Hydraulic Model Studies of Canyon Ferry Dam CX

Spillway Stilling Basin

REC-ERC-74-27
7 . AUTHORISI 8. PERFORMING ORGANIZATION
REPORT NO.
P. H. Burg1
9 PERFORMING 5RGANIZATION NAME A N D ADDRESS ID WORK UNIT NO.

Bureau of Reclamation I 7 CONTRACT OR G R A N T NO. '-

Engineering and Research Center
Denver, Colorado
2 . SPONSORINC AGENCY N A M E A N D ADDRESS
- I3 T Y P E O F R E P O R T A N D PERIOD
COVERED

I
I
I#.SPOkSORING A G E N C Y CODE

5. SUPPLEMENTARY NOTES
-
1 .
.., i-r
\ \
6 . ABSTRACT

Hydraulic model studies were made to determine the cause and recommend b.s_i*$n tor the deposit anc
movement of riverbea material into the Canyon Ferry Dam spillway stilling b&. Tests indicated thal
movement of riverbed material into the basin resulted from operation of the river outlet works at discharge!
greater than 3,000 ft'ls (85 m'lsl. Several solutions to the problem are suggested, including a limitation or
operation of the river outlets to 3.000 ft3/s. Model tests estimated the spillway discharge and length of time
required to clean the existing.900yd3 (688 m3) of riverbed material from the basin. Soundings taken-a1
Canyon Ferry Dam immediately after the suggested spillway release confirmed the model results.

7. K E V WORDS A N D DOCUMENT A N A L Y S I S
. DESCRIPTORS-- / *hydrau!ic models/ hydraulics/outlet works/ 'stilling basin/spillways/ scour/ deposition
'damages/ fluid flow1 riverbeds
.-
:, .,

gs
. I D E N T I F I E R S - - /Canyon Ferry Dam. MT
.. .
. lhl
COSATl ~ i e l d / ~ r o u p
8. DISTRIBUTION S T A T E M E N T 119. S E C U R I T Y C L A S S 21. NO. C F P A G f
I T H E REPORT)
voiioble from the Notion01 Technical Informotion Service. operotions
i v i r i m . Springfield. Virginia 22151. UNCI.ASS1FIED 39
20. S E C U R I T Y C L A S S 22. PRICE
,. , I T H I S ?AGE)
! UNCLASSIFIED
HYDRAULIC MODEL STUDIES
OF CANYON FERRY DAM
SPlLLWAY STILLING BASIN

BY
P. HI. Burgi

-4

December 1974

UNITED STATES DEPARTMENT OF THE INTERIOR * BUREAU O F RECLAMATION

I ACKNOWLEDGMENT I

I Branch, and E.
,draulics Branch
~ h Chief.
. Design
R . Anesi. C h ~ e f .Reservoir Requlation Branch of the
I
 ....
Recommendations for Release o f Surplus Water . . . . . . . . . . . . . . . . 1

Releases Based on Results of Model S:udies . . . . . . . . . . . . . . . . 1

Alternate Methods of Release . . . . . . . . . . . . . . . . . . . . . 1
.
Application . . . . . . . . . . . . . . . . :. . . . . . . . . . . . . 2
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
The Model . . . . . . : :. . . . . . . . . . . . . . . . . . . . . . 2

Description . . . . . . . . . . . . . . . . . . . . . . . . ... ... . . . . 2

Scale Relations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Verification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
.
The Investigation . . . . . .. . . . . . . . . . . . . . . . . . . . . . . 3

Effect of Flow Release Methuds on Riverbed Stability . . . . . . . . . . . . . 3

River outlet works . . . . . . . . . . . . . . . . . . . . . . . . 3

Spillway . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Powerplant and Helena Valley Pumping Plant . . . . . . . . . . . . . 5
Combined river outlet and spillway releases . . . . . . . . . . . . . . . 5

Riverbed Stability . . . . . . . . . . . . . . . . . . . . . . . . . 6

Natural channel . . . . . . . . . . . . . . . . . . . . .. .~...... 6

Modifications . . . . . . . . . . . . . . . . . . . . . . . . . . 6

Spill Required to Clean Stilling Basin. . . . . . . . . . . . . . . . . . . 7

Appendix-Log of Model Tests . . . . . . . . . . . . . . . . . . . . . . 37

LIST OF FIGURES

Figure
1:48 Scale model layout of Canyon Ferry Dam . . . . . . . . . . . . 8
. . . . . . .
Comparison of settling velocities. Canyon Ferry model study 9
Canyon Ferry Dam stilling basin rounding. August 1970 . .
. . . . . . .
.
Model verification t e n 5 . . . . . . . . . . . . . . . . . . . .
. . . . . . .
Canyon Ferry Dam stilling basin sounding. September 1971
10
11
12
Results of river outlet releases. tests 1 and 7 . . . . . .
. . . . . . . 13
Results of river outlet releases. tests 8 and 9 . . . . . .
. . . . . . . 14
. . . . . . . . . . . . . . . .
Result of river outlet release. test 14 15
Result of river outlet release with debris initially in the
basin. test 16 . . . . . . . . . . . . . . . . . . . . . . . 16
10 lsovels-river outlets Q = 9.500 ft3/s . . . . . . . . . . . . . . . . 17
11 . .
Flow patterns resulting from river outlet releases . . . . . . . . . 18
12 .......
Isovels-river outlets Q = 3.000 ft3/s ......... 19
13 .
Asymmetrical operation of two river outlets. test 13 . . . . . . . . . 20
14 .
Symm?trcal operatlon of two rlver outlets. test 15 . . . . . . . . . 21
. . . . . . . . . . . . . . . . .

