DESIGN OF SPILLWAYS
1.0 GENERAL
Spillway is a safety valve provided in the dam to dispose of surplus flood waters from a reservoir after it has been filled to its maximum capacity i.e. Full Reservoir Level.
The importance of safe spillways needs no over-emphasis as many failures of dams have been caused by improper design of spillways or spillways of insufficient capacity especially in case of earth and rockfill dams which are susceptible to breaching, if overtopped. Concrete/Masonry dams can withstand moderate overtopping but this should be avoided.
Further, the spillway must be hydraulically and structurally adequate and must be so located that the overflowing discharges do not erode or undermine the downstream toe of the dam.
2.0 SELECTION OF DESIGN FLOOD
The spillway design flood is generally determined by transposing great storms which have been known to occur in the region over the drainage area. The resulting flood hydrographs are then determined by rational methods. In determining the discharge capacity consideration should be given to all possible contingencies, e.g. one or more gates being inoperative.
IS : 11223 – 1985 provides guidelines for fixing spillway capacity. Inflow design flood for the safety of the dam is guided by the following criterion:
The dams may be classified according to size by using the hydraulic head and the gross storage behind the dam as given below. The overall size classification for the dam would be the greater of that indicated by either of the following two parameters:
Classification
Gross Storage Hydraulic Head
Small 0.5 to 10 million m³ 7.5 to 12 m
Intermediate 10 to 60 million m³ 12 to 30m
Large Above 60 million m³ Above 30 m
The inflow design flood for safety of the dam would be as follows:
Size of dam Inflow Design Flood for safety of dam
Small 100 years flood
Intermediate Standard Project Flood (SPF)
Large Probable Maximum Flood (PMF)
Floods of larger or smaller magnitude may be used if the hazard involved in the eventuality of a failure is particularly high or low. The relevant parameters to be considered in judging the hazard in addition to the size would be:
i) Distance and location of human habitations on the downstream after considering the likely future developments
ii) Maximum hydraulic capacity of the downstream channel
For more important projects dam break studies are done as an aid to the judgement in deciding whether PMF needs to be used. Where the studies or judgement indicate an imminent danger to present or future human settlements, the PMF should be used as design flood.
2.1 Standard Project Flood (SPF)
It is the flood that may be expected from the most severe combination of hydrological and meteorological factors that are considered reasonably characteristic of the region and is computed by using the Standard Project Storm (SPS). While transposition of storms from outside the basin is permissible, very rare storms which are not characteristic of the region concerned are excluded in arriving at the SPS rainfall of the basin.
2.2 Probable Maximum Flood (PMF)
It is the flood that may be expected from the most severe combination of critical meteorological and hydrological condition that are reasonably possible in the region and is computed by using the Probable Maximum Storm (PMS) which is an estimate of the physical upper limit to maximum precipitation for the basin. This is obtained from the transposition studies of the storms that have occurred over the region and maximizing them for the most critical atmospheric conditions.
3.0 FLOOD ROUTING
The process of computing the reservoir storage volumes and outflow rates corresponding to a particular hydrograph of inflow is known as flood routing. It is used for arriving at the MWL for the project. The relationship governing the computation is essentially simple – over any interval of time the volume of inflow must equal the volume of outflow plus the change in storage during the period. If the reservoir is rising, there will be increase in storage and change in storage will be positive, if the reservoir is falling, there will be decrease in storage and the change in storage will be negative.
For an interval of time Δt, the relationship can be expressed by the following expression:
I. t
Where,
I = Average rate of inflow during time equal to Δt
O = Average rate of outflow during time equal to Δt
ΔS = Storage accumulated during time equal to Δt
The following three curves are required for carrying out the computations:
a) The inflow flood hydrograph
b) The reservoir capacity curve
c) The rating curve showing the total rate of outflow through outlets and over the spillway against various reservoir elevations
Flood routing in gated spillways is generally carried out assuming the flood to impinge at FRL assuming inflow equal outflow to at that level. For ungated spillways this would correspond to the spillway crest or a little above this.
The methods generally adopted for flood routing studies are:
i) Trial and Error Method
ii) Modified Puls Method
3.1 Trial and Error Method
This method arranges the basic routing equation as follows:
I1
2 2
The procedure involves assuming a particular level in the reservoir at the time interval Δt, and computing the values on the right side of the above equation. The computed value on the right side of the equation, corresponding to the assumed reservoir level, is compared with the known value on the left side of the equation. If the two values tally, then the assumed reservoir level at the end of the time interval is OK; otherwise a new reservoir level is assumed and the process is repeated till the required matching is obtained.
