Report of stilling basin

Chapter one
Design of stilling basin
Scope :

Energy dissipation at dams and weirs is closely associated with spillway

design, particularly with the chosen specific discharge q, the difference
between the upstream and downstream water levels (H*) and the down-
stream conditions. dealt mainly with the actual spillway inlet works and
certain standard types of conduits conveying the flow from the spillway
inlet, i.e. chutes, tunnels, etc. In this chapter the main concern is the
concept of energy dissipation during the whole passage of the flow from
the reservoir to the tailwater and, in particular, the stilling basin (energy
dissipator) design.
The magnitude of energy that must be dissipated at high dams with
large spillway discharges is enormous. For example, the maximum
energy to be dissipated at the Tarbela dam service and auxiliary spillways
could be 40 000 MW, which is about 20 times the planned generating
capacity at the site (Locher and Hsu,1984).
In the design of energy dissipation, environmental factors have to be
considered; some of the most important ones are the effect of dissolved
gases supersaturation on fish in deep plunge pools, and of spray from flip
bucket jets which can result in landslides and freezing fog.
The passage of water from a reservoir into the downstream reach
involves a whole number of hydraulic phenomena such as the transition
into supercritical flow, supercritical non-aerated and aerated flow on the
spillway, possibly flow through a free-falling jet, entry into the stilling
basin with a transition from supercritical to subcritical flow, and echoes
of macroturbulence after the transition into the stream beyond the basin
or plunge pool. It is, therefore, best to consider the energy dissipation
process in five separate stages, some of which may be combined or
absent (Novak

andCˇábelka,1981)
 Definition sketch for the five phases of energy dissipation
1. on the spillwaysurface;
2. in a free-fallingjet;
3. at impact into the downstreampool;
4. in the stillingbasin;
5. at the outflow into theriver.
Energydissipationonspillways

Energydissipationonthespillwaysurface

The energy head loss on the spillway surface can be expressed as

e = $aV'2/2g (1.1)

where V' is the (supercritical) velocity at the end of the spillway, a is the
Coriolis coefficient and $ is the head loss coefficient, related to the velocity
coefficient ! (the ratio of actual to theoretical velocity) by

1/!2 = 1 + $. (1.2)

The ratio of the energy loss, e, to the total energy E (i.e. the relative energy
loss)
 e $V'2 V2 $V'2 $
——=—
E
ƒ(
—+— =—=1—!2.
2g 2g 1+ $
) (1.3)

2g
For a ratio of the height S of the spillway crest above its toe (or in case of a
spillway with a free-falling jet as in Fig. 1.1 (S–S') above the take-off point)
and the overfall head H, with S/H < S — S'
30 (or spillways (Novak and Cˇ — < 30), and for smooth
ábelka, 1981),

H
!1 = 1 — 0.0155S/H. (1.4)

For a given S, !1 increases as H increases, i.e. if for a given discharge Q the

spillway width b decreases and thus q increases (equation (4.19)). Thus, for
S/H = 5, !1 = 0.92 and the relative head loss is 15%, whereas for S/H =
25, !1 = 0.61 and the loss is 62%.
The value of $ could be increased (and ! decreased) by using a rough
spillway or by placing baffles on the spillway surface. However, unless aer-
ation is provided at these protrusions, the increased energy dissipation
may be achieved only by providing an opportunity for cavitation damage
(Section 4.6).
Stepped spillways may provide an opportunity for additional energy
dissipation (when compared with smooth spillways) pending on the value
of the unit discharge (q) (see also Section 4.7.6). E.g. Rice and Kadavy
(1994) compared (using models) the energy loss on a smooth and stepped
spillway for a 17 m high dam; the result agreed broadly with equation (1.4)
for the smooth spillway and showed a 2–3 times higher energy loss for the
stepped alternative (steps 0.61 m high and 1.52 m deep). Stephenson (1991)
confirms the importance of the unit discharge on the efficiency of energy
dissipation on cascade spillways and Boes and Hager (2003) give a detailed
analysis of the friction factor for the skimming flow regime demonstrating
the effect of the chute slope (friction factor decreases with the slope) which
is much larger than that of the relative roughness (see also equation (4.68)).

Ski-jumpspillways

In many modern spillway designs, increased energy dissipation is

achieved by using free-falling jets, either at the end of a ‘ski-jump’ or
downstream of a flip bucket (Figs 1.2 and 1.3).
The ski-jump spillway was first used by Coyne (1951), and was later
further developed by detailed model studies. Its use brings substantial
Fig. 1.2 Comparison of (a) normal spillway, (b) ski-jump spillway and (c) ski-
jumpspillwaywithjetsplitters(NovakandCˇ ábelka,1981)