.-
15 Result of spillway release. test 2 22
16 Result of small spillway releases. tests 24 and 25 . . . . . . . . . . . 23
. . . . . . . . . . . . . . .
17
18
Result of large spillway release. test 3
.
. of asymmetrical spillway operations. tests 26 27. and
Results
28.-. .I... . . . . . . r' . . . . . . . . . ..........
24

25
19 Basin celiter wall pressures with asym~:-$;al spillway operation . . . .- 2 6 .
20 Results of powerplant and Helena Vallev Pumpins Plant operations . +
tests 6A and 6B . . . . . . . . . . . . . . . . . . . . . . 27
21 Results of simultaneous spillway and river outiet operations.
tests 34 and 38 . . . . . . . . . . . . . . . . . . . . . . . 28
22 Stability of downst. earn river channel. tests 20 and 21 . . . . . . . . . 29
23 .
Modified downstream river channel . . . . . . . . . . . . . . . 30
24 Downstream river channel cleared to bedrock. tests 29. 30. and
31 . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
25 ~ n sill . . . . . . . . . . . . . . .
d wail stud~es.testi32. 33. and 36 33
\\

26
11. and 12
. .
Spill required to clean stilling basin of 400 yd3 tests 10
. . . . . . . . . . . . . . . . . . . . . . . . 34 e-7:
-1

27 . . . . . . . . . . . .
Removal of riverbed material from stilling basin 35
28 . . . . . . . . . . . .
Water surface elevations below dam during spill 36 .
 PURPOSE uptake. This simultaneous operation will also prel!ent
the movement of riverbed material into the stilling
These studies were coiiducted to determine the cause basin. T o prevent riverbed erosion downstream from
and recommend a solution for the movement o f the stilling bisin, this simultaneous operation should be
riverbed material into theCanyon Ferry Dam spillway limited t o a totaiirelease not t o exceed 10,000 ft3/s
stilling basin. Of primary importance were the (283 m3/s) over the spillway and 4.750 ft3/s (135
determination of riverbed stability imm~diately m3/s1 through the river outlets. Observation of the
downstream from the stilling basin and the effect of simultaneous operation i s requiredon the prototype to
flow release methods on,the movement of this riverbed insure that the spillway nappe docs not intersect with
material. Studies wereaiso conducted to determine the the river outlet jets, which could result in cavitation
time and amount of spillway release required to clean damage to ths spillway surface. Analysis of the
deposited material from the stilling basin. dissolved gas uptake will also..have to be performed on
!,'
the prototype.

RESULTS RECOhlhlENDATIONS FOR RELEASE

/
OF SURPLUS WATER ,<,'
1. When the river 'outlet releases exceed 3,000 ft31s /
,
:.,,
(85 m3/s). riverbed-materials move into the spillway: Releases Based on Results of Model Studies
stilling 5asin. ::'
;<.
-;. <; 1.3.000 ft3/s or less.-~ake the total release from
2. Releases f r o m the spillway.~owerplant, or the riv?r outlets, equally distributed through all four
Helena Valley Pumping Plant do not carry riverbed outlets.
material into the stilling basin.
2. More than 3,000ft3/s.-Make the totai release from
3. Model tests mdicated that the deposited riverbed the spillway, equally distributed through a l l four
matertal could be cleared from the stilling basin with spillway gates: :.;. ,.
.
adequate spillway releases. The tlme and amount of -
.- 5
spill requ~redto clear this materlal from the still~ng Alternate Methods of Release ;
,
, ---:::.
basm can be determined from figure 27. 'i
TO minimize gas supersaturation in the rivtr ''\

4. The spilling technique developed in the model downstream from Canyon Ferry am,
simultaneous
s.*

successfully cleared approximately 900 yd3 (688 m31 operation of the outside river outlets and the two
of riverbed material from the Canyon Ferry spillway center spillway gates i s recommended for releases
stilling basin. greater than 3.000 ft3/s.
ir
5. Clearing the river bottom of loose riverbed material 1. 3,000 fr3/s or less.-Make the total release from
down to bedrock for a d~stanceof 100 feet (30 m) river outlets, equally distributed through all four
downstream from the end sill will prevent the outlets.
movement of riverbed material into the basin for river
outlet releases up to the design discharge of 9,500 ft3/s 2. More than a 0 0 0 ft3hand less than 9,500
(269 m3/s). f?/s.-~ake equal releases from the two outside
river outlets and the two inside spillway gates.
6. Operation of two river outlets in either a
symmetrical or asymmetrical pattern did not produce 3. More than 9,500 ft" and less than 14,750
as good a flow distribution in the stilling basin as fP/s.-2,750 ft3/s from each of the two outside
operating all four outlets uniformly. river outlets and 50 percent of the remainder
through each of the two inside spillway gates.
7. Uniform operation of all four spillway gates gave
best spilling results. Acceptable asymmetiical spillway 4. More than 74,750 ft3/s.-~ake the total release
releases through gates 2.3. and 4 were too small, 4,000 from spillway, equally distributed through all four
ft3/s (113 mvs) or less, to be considered as an spillway gates.
alternate spillway release method.
5. When making releases simultaneously from both
8. The simultaneous operation of the two center the river outlets and the spillway, be sure the water
spillway gates (No. 2 and 3) with the two outside river released from the spillway does not intersect the jet
outlets (No. 1 and 4) may minimize the dissolved gas of the water released from the river outlets. I f it
 appears this i s about t o happen, close the two river
outlets and make the total release equally through
each of the four spillway gates. In 1974 hydraulic model studies were requested by the
Upper Missouri Regional Director to:
6. Periodic soundings immediately downstream
from the basin should be made after spiliway (a) Determine the cause (what release method) for
discharges have exceeded 6,000 ft3/s !170 m3/s) in movement of rlverbed material into the spillway
ordei t o monitor any erosion of the river bottom stllling basin,
which may occur i n the area, due t o the
simultaneous operation. ( b l Determine if the existing riverbed, downstream
from the basin, had stabilized. and