This method gives quite reliable results, provided the chosen time interval is sufficiently small, so that the mean of the outflow rates at the start and the end of the interval may be taken as the average throughout the interval.
3.2 Modified Puls Method
This method arranges the basic routing equation as below so that the knowns are placed on the left side and unknowns are placed on the right side of the equation.
Since this equation contains two unknowns it cannot be solved unless a second independent function is available. In the modified Puls method, a storage-indication curve viz. outflow O versus the quantity (S/Δt+O/2) is constructed for the purpose.
In the above equation, it may be noted that subtracting O2 from (S2/Δt+O2/2) gives (S2/ΔtO2/2). This expression is identical to (S1/Δt-O1/2) on the left side of the equation except for the subscripts. Since the subscript 1 denotes values at the beginning of a time increment and subscript 2 denotes values at the end of a time increment, it is apparent that (S2/Δt-O2/2) at the end of one time increment is numerically equal to (S1/Δt-O1/2) for the beginning of the succeeding time increment.
The detailed routing procedure is as follows:
i) Compute the numerical value of left side of the equation for given values of I1, I2, S1 and O1 for the first time increment.
ii) With this numerical value, which equals (S2/Δt+O2/2), refer storage-indication curve and read outflow O2 corresponding to this computed value of (S2/Δt+O2/2). The O2 thus read is the instantaneous outflow at the end of the first time increment.
iii) Subtract this value of O2 from (S2/Δt+O2/2), which gives the value for (S2/ΔtO2/2). The value of (S1/Δt-O1/2) for the second time increment is equal to (S2/ΔtO2/2) for the first time increment. Consequently the left side of the equation can be computed for the second time increment and the entire procedure is repeated.
TYPES OF SPILLWAYS
Spillways can be classified as controlled or uncontrolled depending upon whether they are gated or ungated. Further they are also classified based on other prominent features such as control structure, discharge channel or some such other components.
The common types of spillways used are:
i) Overfall or Ogee ii) Orifice or sluice iii) Chute or trough iv) Side channel
v) Tunnel/Shaft or Morning Glory vi) Siphon
4.1 Overfall or Ogee Spillway
The overfall type is by far the most common and is adapted to masonry dams that have sufficient crest length to provide the desired capacity.
This type comprises a control weir which is ogee or S-shaped. The ogee shape conforms to the profile of aerated lower nappe from a sharp crested weir. The upper curve at the crest may be made either broader or sharper than the nappe. A broader curve will support the sheet and positive hydrostatic pressure will occur along the contact surface. The support sheet thus creates a backwater effect and reduces the coefficient of discharge. The sharper crest on the other hand creates negative pressures, increases the effective head and thereby the discharge.
These spillways are generally provided in Masonry/Concrete dams and also in composite dams as central spillways located in the main river course. Examples are Bhakra dam, Rihand dam, Sriram Sagar dam, Nagarjunasagar dam, Jawahar Sagar dam, Tenughat dam, Srisailam dam, Tawa dam, Ukai dam etc.
A typical ogee spillway is shown in figure below:
4.2 Orifice Spillway
Low crested spillways with either breast wall or sluice type arrangements are now increasingly being provided for flushing out the silt and controlling the silt entry in the power intake which is kept above the spillway crest. These spillways are called orifice or sluice spillways (See figures below).
The orifice spillways have the advantage of having high discharging capacity due to the high water head. At sites where only a limited area and relatively short length of suitable foundation material are available for the spillway structure, the orifice spillway offers the most economic means of passing the design flood. However, the orifice spillways result in high flow concentration, which increases the size and cost of energy dissipation work below.
Orifice spillways are being provided in many of our diversion dams recently in rivers carrying heavy silt load. The power intake is kept above the spillway crest and as close to the spillway as possible. This kind of spillway arrangements thus performs the dual function of passing the flood and managing the sediment in the reservoir.
Examples of spillways with breast wall type arrangements are in Ranganadi H.E. Project (Arunachal Pradesh), Chamera H.E. Project Stage-I (Himachal Pradesh), Rangit H.E. Project Stage-III (Sikkim) etc. and that of sluice spillways are in Tala H.E. Project (Bhutan), Nathpa Jhakri H.E. Project (Himachal Pradesh), Myntdu H.E. Project Stage-I (Meghalaya) etc.
4.3 Chute Spillway
A spillway where discharge is conveyed from the reservoir to the downstream river through an open channel or chute along a dam abutment or through a saddle is called a chute or trough spillway. The chute is the commonest type of water conductor used for conveying flow between control structures and energy dissipators. Chute can be formed on the downstream face of gravity dams, cut into rock abutments and either concretelined or left unlined and built as free-standing structures on foundations of rock or soil.