Flip bucket

economies where geological and morphological conditions are favourable, and

particularly where the spillway can be placed over the power station

oratleastoverthebottomoutletworks(NovakandCˇábelka,1981).
The head loss in the jet itself, whether solid or (more frequently) dis-
integrated, is not very substantial – only about up to 12% (Hor ˇení, 1956). The
energy loss on a ski-jump spillway can be substantially enlarged by splitting the
overfall jet into several streams (see Fig. 1.2) or by using two spillways with
colliding jets.
A substantial benefit for energy dissipation from jet spillways is in the
third phase at impact into the downstream pool. Here most of the energy losses
in the first three phases occur through the collision of masses of water, and
through the compression of air bubbles, both those con- tained in the overfall jet
as well as those drawn into the downstream pool at the point of impact. The
decrease in energy in this phase may be, there- fore, enhanced by having a
dispersed and intensively aerated jet before impact.
 Thecombinedenergylossinthefirstthreephasesofenergydissipa-
tioncanbeexpressedfromavelocitycoefficient!1–3,whichcanbedeter-mined
in model tests from the theoretical supercritical flow depth conjunctive
to the subcritical depth needed to form a stable jump down-
streamofa(ski-jump)spillway.Evenifthisvalueof!maybesubjectto
scaleeffects(theprototype!islikelytobesmallerbecauseofincreased
aeration), the model studies give a very good indication of the relative
merits of variousdesigns.
Generally (Fig. 1.1),

!1–3=f(S'/S,q,geometry) (1.5)

where S' is the height of the ‘take-off’ point above the reference
datum.
A comparison of !1–3 for three designs is shown in Fig. 1.2:
curve a, a normal spillway ending in a stilling basin; curve b, a ski-
jump spillway without baffles at the take-off edge; curve c, a spillway
with baffles which split the jet so that the air entrainment and energy
dissipation is enhanced and the pressures on the stilling basin floor are
reduced. The increase of ! (and decrease of relative energy loss) with
q is again demonstrated. By using a suitable design, the values of !
have been reduced throughout by a factor about 0.7; this results in
substantial energy losses, e.g. for !1–3 = 0.5, e/E = 75% (equation
(1.3)). The optimum (lowest) value of !1–3 is attained for S'/S = 0.6;
at this value the overall nappe gains a sufficient velocity and degree of
turbulence while flowing over the upper part of the overall spill- way
to disperse effectively on the baffles at the take-off edge, and has a
sufficiently long free fall through air to aerate intensively and to break
up the jet core (Novak and Cˇ ábelka, 1981).
The disintegration of a falling circular jet of diameter D was
studied by Ervine and Falvey (1987), who showed that a complete
decay of the solid inner core occurs after a length of fall L, with L/D
in the range 50–100. For flat jets, which are more relevant to spillway
design, Horˇení (1956) established experimentally that the length of
fall from the crest required for total jet disintegration (for q in m2 s—
1) is
L=1.89q0.319=6(q)1/3(m). (1.6)

Flipbucket

The flip bucket (Fig. 1.3) is a version of a ski-jump spillway that

is usually used as an end to a chute or tunnel spillway whenever the
geological and topographical conditions are suitable. Flip buckets (just
as ski-jump spill- ways) are usually tailor-made for a given project,
and the designs are developed with the aid of scale models
 The key parameters for the flip bucket design are the approach flow
velocity and depth, the radius r of the bucket, and the lip angle þ. For a two-
dimensional circular bucket the pressure head can be computed for irrotational
flow; experimental data confirm these values for the maximum pressure head
but (in contrast to the theory) show a non-uniform pressure distribution (Fig.
1.3). At low flow the bucket acts like a stilling basin with water flowing over
the lip and the downstream face; the foundation of the flip bucket has, therefore,
to be protected against erosion. As the flow increases a ‘sweep-out’ discharge is
attained at which point the flip bucket starts to operate properly with a jet.
Thejettrajectoryishardlyaffectedbyairresistanceforvelocitiesbelow20ms—
1
,butforvelocitiesof40ms—
1
thethrowdistancecanbereducedbyasmuchas30%fromthetheoreticalvalue,givenb
y(v2/g)sin2þ.
The designer’s main concern is usually to have the impact zone as far as
possible from the bucket to protect the structure against retrogressive erosion.
Many designs with skew jets and various three-dimensional forms
offlipbucketshavebeendeveloped.Helleretal.(2005)giveananalysisof ski-jump
hydraulics and Locher and Hsu (1984) discuss further the flip bucketdesign.

Stillingbasins

Hydraulicjumpstillingbasin

Thestillingbasinisthemostcommonformofenergydissipatorconverting the
supercritical flow from the spillway into subcritical flow compatible with the
downstream river régime. The straightforward – and often best – method of
achieving this transition is through a simple submerged jump formed in a
rectangular cross-section stilling basin. Vischer and Hager
(1995)giveanoverviewofthehydraulicsofvariousenergydissipators.
Hydraulic jumps have been investigated by many researchers, more
recently by Rajaratnam (1967) and Hager, Bremen and Kawagoshi (1990),
whoalsoextendedthisinvestigationtoajumpwithacontrolsill(Hager and
Li,1992).Theimplicationsofthehydraulicsofthejumpforthe submerged
jumpstillingbasinhavebeenstudiedbyNovak(1955).
Referring to the notation in Fig. 1.4 and to equations (1.1) and (1.2) we
can write

aq2
E=y+ 1
—. (1.7)
2g!y221
y1 q21/2
[ (
y2= —— —1+ 1+8—3 .
2 gy1
)] (1.8)
Fig. 1.4 Definition sketch for hydraulic jump stilling basin