APPLICATION (c) I f not, what would be required t o stabilize the

rlverbed or otherwise prevent the riverbed material
A method of cleaning fiverbed material from the from enterlng the spillway stilling basin.
stilling basin, as determined from the mad4 studies,
has been successfully applied a t Canyon Ferry Dam.
Other results related .to release methods to prevent, THE MODEL
riverbed material from entering the spillway stilling
basin can be applie'd as a guide in future operations at Description
Canyon Ferry Dam.
The model, constructed to a scale of 1:48, included
190 feet (58 ml of the upstream resewoir, the dam,
INTRODUCTION powerplant, Helena Valley Pumping Plant, and 450
feet (137 m) of the downstream river channel. figure 1.
Canyon Ferry Dam i s a feature of the Pick6loan To properly model the various releases from the
Missouri Basin Program and is located 17 miles (27 km) reservoir, provision for controlled releases from the
northeast of Helena, Montana, on the Missouri River. spillway, river outlets, powerplant, and pumping plant
The dam i s a concrete gravity-type structure were included in the model. Each control was
approximately 1,000 feet (305 m) i n length with a calibrated before the test program started. A tailgate
maximum height of 225 feet (68.6 m l above the assembly and sand trap were used t o control the
foundation. The powerplant. on the right side of the downstream tailwater elevation and collect eroded sand
dam. is rated at 50.000 kilowatts. The dam was and gravel. Water was supplied to themodel through
constructed in the period 1949 through 1954. the permanent laboratory sys?.em. U~schargeswere
measured by one of a bank of venturi meters installed
Construction of the cofferdam for the Helena Valley in the laboratory. ,:
Pumping Plant, which i s immediately downstream from
the spillway stilling basin, was started i n May 1957. By Scale Relations
June, the cofferdam was three-fourths complete when
large releases, required t o pass reservoir inflows, T o express the mathematical relationship between the
washed away a major part of ihe earth cofferdam. The hydraulic quantities of the model and the prototype,
cofferdam was reconstructed, and severe. cutting and the Froude Law of model similitude was applied. This
remova! of sand and gravel occurred again during the law isexpressed in equation (1). Hydraulic similitude i s
1958 releases. established when this equality is satisfied.

Soundings have been taken of the stilling basin and of

the river channel immediately downstream from the
basin periodically since 1960. The soundings indicate
that a considerable amount of riverbed material was
carried into the stilling basin. I n 1972 over 17,000yd3
(13.000 m31 of material were removed from the basin
with a 3-yd3 (2.3-m3) clam bucket momted on a
platform barge. The material was carried t.2 a disposal V = velociw
site approximately 1 mile (1.6 km) downstream from g = gravitational acceleration
the dam. Soundings taken i n July 1973 indicated that L = linear dimension, and
approximately 900 yd3 (688 m 3 ) of r:~verbedmaterial
subscripts p and m denote prototype and model. away from the retaining walls and upstream ontc. the
respectively. The scale ratio is denoted by L,, which slop~ngapron of the basln. Compare figure 4a with
means the ratio of the linear dimensions in the model f~gure4b arid f~gure3 w ~ t h
f~gure5.
to corresponding dimensions of the prototype. L, =
Lm During the course of the study, the ~ e ~ i o n i l G i r e c t o r
-.The scale relations, based on the Froude law, can
LP requested that model tests include a study of spillway
be expressed in terms of L, as shown below. releases required to clear approximately 900 yd3 (688
m3I of $and, gravel, and rock fragments from the basin.
Dimension Scale Ratio The tests predicted that a spill of 28,200 ft3/s (799
rn3/s) for approximately 3 hours would sweep the
Length Lr = 1:48 hasin clear of this riverbed material. Shortly after these
Area (L,)' = 1:2,304 model,tests, the suggested release was made at Canyon
Velocity (L,)"~ = 1:6.93 Ferry barn and soundings taken immediately after the
Discharge (L,15" = 1: 15,963 spill verified this model test.
Time (L,)"~ = 1:6.93

The movement of riverbed materials was of special

THE INVESTIGATION
interest in this sludv. To properly represent the
Effect of Flow Release Methods
prototype conditions in the model, determination of
on Riverbed Stability
the size, specific gravity, and in turn the settling
velocity of the model riverbed material was very
Release of surplus water through the Canyon Ferry
important.
Dam spillway and river outlets has resulted in the
deposits of riverbed material in the spillway stilling
Using the Froude Model Law, the settling velocity iatio
basin. The movement of this material has eroded the
of the model to prototype riverbed material was
concrete floor of the basin. The areas of greatest
(L,)li2 or 1:6.93. Figure 2 relates the settling velocity
concrete erosion, up to 1.5 feet (0.46 m), have
of the model and pototype riverbed material. The top
occurred on the sloping apron of the stilling basin.
30 percent of the model material i s very close to the
figure 3. The model was tested to determine the effect :
scaled settling velocity of the prototype material. The
of flow release metho* on t h e movement and
lower 70 percent of the model material i s lighter than
deposition of riverbed material.
the prototype material.
River outlet works.-The river outlet works consist of
Verification four 86-inch (218-cm) diameter, horseshoe.shaped
conduits placed horizontally through the sp~llway
The scaling technique used to model the riverbed section, which exit on the face of the spillway chute at
material indicated the rnodel sand selected for the an invert elevation of 3649.91 (1112.5 m). Each
study adequately represented the prototype sands and conduit has a 77-inch (195.6-cm) high-pressure
gravels. regulating gate. The design discharge for the fpur river
outlets is 9,500 ft3/s (269 m3/s). Although ttic outlet
To verify the Canyon Ferry rnodel, the August 1970 conduits are symmetrical with the stilling basin center
sounding of the prototype stilling basin and wall, they are not centered in the stilling basin bays,
downstream river channel (fig. 3). was modeled (fig. figure 3.
4a). The significant rlver outlet and spillway releases
were applied to the model on a compressed time scale A series of tests. 1. 7, 8. 9, and 14, with releases of
representing the 1971 releases from the prototype, test 9,500, 4.000. 5.000, 6.000, and 3.000 ft3/s (269, 113,
5. (See Appendix for test conditions.) A ~najorportion 142, 170, and 85 m3/sl. respectively, was conducted
of the 1971 releases involved the river outlet works. for time intervals representing 43 hours each in the
prototype. Figures 6, 7, and 8 illustrate the
The results of the test (fig. 4bl were compared with the downstream erosion and deposit of riverbed material in
September 1971 sounding of the prototvpe stilling the basin after each test.
basin and downstream river channel (fig. 51. Although
the test results did not fully duplicate the 1971 Tests 1, 7, and 14 initially had clean stilling basins.
sounding, the configuration of the relocated material in exceptfor the grouted,gravel in the downstream left
the model was similar t o that in the prototype. In both corner of the stilling basin. The initial river bottom
the rnodel and prototype, the 1970 deposition moved configuration for these tests was similar to figure 8.
,.,-
. ... ..,.
Initial conditions for tests 8 and 9 were the conditions
resulting from t e s t r 7 and 8, figures 6b and 7a.
respectively.
~~, - - - ~ ~