These are mostly used with earth/rockfill dams and have the following main advantages:
i) Simplicity of design
ii) Adaptability to all types of foundations and
iii) Overall economy by using large amount of spillway excavation in dam construction
Examples of chute spillways are Beas Dam, Ram Ganga Dam, Kolar Dam, Tehri Dam etc.
A typical chute spillway is shown in figure below:
4.4 Side Channel Spillway
The distinctive feature of side channel spillway which distinguishes it from chute spillway is that whereas in the chute spillway the water flows at right angle to the axis of the dam, in the side channel spillway, the flow is initially in a channel parallel to the axis of the dam and thereafter it flows in a discharge channel at approximately right angle to the dam axis.
This type of spillway is suited to narrow canyons with steep sides which rise to a considerable height above the dam. This type of spillway is also provided at sites where the overfall type is ruled out for some reason and where saddle of sufficient width is not available to accommodate a trough (chute) type spillway. It is assumed that all the energy of the overfalling water is dissipated in turbulence in the side channel. Example of side channel spillway is Pancheshwar Project.
A typical layout of a side channel spillway is illustrated in figure below:
4.5 Tunnel/Shaft or Morning Glory Spillway
In this type of spillway water enters over the lip of a horizontal circular crest and drops through a vertical or sloping shaft and then flows downstream through a horizontal conduit or tunnel. The spillway is suitable to dam sites in narrow canyons where room for a spillway restricted.
In some instances advantage of the existing diversion tunnel has been taken for conversion into tunnel spillway. A disadvantage of this type is that the discharge beyond a certain point increases only slightly with increased depth of overflow and therefore does not give much factor of safety against underestimation flood discharge as compared to the other types.
Examples of Tunnel/Shaft spillways are Tehri Dam, Itaipau Dam etc. A typical Tunnel/Shaft Spillway is shown in figure below:
4.6 Siphon Spillway
Siphon spillways are based on the principle of siphonic action in an inverted bent pipe. If such pipe is once filled with water, it will continue to flow so long as the liquid surface is higher than the lower leg of the pipe unless of course, the upper leg gets exposed earlier.
Siphon spillways are often superior to other forms where the available space is limited and the discharge is not extremely large. They are also useful in providing automatic surface-level regulation within narrow limits. The siphon spillways prime rapidly and bring into action their full capacity. Therefore, they are especially useful at the power house end of long power channels with limited forebay capacity where a considerable discharge capacity is necessary within a very short time in order to avoid overflow of the channel banks.
However, siphon spillways are not very common mainly because of:
i) Possibility of clogging of the siphon passage way and siphon breaker vents with debris, leaves etc.
ii) The occurrence of sudden surges and stoppages of outflow as a result of the erratic make and break action of the siphon, thus causing fluctuations in the downstream river stage.
iii) The release of outflows in excess of reservoir inflows whenever the siphon operates, if a single siphon is used. Closer regulation which will more nearly balance outflow and inflow can be obtained by providing a series of smaller siphons with their siphon breaker vents set to prime at gradually increasing reservoir heads.
iv) Vibration disturbances which are more pronounced than in other types of spillways.
A siphon spillway through a dam is shown in figure below:
4.0 HYDRAULIC DESIGN OF OVERFALL OGEE SPILLWAYS (Refer IS:
6934)
Overfall ogee spillway has its overflow profile conforming, as nearly as possible, to the profile of the lower nappe of a ventilated jet of water over a sharp crested weir. These spillways are classified as high and low depending on whether the ratio of height of the spillway crest measured from the river bed to the design head is greater than or equal or less than 1.33 respectively. In the case of high overflow spillways the velocity of approach head may be considered negligible.
5.1 Shape of Ogee Profile
i) Spillways with vertical upstream face
Upstream Quadrant
The upstream quadrant of the crest may conform to the ellipse:
The magnitude of A1 and B1 are determined from the graph P/Hd vs A1/Hd and B1/Hd respectively in fig.2 of IS:6934, where,
P = Height of crest from the river bed
Hd = Design Head
Downstream Quadrant
The downstream profile of the crest may conform to the equation:
X 21.85K 2 .Hd 0.85 .Y2
The magnitude of K2 is determined from the graph P/Hd vs K2 in fig.2 of IS:6934.
ii) Spillways with sloping upstream face
In the case of sloping upstream face, the desired inclination of the face is fitted tangentially to the elliptical profile described under (i) above, with the appropriate tangent point worked out from the equation. The profile of the downstream quadrant remains unchanged.