The stilling basin depth is then given by

y'=y'2—y0=o'y2—y0 (1.9)

and the length by

L=K(y2—y1) (1.10)

whereo'andKarecoefficients(derivedfromlaboratoryandfieldexperi-
ments).
When applying equations (1.7)–(1.10) we start with a known dis-
charge q and the corresponding downstream depth y0. For a suitably
chosen !(Section 1.2) and a value of E corresponding to the total energy
available above the stilling basin floor, y1 can be computed from equation
(1.7), y2 from equation (1.8) and y' from equation (1.9) (from a chosen
value of safety coefficient). E is, of course, initially not known and thus it
is best to apply the above procedure by iteration, initially assumingy'= 0,
i.e. taking the energy datum at the downstream river-bed level. This com-
putation, carried out for several discharges, can produce five alternatives:

1. y2>y0throughouttherangeofq;
2. y2=y0throughouttherangeofq;
3. y2<y0throughouttherangeofq;
4. y2>y0onlyathighdischarges;
5. y2>y0onlyatlowdischarges.

Case 1 is the most frequent one, and shows that a stilling basin is
requiredforalldischargesinordertoproduceasubmergedjump.Forsafety
thesameisrequiredincase2(whichisreallyonlyatheoreticalpossibility).
Fory2<y0nostillingbasinisnecessary,andahorizontalapronprotecting
theriverbeddownstreamofthedamissufficient,asasubmergedjumpwill
resultnaturally.Thestillingbasindesignforcase4hastobebasedonthe
maximum discharge (the same as for case 1) and, for case 5, on the dis-
chargegivingamaximumdifferencebetweeny2andy0(withQd<Qmax);this
canresultinasmallstillingbasinatthetoeofthedamfollowedbyahori-
zontalapron(orviceversa),orinaslopingaprondesign.
Where the result of the first computation shows that a stilling basin is
required, the procedure is repeated for a new value of E (in equation(1.7))
which takes into account the lowering of the energy datum by a suf- ficient
amount (see Worked example1.1).
Thevaluesofthecoefficientso'andKinequations(1.9)and(1.10)
canbetaken(NovakandCˇábelka,1981)as1.1<o' <1.25and
4.5 <K <1.5, where the lower value of K applies for Fr1>10 and the
higherforFr1“3.BecauseatlowsupercriticalFroudenumbersthejump
isnotwelldevelopedandcanbeunstable,itisratherdifficulttodesignan
economicallydimensionedbasininthesecaseswithoutmodelstudies.
Equations (1.8) and (1.10), and thus the design under discussion, apply
to basins with a horizontal floor only. In sloping channels the value of y2/y1
increases with the slope; for a slope S0=0.2, y2/y1 is twice the value of a
horizontal channel, with the same Froudenumber.
The quoted values of K, and particularly of o', are fairly low (as dic-
tated by economy) and dependent on a good assessment of the coefficient
!,andparticularlyofthedownstreamdepthy0which,inturn,dependsusuallyona
nassumedvalueofManning’snfortheriver.Ifconservative valuesofnand!
areassumed(i.e.lownandhigh!)thenasmallvalueof
o'(say,1.1)issufficient,otherwiseahighervaluemayhavetobeselected.
Itisalsoveryimportanttoassessthepossiblelong-termriver-beddegra- dation
downstream of the dam, which could result in a lowering of the
downstreamwaterlevelsandofy0.
A simple end sill with a 1 in 3 slope is usually as good as more com-
plicated sills (Section 1.3.3).
Itisevidentfromequations(1.7)–(1.10)thatthelowerthevalueof!
(thehigherthevalueof$)thesmallerwillbetherequiredstillingbasin;! in
equation (1.7) refers to the total losses between the spillway crestand
entryintothestillingbasin,i.e.to!1–3(equation(1.5)).
The energy loss in the fourth and fifth phases of energy dissipation
(Section 1.1) can be expressed as
e4,5=(y2—y1)3/4y2y1. (1.11)
Downstream of the jump at the outflow from the basin there is still a
substantial proportion of excess energy left, mainly due to the high
turbulenceofflowwhichcanbeexpressed(NovakandCˇábelka,1981)as
e5=(a'—a)V20/2g (1.12)
 wherea'istheincreasedvalueoftheCorioliscoefficientreflectingthe
highdegreeofturbulenceandunevenvelocitydistribution;2<a'<5for
3<Fr1<10,whilea=1.
From equations (1.11) and (1.12) we obtain
e4 —=1—4(a'—1)
—=1— e5 1+(1+8Fr
1 ) .
2 1/2
(1.13)
——— 21/23
e e
4,5 [—3 +(1 + 8Fr1) ]
4,
5
Equation (1.13) shows that the efficiency of energy dissipation in the
jump itself within the stilling basin decreases with the Froude number,
leaving up to 50% of the energy to be dissipated downstream of the basin
at low Froude numbers (Section1.3.3).
The hydraulic jump entrains a substantial amount of air additional
toanyincomingaeratedflow.Aconstantairconcentrationthroughout
thejump(C¯1=C¯2)resultsinalowerheightofthejumpthanforthe
casewithoutair,whileforC¯1>0andC¯2=0(whichisamorerealistic
assumption) a slightly higher y2 is needed than for no air (Naudascher,
1987). Thus the main significance of the presence of air in the jump
region
istherequirementofhigherstillingbasinsidewallsduetothehigherdepth of
flow (equation (4.33)). The effect of air entrainment by hydraulic jumps
onoxygenconcentrationintheflowisbrieflydiscussedinSection9.1.7.
The highly turbulent nature of the flow in the hydraulic jump
induces large pressure fluctuations on the side walls and particularly on
the floor of the basin which, in turn, could lead to cavitation. Using a
cavitation number,o,intheform(¯p¯'
1 ¯) /(1/2pV2)
2 1/2