Erosion of the downstream river channel and from the river outlets. The design capacity of 2,375
deposition of the eroded material on the sloping apron ft3/s (67 rn3/s) was released through outlets No. 3 and
of the stilling basin increased as the river outlet 4 for a total dischargecif 4.750 ft3/s (135 m3/s), figure
discharge increased from 3,000 to 9,500 ft3/s. The test 13. In comparing figure 13b with figure 6a, test 1, i t i s
results clearly indicate that Operation of the river apparent that the erosion is more severe near the
outlets can carry large amounts of riverbed material pumping plant and at the sheet piling with the
into the stilling basin. River outlet releases limited to asymmetrical operation. Although a comparable
3.000 f t 3 / n o r less result in very little movement of amount of material appeared to be carried into the
riverbed material into the stilling basin, figure 8. 'basin' on the left side, it was not a l l carried onto the
sloping apron as in test 1. Approximately one-third
Test 16 was conducted to determine the effect of a remained in the downstream portion of the basin.
3.000.ft3/s river outlet release on the movement of
riverbed material already deposited in the downstream : Test 15 (fig. 141 compared the operation of outlets No.
section of the stilling basin. An initial discharge of 1 and 4, releasing a total discharge of 4.750 ft3/s, with
7,900ft3/s (224 m3/s) through the river outlets carried test 1, releasing the design discharge of 9,500 ft3/s
material into the basin. The spillway gates were then through four outlets, and with t e s t 8, releasing a total
opened releasing 15.000 ft3/s (425 m3/s), which discharge of 5,000 ft3/s through four outlets. As
moved the accumulated material into the down4ream expected, because of the smaller discharge, both the,
section of the basin, figure 9a. The river outlets were downstream erosion and the amount of material
then operated at a reduced flow of 3,000 ft3/s, which carried into the basin were less in test 15 than in test 1.
produced the deposition shown in figure 9b. In figure 6a. However, test 8 showed less downstream
mmparing figure 9b with figure 8, i t is evident that erosion than test 15; compare figures 7a and 14. The
more riverbed material will be carried upstream when total release was approximately the same for these two
there i s an initial deposit in t i l e lower section of the tests, but test 8 used all four river outlets, while test 15
basin as compared to an initial clean basin. used only outlets No. 1 and 4. :

Velocity measurements in the model determined the These tests indicate that all four river outlets should be
direction of flow (in or out) and the velocity a t three operated uniformly to achieve the best flow
half-sections in the stilling basin for the 9.500 ft3/s distr~butionpossible.
river outlet release, figure 10. The flow pattern i s
assumed symmetrical about the basin centerline. As Spillway-It was noted early in the testing program
indicated earlier by the erosion patterns, a strong that small spillway releases would not weep the basin
undercurrent moves upstream in the basin. A t the sill floor clean. Test 2, with a spillway relisse of 9,400
section, Station 4+10, the core of the upstream curre& ft3/s (266 m31s), did not clear the riverbed material
lies on the sill approximately 75 feet (23 ml from the initially in the basin, figure 15.
training wall. The core risesfrom elevation 3605 at the
sill to 3623 at Station 2+74. The core velocity of the The results of test 2 indicated the wssibility that
undercurrent increases from 5 ftls (1.5 m/s) a t the sill. riverbed material might be c a r ~ i e d i nthe
t ~ basin at low
to 7.5 ft/s (2.3 m/sl at Station 2+74. spillway releases. Since th'bre was some material
initially in the basin for t e s t 2, it was difficult to
Figure 11 illustrates the general flow pattern in the determin? whether the material present at the end of
basin. The jet leaving the river outlet conduit stays in the test W ~ S , in fact, carried in. Tests 24 and 25, with
the upper 20 fee: (6.1 ml of the basin depth and does spillway releases of 4,100 and 6.600 f t v s (116 and
not penetrate to the floor of the basin. A large 187 m3/s), respectively, were conducted to clear up
longitudinal eddy i s established in the vertical plane, the question. Figure 16 illustrates no significant
providing the means for carrying riverbed material into movement of material into the basin at low spillway
the basin. releases. However, material initially present in the basin
will be exposed t o secondary currents and will
Similar velocity measurements were also made at the continue to erode the concrete floor when spillway
sill. Station 4+10, for a river outlet release of 3,000 releases are not large enough to sweep the basin clean;
ft3/s (85 rn3/s), figure 12. The core of the figure 15b.
undercurrent was located about 95 feet (29 m) from
Test 3 (fig. 17) was conducted with a spillway release
of 24.000 ft3/s (680 mvs) using the results of test 2
(fig. 15) as the initial conditioc. As figure 17 did not affect the movement of riverbed material ihor
illustiates, the basin was swept clear of riverbed near the stilling basin. Figure 20 illustrates the results
material. Minor erosion occurred in the channel of Tests 6A and 66.
immediately downstream from the stilling basin end
sill. The spill required to clean varlous amounts of Combined river outlet and spillmy releases.-Model
material from the basin, will be covered in a later tests indicated that the movement of riverbed material
sectton of this report. can be controlled by limiting the four river outlets to a
total release of 3.000 ft3/s. When larger releases are
Because of the spray associated with spillway operation ;quired, the river outlet works should be closed and
and the resulting maintenance required on the all release's made over the spillway. ,.
electrical equipment located on top of the powerplant, i;

