water Power, Vol.
19, Mar., 1966, p. 103, and Apr. , 1966, p. 143.
254 DECEMBER 1981 EY2
APPENDIX 11.-NOTATlON

The following symbols are used in this paper:

d turbine throat diameter, in meters;

G generator casing size measured to outside diameter or to outside
concrete casing surface, in meters;
h rated turbine head, in meters;
MW,kW generator rating, in Megawatts and kilowatts;
N number of units in a power house;
R = erection bay length, in meters;
s distance between unit center lines, in meters;
v, total power house concrete, in cubic meters; and
v. concrete volume per unit, in cubic meters.
 l
I
I
I f

;-- - - - - - - - - - - - - - - - - - - - ------ - - --- - - - - --- -•'.

)
1
i
16728 CONCRETE IN HYDROELECTRIC POWERHOUSE
KEY WORDS: Capacity; Casings; Concrete;
Hydroelectric powerplants; Power senes; Turbines
Design; Estimates;
:
'
:
~ I

!, ABSTRACT: The volume of concrete in a hydroelectric powerhouse varies by a iarge :

i amount, depending mostly on the size of turbine, size of generator, and the foundation :
:,
,.1, ' conditions, An analysis based on data from 40 powerplan ts ranging in unit size from
, 0,5 Mw up to 600 Mw, and operating under heads ranging from 12,2 m up to 273 m is
presented. Formulae are developed for estimating concrete volume for outdoor
powerhouses containing vertical shaft reaction turbines with steel spiral casings based
spiral on turbine throat diameter for heads less than about 110 m, and on generator
casing diameter for heads above 110 m, Formulae for use in preliminary estimates are ),
also developed based on unit capacity and head, where head is below 100 m, and on }
I
unit capacity only where head is above 110 m. All formulae include a factor for the
erection bay size,
REFERENCE: Gordon, James L (Vice Pres, and Hydro., Montreal E ngineering Co,,
Ltd., P,O, Box 6088, Station A, Montreal, Canada H3C 3Z8), " Medium Head Hydro
Power House Concrete Volumes," Journal of the Energy Division, ASCE, YoL 107,
/ No, EY2, Proc. Paper 16728, December, 1981, pp, 237-254
t