(equation(4.14)),wherep'isthedeviation of the instantaneous pressure p

from the time-averaged pressure
p¯(p=f(t)canbeobtainedfrompressuretransducerrecords),therelation- ship
between o and x/y1 (where x is the distance from the toe of the jump) for a
free and submerged jump at Fr1= 5 is shown in Fig. 1.5(Narayanan,
1980; Locher and Hsu, 1984). Assuming the length of the jump to be
approximately6(y2—y1),thehydrostaticpressure atthepointofmaximum
pressurefluctuations,i.e.inafreejumpatx/y1=12,willbepgy¯,with
ȳ y 12y = 3y .
=
y
+ 2
—
y
1
1 —— 1 1
6(y2— y1)

For o = 0.05, cavitation will occur if

p+pgy¯—0.05kp V2
=p=0
1
(1.14)
0 — v
2
wherek=p'/(¯p'¯2¯)1/2>1(1<k<5).
The value of k can be computed from equation (1.14) and assuming,
for example, a normal distribution of pressure fluctuations, the intermit-
tency factor, i.e. the proportion of time for which k is exceeded (the
probability of occurrence of cavitation) can be computed (worked
example
 Comparisonofstreamwisevariationofpressurefluctuationsin
freeandsubmergedjumpsforFr=5(afterLocherandHsu,1984)

1.1). For k > 5 there is practically no cavitation danger; for k = 3.5 there is a
2% probability of the occurrence of cavitation during the time of basin
operation (for well constructed basins). In reality, this probability of cavita-
tion could be appreciably higher because the inception of cavitation will
occur at pressures higher than pv, irregularities in the basin floor will cause
local pressure reduction, and the incoming flow upstream of the jump may
already have a substantial degree of turbulence (Narayanan, 1980).
Physical models provide, with certain safeguards, a suitable tool for
evaluating the amplitude and frequency characteristics of macroturbulent
pressure fluctuations and for assessing the tendency towards cavitation with
intermittent cavitation in prototype (Lopardo, 1988) (Chapter 16).
Potential cavitational damage is not the only danger in hydraulic jump
stilling basins (as well as in other types of basins). Probably the most
serious structural problem is the effect of uplift pressures due to the dam
drainage system or the tailwater level or the water table in the basin bank.
This pressure excess over the hydrostatic pressure in the basin is aggra-
vated by the macroturbulent pressure fluctuations underneath (and on the
side of) the jump. Although the pressure fluctuations have an unequal
spatial distribution (which alleviates this part of the problem) it is only
sensible to design the floor slab for the more severe of the two situations:
either the full downstream uplift pressure applied over the whole area of the
floor with the basin empty or the uplift pressure head equal to the root mean
square (r.m.s.)
value of pressure fluctuations of the order of 0.12V2/2g (V is the inlet
super-
1 1
critical velocity) applied under the whole full basin (this is half of the
maximum point pressure fluctuation resulting from o index max=0.05 and
kmax=5). Furthermore, all contraction joints should be sealed, no drain
open- ings should be provided, and floor slabs should be as large as possible
and connected by dowels and reinforcement (ICOLD, 1987
 Farhoudi and Narayanan (1991) investigated the forces acting on
slabs of different lengths and widths beneath a hydraulic jump giving
further details of effects of slab size, position and width–length ratio. Pin-
heiro, Quintela and Ramos (1994) give a summary and comparative
analy- sis of methodologies for computing the hydrodynamic forces
acting on hydraulic jump stilling basin slabs.
The prevention of vibration of basin elements (due to the turbulence
of the flow) also requires massive slabs, pinned to the foundation when
possible (see also Fiorotto and Salandin, 2000).
Abrasion of concrete in the basin could take place if this is also used
for bottom outlets carrying abrasive sediment (although this is unlikely to
happen for velocities below 10 ms—1, or from sediment drawn into the
basin from downstream either by bad design or operation. The basin
should be self-cleaning to flush out any trapped sediment.
Uplift, abrasion, and cavitation are, of course, closely connected,
and provision for maintenance and repairs should be considered in the
basin design.
The discharge used in stilling basin design is in most cases the spill-
way design (maximum) discharge (in cases 1, 2 and 4 above). This,
however, need not always be the case. Sometimes it may be more
econom- ical to take a calculated risk and design the basin for a smaller
and more frequently occurring discharge (say Q1000 or smaller instead of
PMF – Chapter 4) and carry out repairs should this chosen Q be
exceeded. Great care and experience is necessary when opting for this
alternative.