it was requested that ihe model spillway be operated With the recent interest'!in the effect of gas
asymmetrically using gates 2, 3, and 4. with gate 1. supersaturation on fish life in the Columbia River,
closest to the powerplant, closed. Tests 26, 27, and 28 spillways with relatively deep stilling basins have .-
were conducted operating gates 2, 3, and 4 uniformly become suspect. Water released over the spillway-
with total spills a t 14.500,8,300, and 3$0O ft3/s (411, carries large quantities of air deep into the stilling
235, and 110 m3/s), respectively. During these tests, basin. The hydrostatic pressure in the basin forces gas
e d f t 3 / s (170 m3/s). The
the p ~ w e r ~ l a ~ t r e l e a s6.000 :?to solution, resulting in supersaturatedwater. Fish
larger spills of 14,500 and 8,300 ft"s~esulted in severe swimming in these waters take in dissolved gases
erosion downstream from the end sill, figures 16a and through their gills and in turn these gases are
18b. The 3,900 ft3/s spill caused an insignificant transported into the body tissue by the bloodstream.
amount of erosion downstream from the basin, figure Gas-bubble disease results v:hen the fish swim into
18c. The only acceptable asymmetrical spillway release waters of lower pressure, where the dissolved gas
of 3,900 ft3/s was too small to be considered as an returns to i t s gaseous state.
alternate method of spillway release. It is
recommended that a l l four spillway gates be operated The Canyon Ferry river outlets discharge across the
uniformly. water surface of the stilling basin in contrast to the
deep plunging-type discharge of the spillway. With
The center wall in the stilling basin wasequipped with respect to supersaturated water, the Canyon Ferry river
six piezometers to determine the pressure on the wall outlets pro!!ide a more acceptable release method than
during the asymmetrical spillway operation of t e s t s 26. the spillway.
27, and 28. Figure 19 shows the piezometnr locations
and also indicates the average, and '.'maximum In anticipation of future field tests to determine the
instantaneous differential heads on the wall for each effect of release methods on gas supersaturation, a
test. The maximum instantaneous differential head laboratory test: was conducted using the left river
occurred in test 26 a t piezometer No::P<where the outlet to determine the effect of such a field test on
pressure differential was 6.0feet (1.83 m). riverbed stability. The test represented a .%-hourfield
test where the left (No. 4) river outlet would be
Powerplant and Helena Valley Pumping Plant.-Test opened in 10-minute intervals and held constant for
6A determined the effect of the powerplant operation 20-minute intervals for releases of 590, 1,180. 1.770,
on the movement of riverbed mater~alnear and in the and 2,380 ft3/s (17, 33, 50, and 67,m3/s). During the
stilling basin. Before the start of the test some very fine field test, saturometer measurement%vill be recorded
material was observed on the stilling basm floor. The for the various outlet openings during the 20-minute
powerplant discharged 6.000 ft3/s for approximately hold intervals. The laboratory test indicated that the
31 hours (prototype time scale). There was no amount of debris carried into the basin would be
movement of the fine materiel initially present on the minimal, less than that shown in figure 8.
floor of the basin over this time span. Test 6B was an
extension of test 6A with the powerplant discharging Simultaneous operation of the two outside river outlets
6,000 ft3/s and the Helena Valley Pumping Plant (No. 1 and 4) and the two center spillway gates (No. 2
turbines releasing 463 f t v s 113 m3/s) into the and 3) should result in less gas supersaturation than
downstream channel for a timo span representing 10 spillway-only operation and also will result in less
hours in the prototype. Again, there was no indication movement OF riverbed material into the basin than
of any movement of the fine material on the stilling rivernutletsnnly operation. Tests 34 and 38 were
basin floor. A l o c ~scour
l hole and buildup occurred in
-
-
conducted releasing 4.750 ft3/s (135 m3/s) through
Test 3 (fig. 17) was conducted with a spillway release
of 24,000 ft3/s 1680 m3/s) using the results of test 2
(fig. 15) as the initial condition. As figure 17 did not affect the movement of riverbed material ireor
illustrates, the bas~n was swept clear of riverbed near the stilling basin. Figure 20 illustrates the results
material. Minor erosion occurred in the channel of Tests 6A and 6B.
immediately downsiream from the stilling basin end
sill. The spill required to clean various amount* of Combined river outlet and spillway releases.-Model
material frcm the basin, will be covered in a later tests indicated that the movement of riverbed material
section of this report. can be controlled by limiting the four river outlets to a
total release of 3,000 ft3/s. When larger releases are
Secause of the spray asraciated with spillway cperation required, the river outlet works should be closed and
and the resulting maintenance required"'on the all releases made over the spillway.
electrical equipment located on top of the powerplant,
it was requested that tne model rnillway be operated With the recent interest in the effect of gas
asymmetrically using gates 2, 3, and 4, with gate 1, supersaturation on fish life in the Columbia River,
closest to the powerplant, closed. Tests 26.27, and 28 spillways with ralatively deep stilling basins have
were conducted operating gates 2, 3, and 4 uniformly become suspect. Water released over the spillway
with total spills at 14,500, Rz300,and 3,900 ft3/s (411, carries large quantities of air deep into the stilling
235, and 110 m3/sl, respecti:&ly. During these..rcsts, basin. The hydrostatic pressure in the basin forces gas
the powerplant. released 6,000,.ft3/s (170 m"i;i: The into solution, resulting in supersaturated water. Fish
larger spills of 14,500 and 8,300 ft3/s resulted,ii?.sev&e swimming in these waters take in dissolved gases
erosion downrtrearn from the end sill, figurei5i:Sa and through their gills and in turn these gases are