Figure 2 – IS:6934
iii) Spillways with crest offsets and risers
Whenever structural requirements permit, removal of some mass from the upstream face leading to offsets and risers as shown in fig.1 of IS:6934 results in economy. The ratio of risers M to the design head Hd i.e. M/Hd should be at least 0.6 or larger, for the flow condition to be stable. The shapes of u/s and d/s quadrants defined for spillways with vertical upstream face are also applicable to overhanging crests, for the ratio M/Hd > 0.6.
5.2 Discharge Computations
The discharge over the spillway is generally computed by the equation
Q 2 gC.L.H 3/ 2
where,
C = Coefficient of Discharge
L = Effective length of crest
H = Head over crest
i) Effective Length of Overflow Crest
The net length of overflow crest is reduced due to contraction caused by abutment and crest piers. The effective length L of the crest may be calculated as follows:
2H(N.Kp Ka)
where,
L’ = Overall length of the crest excluding piers
H = Head over crest
N = Number of piers
Kp = End contraction coefficient of piers
Ka = End contraction coefficient of abutment
The pier contraction coefficient, Kp is affected by the shape and location of the pier nose, thickness of pier, the actual head in relation to the design head and the approach velocity. The average pier contraction coefficients may be taken as follows:
Type Kp
Square-nosed piers with rounded corners of radius about 0.1 times pier thickness
0.02
Round-nosed piers 0.01
Pointed-nosed piers 0.0
The abutment contraction coefficient, Ka is affected by the shape of the abutment, the angle between the upstream approach wall and the axis of flow, the actual head in relation to the design head and the approach velocity. The average abutment coefficient may be taken as follows:
Type Ka
Square abutments with head wall at 90o to direction of flow 0.2
Rounded abutments with head wall at 90o to direction of flow, when 0.5Hd > R >
0.15Hd
0.1
Rounded abutments where R > 0.5Hd and head wall not more than 45o to direction of flow 0.0
ii) Coefficient of Discharge
The value of coefficient of discharge depends on the following:
a) Shape of the crest
b) Depth of overflow in relation to design head
c) Depth of approach
d) Extent of submergence due to tail water
e) Inclination of the upstream face
Fig. 3 of IS:6934 gives the coefficient of discharge C for the design head as a function of approach depth and inclination of upstream face of the spillway. These curves can be used for preliminary design purposes.
Fig. 4 of IS:6934 gives the variation of coefficient of discharge as a function of ratio of the actual head to the design head (i.e. H/Hd). This curve can be used to estimate C for heads other than design head.
The coefficient of discharge is reduced due to submergence by the tail water. The position of the downstream apron relative to the crest level also has an effect on the discharge coefficient. Fig. 5A and 5B of IS:6934 give the variation of C with the above parameters.
iii) Design Head
When the ogee crest is formed to a shape differing from the ideal shape or when the crest has been shaped for a head larger or smaller than the one under consideration, the coefficient of discharge will differ.
A design head grater than the actual head will push the crest surface into the theoretical nappe and result in greater pressure along the curved surfaces and in lower discharge capacities. Conversely, a design head lower than the actual head pulls the crest surface below the theoretical nappe, resulting in sub-atmospheric pressures over some portion of the crest curve. At the same time the discharge capacity of such a crest curve is increased.
Excessive sub-atmospheric pressures can result in pulsating, inefficient spillway operation, and possibly damage to the structure as a result of cavitations. A certain amount of sub-atmospheric pressure can be attained without undesirable effects. Figure provides a guide for determining the minimum pressures on the crest for various ratios of design head and actual head on the crest.
Designing the crest shape to fit the nappe for a head less than maximum head expected often results in economies in construction. The resulting increase in unit discharge may make possible a shortening of the crest length, or a reduction in freeboard allowance for reservoir surcharge under extreme flood conditions.
Because the occurrence of design floods is usually so infrequent, the spillway crests are fitted to the lower nappe of a head which is 75% of that resulting from the actual discharge capacity. Tests have shown that the sub-atmospheric pressures on a nappeshaped crest do not exceed about half of the design head when the design head is not less that about 75% of the maximum head. An approximate diagram of the sub-atmospheric pressures, as determined from model tests, is shown by figure. The design head is normally kept as 80% to 90% of the maximum head corresponding to MWL.
The minimum crest pressure must be greater than cavitation pressure. It is suggested that the minimum pressure allowable for design purposes be 20ft of water below sea level atmospheric pressure and that the altitude of the project site be taken into account in making the calculation. For example, assume a site where the atmospheric pressure if 5ft of water less than sea level pressure, and in which the maximum head contemplated is 60ft; then, only 15 additional feet of sub-atmospheric pressure is allowable.