Other types of stillingbasins

Althoughthestillingbasinbasedpurelyonasimplehydraulicjumpworks
wellandrelativelyefficiently,incertainconditionsothertypesofbasins may
produce savings in construction costs. Standard basins were developed
with baffles, chute blocks and special end sills by the USBR (Bradley and
Peterka, 1957; Peterka, 1963; US Bureau of Reclamation,
1987).Anexampleofabasinwithchuteandbaffleblocks–USBRType III–
which canbe used forvelocitiesV<18.2ms—1 and q<18.6m2s—1is
showninFig.1.6.Asthisbasinisshorterthanothers,thetemptationisto use it
outside these limits; however, the danger of cavitation damage in these
cases is substantial and great care must be exercised in the design
andpositioningoftheblocks.Basco(1969)andNothaft(2003)carriedout
adetailedinvestigationofthetrendindesignofbaffledbasinsandofdrag forces,
pressure fluctuations, and optimum geometry; the whole area of
baffledbasinsisalsoreviewedbyLocherandHsu(1984).
The plain and slotted roller bucket dissipators developed mainly in the
USA (Peterka, 1963) (Fig. 1.7) require substantially higher tailwater
Stilling basin with chute blocks and baffles, USBR Type III (after US Bureau
of Reclamation, 1987)

Plain and slotted roller buckets (US Bureau of Reclamation, 1987)

levels than conventional hydraulic jump basins and, in the case of gated
spillways, symmetrical gates operation (to prevent side rollers whichcould
bringsedimentintothebucketwhich,inturn,coulddamagethedissipator).
The stilling basin with a surface regime hydraulic jump uses a sub-
mergedsmallshallowflipbucket(Fig.1.8(b));thetheoryanditsapplication
 Spatial hydraulic jump basins: (a) change in width; (b) change in depth;
(c) flow from lateral channels (after Locher and Hsu, 1984)
to low dams was developed particularly in the USSR (e.g. Skladnev, 1956)
andisreviewedbyNovakandCˇábelka(1981).Thistypeofbasinisreally only one
example of spatial hydraulic jump basins. Others involve a sudden change in
width (Fig. 1.8(a)) or a jump combined with side inflows form chutes (Fig.
1.8(c)).
Yasuda and Hager (1995) investigated the formation of a hydraulic jump
in a linearly contracting channel with the conclusion that although there was
substantial agreement with the classical jump (equation (1.8)) there were
significant differences in its structure. Bremen and Hager (1993) investigated
jumps in abruptly expanding channels with theconclu- sion that jumps with the
toe just in the upstream channel are more effi- cient than classical jumps, but in
stilling basin construction this advantage has to be offset against the fact that
they may become asymmetric for expansion ratios larger than 1.4 and – being
longer – require more excava- tion volume than basins based on the
classicaljump.
For gated barrages various types of stilling basins other than those already
discussed have been developed; one of the most common is a sloping apron
(glacis) followed by a horizontal sill with or without appur- tenances. The
shape of the basin is usually dependent on the morphology of the river bed and
the amount of excavation needed for its construction and its function has to be
considered together with various modes of gate operation. For further details
see Section 9.1.
All these and other types of stilling basins and energy dissipatorsare
bestdevelopedwiththeaidofscalemodelstudies(NovakandCˇábelka,
1981).
Mason (1982) carried out a survey of 370 dissipators constructed since
1950; the survey included rock basins (unprotected bed with sub-
mergedrollerbucket),simplejumpbasins,bafflebasinsandvarioustypes of
dissipators involving free trajectory jets. The results of the surveyare
summarized in Fig. 1.9, and confirm that to operate successfully baffle
basins need a certain minimum inflow velocity (head), but their rangeis
limitedbycavitationproblemsforH>30m.Simplehydraulicjumpshave
beenusedforheadsthatarelargerthanindicatedonthefigure(H>50m) but great
care is needed in design and construction. Various types of jet
dissipatorsneedaminimumhead(H>10m)toworkproperly.

Erosiondownstreamofstillingbasins

It has been demonstrated in Section 1.3.1 that at the outflow from the
basin there remains a certain proportion of energy to be dissipated. Because of
this and the uneven velocity distribution, there will always be some local
erosion downstream of the basin. To eliminate this is almost impossible and,
above all, uneconomical. The main purpose of the basinis
 12000
Key
simplehydraulic jumpbasins

Dischargecapacityofdissipator,Q(m3s—1)
10000
bafflebasins

8000

6000

4000
free trajectory jetdissipators

2000

0
0 20 40 60 80 100 120
Head drop from reservoir level to tailwater level, H* (m)