?3h. The 3,900 ft3/s spill caused an in&nificsnt transported into the body tissue by the bloodstream.
amount of erosion downstream from the basin, figure Gas-bubble disease results when the fish swim into
18c. The only acceptable asymmetrical spillway release waters of lower pressure, where the dissolved gas
of 3,900 ft3/s was too small to.he considered as an returns to itsgaseous state.
alternate method of spillway release. I t i s
recommended that all four spillway gates be operated The Canyon Ferry river outlets discharge across the
uniformly. water surface of the stilling basin in contrast to the
' deep plunging-type discharge of the spillway. With
The center wall in the stilling basin was equipped with respect to supersaturated water, the Canyon Ferry river
six piezometcrs to determine the pressure on the wall outlets provlde a more acceptable release method than
during the asymmetrical spillway operation of tests 26, the ?pillway.
27. and 28. Figure 19 shows the piezometer locations
and also indicates the average and maximum In anticipation of future field tests to deterrnine the ,,
instantaneous differential heads on the wall for each effect of release methods on gas supersaturation, a
test. The maximum instantaneous differential head labdiatory test was conducted using the left river
occurred ir: t e s t 26 at piezometer No. 6, where the outlet to determine the effect of such a field test on
pressure differential was 6.0feet 11.83 m). riverbed stability. The test represented a 2-hour field
test where the left (No. 4) river outlet would be
Powerplan t and Helena Valley Pumping Plan t.-Test owned in 10-minute intervals and held constant for
6A determined the effect of the powerplant operation 20-miflute intervals for releases of 590. 1,180, 1,770,
on the movement of riverbed material near and in the and.2.380 ft3/s 117, 33, 50. and 67 mvs). During the
stilling basin. Before the start of the test some very fine , saturometer measurements will be recorded
material was observed on the stilling,basin floor. The arkus outlet openings during the 20-minute
powerplant discharged 6.000 ft3/s for approximately rvals. The laboratory t e s t indicated that the
.,.
31 hours (prototype time scale). There' was no t of debris carried into the 'basin would be^
movement of the fine material initially present on the al, less than that Shown in figure 8.
floor of the basin over this time span. Test 6B wasar;
extension of test 6A with the powsrpl+nt dischzrging Simultaneous operation of the two outside river outlets
6.000 ft3/s and the Helena Valley Pumping Plant (No. 1 and 4) and the two center spillway gates (No. 2
turbines releasing 463 ft3/s (13 m3/s) in:? the and 3) should result in less gas supersaturation than
downstream channel for a time span representing 10 spillway-unly operation and also will result in less
hours in the prototype. Again, there was no indication movement of riverbed material into the basin than
of any movement of the fine material on the stilling
basin floor. A local scour hole and buildup occurred in
the two outside river outlet conduits and 4.750 (test
34) and 10.000 ft3/s (283 m3/s) (test 381, through the
two center spillway gate;. Figure 21 illustrates the continued abrasion damage and high costs for material
success of the simultaneous operation controlling removal. As an alternativg, the downstream river
the riverbed movement. bottom could be artificially stabilized with concrete
grout or bituminousgrout.
Two possible adverse conditions which should be
monitored a t the prototype structure are: (1) Modifications.-A series of three tests (No. 29. 30, end
cavitation damage to the spillway flow surface and (2) 31) was conducted to determine the distance
revere :.erosion of the riverbed. To prevent the downstream from the basin end sill whicii ~ o u l d need
formation of cavitation and the potential for concrete to be cleared to bedrock to eliminate the mJvement of
damage to tfie spillway flow surface, it i s essential that riverbsd material into the basin. For each test the
the spillway nappe and the river outlet jets do not riverbsd was, cleared of material to the simulated
intersect. Such interaction could result i n bedrock, the full width of the basin extending 10 feet
subatmospheric.pressures on the spillway flow surface (3 m) beyond the right training wall. The lengths of
leading to cavitation and possible concrete erosion. riverbed cleared to bedrock downstream from the basin
Figure 21b illustratesthe severe erosion of the riverbed end sill were 50, 75, and 100 feet 115, 23, and 30 m)
material to bedrock downstream from ths spillway for test5 29, 30. and 31. respectively: Downstream
stilling basin at the higher spillway release. Periodic from the cleared area, the invert sloped upward on a 4
soundings immediately downstream from the basin to 1 slope to the existing riverbed, figure 23. For each
would allow the project to monitor any erosion of the t e s t the discharge was 9,500 ft31s through the river
bedrock in this area. outlets and 8,000 ft3/s from the powerplant, for a time
period representing 43 hours in the prototype. These
Riverbed Srability tests indicated that the river bottom should be cleared
to bedrock for a distance of 100 feet downstream from
The term "riverbed stat;lity." as used in this report. the stilling basin end sill, figure 24. The kft riverbank
~~~~~~refer to the noticeable movement of riverbed between the stilling basin and the Helena Valley
material with timd, and,.particularly in dference to Pumping Plant would also require extensive
material moving upstreaminto the stilling basin. stabilization to preven$the bank from sloughing into
the excavated area. 1
One objectwe of the study was to determine if the river
channel downstream from the spillway stilling basin Studies were also conducted to determine the height
was stable. Tests 1, 7, 8, 9, and 14 indicated that the that would be required of a wall on top of the end s i l l
model riverbed wss not stable when the river outlets to prevent movement of riverbed material into the