Preferredrangesofuseforthemaintypesofdissipaters(after Mason,1982)

to reduce and localize the scour (in a position where it can be

controlled and is not harmful to the dam), not to eliminate it. The end sill of
the basin has to be protected against retrogressive erosion and/or designed
in such a way as to encourage sediment accumulation against the sill rather
than erosion (Fig. 1.4).
The extent and depth of the local scour depend on hydraulic para-
meters, geological structure (erodibility index (Annandale, 1995)) and basin
geometry. Several methods are in use for its computation including model
studies, but perhaps the simplest is to estimate the scour depth as a
percentage of the depth which would occur at the foot of a free overfall
without a basin; this in turn can be computed from several equations
(Schoklitsch, Veronese, Jaeger, etc.). Using Jaeger’s form Novak (1955)
expresses the scour depth (downstream of hydraulic jump basins as deter-
mined from model experiments of relatively low head structures – say S <
10 m – with coarse sand and limited field observations) as

ys=0.55[6H0.25q*0.5(y/d )01/3—y]
90 0 (1.15)

whereys is the scour depth below the river bed (m), H* is the
difference between upstream and downstream water levels (m), y0 is the
tailwater depth (m), q is the specific discharge (m2 s—1), and d90 is the
90% grain size of sediment forming the river bed (mm). Equation (1.15)
thus indicates that the stilling basin reduces the potential scour by 45–50%.
For erosion downstream of an apron see e.g. Dey and Sarkar (2006).
Plungepools

Downstream of free-falling jets (jet, ski-jump spillways, flip buckets)

energy dissipation takes place in stilling basins, or more frequently, in
plunge pools, usually excavated fully or partially in the stream bed during
dam construction, but sometimes only scoured by the action of the jet itself.
The scour and protection works have been reviewed by, for example,
Hartung and Häusler (1973), and the equations of the maximum scour
y's(m) by Locher and Hsu (1984), Breusers and Raudkivi(1991), Annandale
(2006) and Liu (2005) (who considered both the energy dissi- pation
mechanism and submerged jet diffusion properties in the scour
development).
Bollaert and Schleiss (2005) using investigations of transient water
pressures in rock fissures developed a physically based model evaluating
the time evolution of scour in jointed rocks. Pagliaraet al. (2006) exam-
ined the effect of the densimetric Froude number, jet impact angle and air
content, tailwater elevation, sediment nonuniformity and upstream flow
Froude number on the plane plunge pool scour with the result that the
densimetric Froude number (involving average jet velocity and the particle
determining size) had the main influence on thescour.
Thegeneralformofthesimplifiedequationforthemaximumscour
depth,y's,measuredfromthetailwatersurface(y's=ys+y0)is

y's=CqxH
* þ /d
y w z
(1.16a)

whereCisacoefficient,þistheangleoftheflipbucketwiththehorizontal, d is the
particle size (mm) and q and H* are the same as in equation (1.15); the
range of the coefficient C and exponents x, y, zis

0.65 <C <4.7, 0.5 <x <0.67, 0.1 <y <0.5,

0 <z <0.3, 0 <w <0.1.

Thus the range of y'sis wide, as is only to be expected because equation

(1.16a)coversawiderangeofstructureswithdifferentdesigns,degreesof air
entrainment, and geologicalconditions.
Acceptingthepossibilityofsubstantial‘errors’,asimplifiedequation
byMartins(1975)withx=0.6,y=0.1,w=z=0andC=1.5canbeused:

y's=1.5q0.6H
*
0.1
(m) (1.16b)

(in equation (1.16b) H* is the drop from the reservoir level to the flip bucket
lip).
To reduce the variations in the powers in equation (1.16a) Mason
(1989) investigated the effect of air content in the plunge pool and
260 ENERGYDISSIPATION

proposed an equation (applicable to model and prototype) where air

entrainment replaces the effect of H *:

ys=3.39(q0.6(1+þ)0.3y0.16
0 )/(g d )
0.3 0.06
(1.16c)

whereþ=qa/q=0,13(1—vc/v)(H*/t)0.446;tisthejetthicknessandvcthe
minimumvelocityrequiredforairentrainment(1.1m/s).
Equation (1.16c) can be approximated by an earlier author’s equa-
t
i
o ys=3.27(q0.60H*0.050 y0.15)/(g0.3d0.10). (1.16d)
n

Lopardo (1997) proposed the equation

ys/H*=2.5(q/(gH3)0.5)0.5
* (m) (1.16e)

which translates to :

ys=1.41q0.5H0.25
* (m) (1.16f)

Equation (1.16e) is based on over 60 laboratory and 17 prototype results

and with a safety factor 1.3 covers all recent field data, e.g. from Tarbela
and several dams in China, Africa and South America; it can give higher
results than (1.16b), (1.16c) or (1.16d) (particularly when the safety
factor
1.3 is applied). Equations (1.16c–e) are dimensionally correct and all
equa- tions (1.16) confirm again the decisive influence of the unit
discharge in dam design (see also Sections 4.7.1 and 1.2.1).
For flip bucket spillways Tarajmovich proposed the equation
(1978):

ys=6ycrtanþ1 (1.17)

whereys=y's—y0(asinequation(1.15)),ycristhecriticaldepth,andþ1is
theupstreamangleofthescourhole,whichisafunctionoftheflipbucket
exitangleþbutdoesnotvarywidely(14°<þ1<24°for10°<þ<40°).
The pressure fluctuations on the floor of a plunge pool underneath
aplunging jet can be very considerable; it will be a function of plunge
length, pool depth, jet size and shape and can reach a substantial percent-
age of the head H* (Ervine, Falvey and Withers, 1997). If the plunge pool
is not deep enough to absorp the energy of the jet it may have to be pro-
vided with slabs forming a stilling basin floor (see Section 1.3.1). In this
case slab joints should be located outside the zone affected by the jet
deflection to avoid strong uplift forces in case of a joint waterstop
fracture or if using open joints (which may be advantageous with
effective founda- tion drainage). For further details of forces acting on
plunge pool slabsand their design see Meloet al.(2006).
Energydissipationatbottomoutlets

The flow from outlets occurs most frequently in a concentrated stream of

high velocity. The outlet may terminate below or above the downstream
water level, with or without an outlet regulating valve at its end (Sections
4.8 and 6.3). These variations in design are also reflected in the methods of
energy dissipation.
The two main design trends are either to disperse artificially and to
aerate the outflow jets (outflow above tailwater with or without control gate
at its end) or to reduce the specific discharge at entry into the stilling basin.
This basin may be a common one with the spillway – the best solu- tion
when feasible (e.g. Fig. 1.2) – or a separate one. The reduction in spe- cific
discharge for a high-velocity stream can be achieved either by depressing
the soffit of the outlet simultaneously with its widening, or by using blocks
and sills or guide walls just downstream of the outlet and before the entry
into the stilling basin, or in the basin itself, or by a combi- nation of the
various methods. The first method has the advantage of avoiding cavitation
and/or abrasion and is particularly effective when used in conjunction with
a spillway stilling basin because the stream from the
outletcanbesuitablydirectedintothebasin(Fig.4.22).
When submerged deflectors are used as, for example, in outflows from
tunnels, care must be taken in shaping the end of the deflector to avoid
cavitation. For further design details, see US Bureau of Reclamation
(1987)andNovakandCˇábelka(1981).
For the design of a gradually widening transition for the free super-
critical outflow from an outlet which terminates in a separate stilling basin,
Smith (1978), on the basis of work by Rouse (1961), recommends for an
initial width B0 (at the outlet), final width B1 (at the entry into a hydraulic
jump basin), and straight side walls diverging at an angle 0 from outflow
axis, the equations

B1=1.1Q1/2, (1.18)
B1 1/3
tan0= ( —1) ƒ(4.5+2Fr) (1.19)
— 0
B0

whereFr0=V0/(gB0)1/2forasquareconduitorFr0=V0(gD)1/2foracircular
conduit section. This rate of flare produces a reasonably uniform and
steadydistributionofflowatthestartofthejumpandalsopermitsacon-
tinuedexpansionoftheflowatthesameratethroughthestillingbasin;for
further details of the jump computation and basin design, see Smith
(1978).
For small-capacity outlets (5–10 m 3 s—1) under high heads, vertical
stilling wells provide a compact means of energy dissipation.
 Sudden expansion energy dissipators(Locher and Hsu, 1984) which
utilize the principle of energy loss at a sudden enlargement are a fairly
recent development. Although almost inevitably associated with cavita-
tion, this occurs away from the boundaries without undue danger to the
structure. As an example of this type of structure, the section of one of
the three circular cross-section expansion chambers of the New Don
Pedro dam, designed to pass a total of 200 m 3 s—1 at a gross head of 170
m, is shown in Fig. 1.10. The chambers dissipate about 45 m of head and
the remainder is dissipated in pipe resistance and the 9.14 m diameter
tunnel downstream of the gates.