released flows larger than 3,000 ft3/s (85 m3/s). In stilling basin. Tests 32, 33, and 3S were conducted
comparing figures 3 and 5, it is quite evident that the with the river outlets discharging 9,500 f t v s for time
prototype riverbed is also not stable. Therefore, studies intervals representing 43 hours each. Test 32 consisted
were conducted to determine what would be required of two !%foot (1.5-m) high. 20-foot (6.1-m) long walls,
to make the riverbed steble. centered 65 feet (19.8 m) from the training walls. The
walls in tests 33 and 36 were 8 feet (2.4 m) and 12 feet
Natural channel.-Plywood was placed in the model to (3.7 m) high, respectively, and extended the full width
represent the location of bedrock in the prototype of the basin. Figure 25 illustrates the results of each
river channel. Three tests were conducted, releasing test. The riverbed material moved around the walls in
9,500 ft3/r (269 m3/sJ for a time representing 42 test 32. In t e s t 33 the riverbed material accumulated
hours in the prototype. After each t e s t riverbed agair~stthe downstream face of the wall and then
material was removed from the stilling basin, measured, overtopped it. The 12-foot-high wall in test 36 was
and not returned to the model. The material carried sufficiently high to prevent river material from
into the basin was 1,300,700. and 530 yd3 (994.535, overtopping into the basin. The small amount of
and 405 m3) for tests 20. 21, and 22, respectively. material collected in the basin during the test is
Figure 22 illustrates the movement of riverbed material believed to have come in under the wall where it '
after tests 20 and 21. contacts the right training wall.

The tests indicated that the riverbed downstream from

the stilling basin will stabilize in time with repeated
operation of the river outlets and subsequent removal
have a detrimental effect on the stilling action of the basin. It was later verified by divers that the spill had
basin at larger spillway discharges. swept the basin clean of all material except that
material lodged under exposed reinforcement.
Spill Required to Clean Stilling Basin

Since spillway or river outlet releases are necessary

approximately 6 out of every 7 years at Canyon Ferry
Dam, the idea of sweeping future deposits of material
out of the basin by spillway releases was considered.
Tests 10. 11. and 12 were conducted to determine the
minimum spillway release required to sweep the basin
clean with approximately 400 yd3 (306 m3) of
mater~aldeposited on the sloping apron. Figure 7b
illustrates the approximate init~alcondition for these
tests. Figure 26 illustrates locations of the deposited
riverbed rnater~alafter spiilway operation at discharges
of 11,000. 20,800, and 24,400 ft3/s (312. 589, and
691 m3/s), tests 10, 11, and 12. respectively. The
24,400 ft31s spill essentially cleaned the basin in 2
hours as compared to 6 hours for the 20.800 ft3/s.

Tests 17. 18, and 19 were conducted to determine the

spillway discharge and time requlred to clear the basin
of 1,000 vd3 (765 m3) of debris. The 24.300, 28,200,
and 30.200 ft3/s (688, 799. and 855 m31sl spills
cleaned the basin in 6, 3, and 2.112 hours, respectively.
These tests verif~edthe fact that the larger releases
cleared the basin of debris with less time and total
water than the small releases. The curves in figure 27
illustrate the relationship of the variables, spillway
discharge (01,time of spill (TI, and volume of
deposited material to be removed from stilling basin
(Vm).

The Upper Missouri Regional Office used the results of

these tests to formulate an operating procedure to
remove approximately 900 yd3 (688 m31 of debris
from the Canyon Ferry Dam spillway stilling basin. On
May 8. 1974, the Canyon Ferry Dam spillway gates
were opened in increments of 5,000 ft3/s (142 mvsl
over a 10-minute period and held at each increment for
20 minutes until a total spillway discharge of 28,200
ft3/s was achieved. The spillway discharged for 3 hours
at 28,200 ft3/s and then the gates were closed over an
18-minute interval. During the spill, the powerplant
re leases continued unchanged, discharging an
additional 5T900 ft3/s.(167 m3/s) into the downstream
channel. The Helena Valley Pumping Plant turbine also
operated to prevent riverbed materials from being
washed into the turbine draft tubes. The peak
discharge into the river was about 34,500 ft3/s (977
m3/s) for 3 hours. Four staff gages monitored the river
level below Canyon Ferry Dam. The changes in stage at
the gaging stations are shown in figure 28. Soundings
taken on May 9 and 10 indicated that the spill was
successful in flushing the 900 yd3 of material from the
 Spillway radial gates

Helena Valley Pumping Plant

Figure 1 . 1:48 Scale model lavoul of Canyon Ferry Durn. Photo

P296-D.75729
 STATION 2 + 74 1
E

:I

i
0' /
-
-,
,,
2 5' 50'
STATION 3 +81
75' 100'

, -.$

I I 6ND SILL I 1
0' 2 5' 5 0' 7 5' 100' 1
STATION 4 + 10

Approximate A r e a o f Upstream F l o w

Figure 10. Irovels-river outlets Q = 9,500 ft31s.

.
-
3600 I -
2
.,
\
I I i
0' 2 5' 5 0' 7 5' 10 0'
STATION 4 + 10
A p p r o x i m a t e A r e a o f Upstream Flow

Flgure 12. Irovels-raw, outlets Q = 3.000 ft3h.

a . Vlcw o l borm and downstream rlvcr channel before tcrt 2. Photo P296-D-75709

b. V l e v o f basin and downrtruom river chonncl after tcrl 2. Pholo PZLJS-D.7511U

Figure 1 7 . Result of large s p i l l w o y rclcarc, test 3.

24
a. Rerulcr of test 26. gates 2, 3. and
4. total 0 = 14,500 fr3/r. Photo
P296.0.75740
1 GATE
*4
1
GATE
*3
I
GATE
*2

11

SECTION A-A

P I E Z O M E T E R LOCATIONS I N CENTER WALL

D I F F E R E N T I A L PRESSURE R E A D I N G S (feet o f H

Flours 19. Basin center well pressures wlth asymmetrical spillway operation.
c l e x c U to bedrock for 100 i t , river
outler 0 = 9,500 f i J l r . Campor~teof
photos P296.D-75748 and -75749

Figure 24. Icominucd).

32
 28,200 FT%S
,-----
1---
- 15,000 FT%
---7'x-L 10,000 FT3/s
r- 5,000 FT'/S
I
I
I 2 3

TIME IN HOURS
Figure 28. Water surfam elevations below dam during spill.

36
 APPENDIX

Log of Model Terw

I

comments

1 6.25
2 6.25
3 GOO
4 A 3.00
B 3.50
5 A 8.00
6 4.00
C 4.00

6 A 4.50
6 1.50
7 6.25
8 6.25
9 6.25
10 2.75
11 2.26
12 1.31
13 6.56
14 6.25
15 5.75
16 A 4.W
6 1.60
C 6.17
17 1.20
18 1.20
19 1.20
20 6.25
21 6.00
22 6.00
23
24 6.25
25 6.00
26 3.00
27 3.00
26 3.00
29 6.25 Riverbed cleared ro bedrock 50 fee, down.
Itredm from barn
30 6.25 Riverbed rtaared to bcdrock 75 feeidawn-
siream from barn
31 6.25 Riverbed cleared to bedr&k 100 feeldown-
sfream from bam
32 6.25 Two 5.foof.high. 20-foot.long walls on lop
of end rill
33 6.25 8-fool-highwall on i r p of end rill, exrended
full width of barin
34 6.25 jimulianeoul operalion
35 2.17 Initial condition. 1.9W ydz debris in barin
36 6.25 12.fool-highwall on rooof end rill, extended
full width of basin
37 5.67 12faol-high wall on topoie,>drill. wrwnded
full width of basin
36 6.25 5~muImne0~1 weration
 ...................
7-1750 13.71)

CONVERSION FACTORS-BRITISH TO METRIC UNITS OF MEASUREMENT

The following convetrion facmrr adopted b y The Bureau o f Reclamation are thore published by the American
Society for Testing and Materials IASTM Metric Practice Guide. E 380-681 except that additional f s R o n 11.
c o m m o n l ~used in the Bureau have been added. Further dlreurrlon o f definitions of quantiliesand units ir given in
the ASTM Metric Practice Guide.

The metric u n i u and conversion factors adapted by the ASTM are bared on the "International System of Unirr"
lderisnred Si for Syrteme International d'Uniterl, fixed by the International CommiUes for Weighs and
Measurer: this ryrtem is alro known a the Giorgi or MKSA Imerer-kilogram Imasl-second-amperel r{rtem. This
ryrtem has been adopt4 by the International Organization for Sandardimrion in IS0 Recommendation R.31.

The metric technical unir of force is the kilogram-force; thir 1% the force whirs .when applicd to a bcdy hming J
m a s of 1 kg, giver it an aceleration of 9.80665 mlreclrec. the nandard aceel m o f free fall toward the earth's
center for sea level at 45 deg latitude. The metric unit of force i n 51 units I. newton IN), shich ir defined sr
that force whish. when applied t o a body having a mar5 of 1 kg, giver i: an m ltion d 1 mhedrec There unitr
must be distinguished from thc (inconstant) local w i g h t of a body having a i of 1 kg, that is. the weight of a
body is that fome with which a body ir attracted t o the earth and is equal tort, m a s of a body multiplied by the
acceleration due to gravity. Howewer, because it is general practice to u w "pcund" rather than the technically
correct term "pound.force." the term "kiiogram" (or derived mars unit1 has been u w d in thir guide instead of
"kilogram.force" i n exprerriog the conversion factors for forcer. The newton unit o f force will find increasing use.
and is euential in SI unitr.

Where approximate or nominal Englih unitr are u w d to exprcrr a value or range o f valuer. the converted metric
;/ w i l l i n parentheros are also approximate or nominal. Wheie precise English units are used, the convened metric
units are expressed as equaily significant valuer.
/i
Table I

QUANTITIES AND UNITS OF SPACE

Multiply BY TO o b t a i n

LENGTH

Mil . . ..............." 25.4 (exactlyi . . . . . . . . . . . . . . . . . . . . . . Micron

............... 25.4lexactlyl . . . . . . . . . . . . . . . . . . . Millimeters

. Inches
lncher
Feet
Feet'.
...............
................
...............
Feet . . . . . . . . . . . . . . . .
2.54 Iexactlyl. . . . . . . . . . . . . . . . . . . Ceodmeterr
30.48 lexactlyl . . . . . . . . . . . . . . . . . . CcnIimcterr
0.3048 lexactlvl' . . . . . . . . . . . . . . . . . . ? Meterr
0.0003048 1exactlyl' . . . . . . . . . . . . . . Kilome:crr
Yards . . . . . . . . . . . . . . . 0.9144 lexactlyl .................... Meters
Miles lrwtutel .......... 1,809,344 lexesllyl' ..................... Meters
Mfler ................ 1.609344 lexacclyl ............... Kilometers

AREA

;
Square incher
Square feet
...........
............
Squarefeet . . . . . . . . . . . .
Square yards ...........
6.4516lcxactlyl
'929.03
0.092903
.............
....................
....................
Square se8:tin~etsir
Square cenrimeten
Square meter.
0.836127 . . . . . . . . . . . . . . . . . . . . Square meters
.....
Acrer ................ .0,4M69 ........................ Hectares
Acrer . . . . . . . . . . . . . . . . .4.048.9 ........................ Square meterr
Acres . . . . . . . . . . . . . . . . '0.0040469 . . . . . . . . . . . . . . . . Square kilometers
Square miles ........... 2.58999 . . . . . . . . . . . . . . . . . .
- Square kilometerr

Cubic Inches ........... 16.3871 . . . . . . .,f. . . . . . . . . . . Cubic centimeters

Cubicf~et ............. 0.0283163 . . . . . . . . . . . . . . . . . . . . .Cubic meters
Cubic yards ............ 0.764555 .............
. . . . .'i. Cubic me:m

CAPACITY

Fluid ounces 1U.S.I .......

Fluid ounces 1U.S.I .......
Llquld p l n n lU.S.1 ........ 0.473179 . . . . . . . . . . . . . . . . . . Cubic decimeters
Liquid pints I U S.1 ........ 0.473166 . . . . . . . . . . . . . . . . . . . . . . . . Liters
sum: 1u.s.1 : . . . . . . . . . . '946.358 ................... Cubiscantimeterr
a u s n s iu.s.i . . . . . . . . . . . '0.946331 . . . . . . . . . . . . . . . . . . . . . . . . Liters
Gallons1U.S.I . . . . . . . . . . . -3,785.43 . . . . . . . . . . . . . . . . . . . . Cubic centimrterr
Gallons 1U.S.I . . . . . . . . . . . 3.78543 . . . . . . . . . . . . . . . . . . . C u b i ~d ~ i m e t e r r
Gallons lU.S.1 . . . . . . . . . . . 3.78533. ........................ Literr
Gallons 1U.S.l . . . . . . .: . . . .0.00378543 .................... Cubic meters
.-.\. .GallonrlU.K.~ . . . . . . . . . . 4.54803 . . . . . . .:. . . . . . . .:'. . Cubic decimeters
,'I Gallons 1U.K.I . . . . . . . . . . 4.54596 . . . . . . . . . . . . . . . . . . . . . . . . .Liters
I'
Cubic f ~ e t ............. 28.3160.. ........................ Liters
Cubic yards . . :: .......... '764.55 ........................... Lltarr :C
Acrefeet ............. '1,233.5 ........................ Cubic meters :'
Acrefeet ............. '1.233.500 ............................. Literr .;
.,,.