Sudden expansion dissipator, New Don Pedro dam; dimen- sions in

metres (Locher and Hsu, 1984)
References
Annandale, G.W. (1995) Erodibility. Journal of Hydraulic Research, IAHR, 33, No. 4:471–94.
––––(2006)ScourTechnology:MechanicsandEngineeringPractice,McGraw-Hill, NewYork.
Basco, D.R. (1969) Trends in Baffled Hydraulic Jump Stilling Basin Design of the Corps of Engineers
since 1947, Miscellaneous Paper H-69-1, US ArmyEngi- neers Waterways Experimental
Station,Vicksburg.
Boes,R.M.andHager,W.H.(2003)Hydraulicdesignofsteppedspillways,Journal of Hydraulic Engineering
ASCE, 129, No. 9:671–
Bollaert, E.F.R. and Schleiss, A.J. (2005) Physically based model for evaluation of rock scour due to high-
velocity jet impact, Journal of Hydraulic Engineering ASCE, 131, No. 3: 153–61.
Bradley, J. and Peterka, A.J. (1957) The hydraulic design of stilling basins. Journal of the Hydraulics
Division, American Society of Civil Engineers, 83 (Hy 5), Papers 1401–6, 130 pp.
Bremen, R. and Hager, W.H. (1993) T-jump in abruptly expanding channel.
Journal of Hydraulic Research, 31 (1): 61–78.
Breusers, H.N.C. and Raudkivi, A.J. (1991) Scouring, IAHR Hydraulic Structures Design Manual, Vol.2,
Balkema, Rotterdam.
Coyne, A. (1951) Observation sur les déversoirs en saut de ski, in Transactions of the 4th Congress of
ICOLD, New Delhi, Vol. 2, Report 89: 737–56.
Dey, S. and Sakar, A. (2006) Scour downstream of an apron due to submerged horizontal jets, Journal of
Hydraulic Engineering ASCE, 132, No. 3: 246–57.
Ervine, D.A. and Falvey, H.T. (1987) Behaviour of turbulent water jets in the atmosphere and in plunge
pools. Proceedings of the Institution of Civil Engi- neers, Part 2, 83: 295–314.
Ervine, D.A., Falvey, H.T. and Withers, W. (1997) Pressure fluctuations on plunge pool floors.Journal of
Hydraulic Research, IAHR, 35, No. 2: 257–79.
Farhoudi, J. and Narayanan, R. (1991) Force on slab beneath hydraulic jump. Journal of Hydraulic Engineering,
American Society of Civil Engineers, 117 (1): 64–82.
Fiorotto, V. and Salandin, P. (2000) Design of anchored slabs in spillway stilling basins. Journal of
Hydraulic Engineering, ASCE, 126, No. 7: 502–12.
Hager, W.H. and Li, D. (1992) Sill-controlled energy dissipator.Journal of Hydraulic Research, 30 (2): 165–
81.
Hager, W.H., Bremen, R. and Kawagoshi, N. (1990) Classical hydraulic jump: length of roller. Journal of
Hydraulic Research, 28 (5): 592–608.
Hartung, F. and Häusler, E. (1973) Scour, stilling basins and downstream protec- tion under free overfall jets
at dams, in Proceedings of the 11th Congress on Large Dams, Madrid: 39–56.
Heller, V., Hager, W.H. and Minor, H.-E. (2005) Ski-jump hydraulics, Journal of Hydraulic Engineering
ASCE, 131, No. 5: 347–51.
Horˇení, P. (1956) Disintegration of a free jet in air, in Práce a Studie, No. 93, VUV,
Prague (inCzech).
ICOLD (1987) Spillways for Dams, Bulletin 58, International Commission on Large Dams,Paris.
Liu, P. (2005) A new method for calculating depth of scour pit caused by overflow water jets, Journal of
Hydraulic Research, IAHR, 43, No. 6: 695–701.
Locher, F.A. and Hsu, S.T. (1984) Energy dissipation at high dams, in Develop- ments in Hydraulic
Engineering, Vol. 2 (ed. P. Novak), Elsevier Applied Science,London.
Lopardo, R.A. (1988) Stilling basin pressure fluctuations (invited lecture), in Pro- ceedings of the
International Symposium on Model–Prototype Correlation of Hydraulic Structures (ed. H. Burgi),
American Society of Civil Engineers, New York:56–73.
—— (1997) Hydrodynamic actions on large dams under flood conditions, 3rd Int.
Conf. on River Flood Hydraulics, Stellenbosch, HR Wallingford.
Martins, R.B.F. (1975) Scouring of rocky river beds by free jet spillways. Water Power & Dam
Construction, 27 (4): 152–3.
 Chapter two
Construction ofstilling basin procedure
Diverting the river

A diversion dam and stilling basin is usually constructed across a river to create
a reservoir in the valley behind by storing the water that flows into it naturally.
Sometimes, they are built across dry valleys, or valleys with small streams, to
create a storage area for water that is transported from elsewhere.

Preparing the foundation:

A dam and basin is built on a soil or rock foundation, depending on the type of
construction . The foundation must be strong enough to carry the weight of the
dam, and the water pressures acting on the dam. The dam foundation is below
the original ground level. It is covered by weaker soils or rock that has to be
removed either by excavation or by blasting with explosives.

Sometimes, areas of weaker soils or rocks are found below the planned
foundation level and these have to be removed and replaced with stronger
materials

Some rock foundations have cracks and fissures in them. These have to be filled
with grout to stop water leaking out from the reservoir through the cracks once
the dam is completed. This is done by drilling holes down into the rock, and by
pumping grout into them, which spreads outwards to fill the cracks.

Building a concrete stilling basin ::

As so much concrete is required in a concrete site , it is mixed together in a

special building called a concrete batching plant, which is built on the
construction site. For very large dams, there may be a number of batching
plants. The concrete can be placed in the dam by two different methods.

Pouring Concrete:

The traditional method is to pour a wet mixture of concrete into a mould made
in the required shape of the dam. The mould, called formwork, is usually made
from sheets of timber. The mould is not made to the full height of the dam, as
the dam is built upwards in stages of about 1 to 2 metres at a time. The concrete
is left to dry before the next section is formed on top.

The concrete can be transferred from the batching plant to the dam by a number
of methods. Sometimes, a system of conveyor belts is used and sometimes the
concrete is taken by trucks to the bottom of the dam where it is poured into
skips which are lifted to the top of the dam by cranes