Proc. Instn Civ. Engrs, Part 2, 1976,61, Dec.

, 685-696

7942 A flow equation for submerged

rectangular weirs

M. M. ABOU-SEIDA, MSc, PhD*

A. A. QURAISHI, MSC. PhD*

An equation is derived to obtain the discharge over a sharp-edged rectangularweir in

the case of free and submerged conditions. The presence of an upstream sloping face
increases the discharge. This increase depends on the ratios between the downstream
head Ha, the upstream head HI and the angle of slope of the upstream face. The dis-
charge over a narrow crested rectangular weir with upstream edge rounding is greater
than the discharge over a sharp-edged weir for the same value of Ha/H1. This increase
in the discharge depends on the width of the weir in the direction of flow and on the
ratio HaIHl.

Notation
weir breadth normal to flow direction
weir width in the direction of flow
coefficient of discharge in the weir equation(9)
coefficient of dischargefor the flow over a weir
coefficient of dischargefor the flow through a submerged orifice
acceleration due to gravity
head on the upstream side of the weir over crest
head on the downstream side of the weir over crest
K = 3C-\/(&db
n exponent of Hl in the free discharge equation
, P heightoftheweirabovebedlevel
Ql discharge at head HI for free discharge case
Q discharge for submerged case
r radius of the rounded upstream corner for narrow crested weir
y vertical distance between the weir crest and the sloping face upper edge
a correction in the weir equation (9) for the case of the presence of an upstream
sloping face
P=HaIHl
0 angle of the sloping upstream face with vertical
A correction in the weir equation(9) for the caseof narrow crested weir with edge
rounding
+ ratio between the discharge in the submerged condition to the discharge in the
free condition

Written discussion closes IS February, 1977,for publication in Proceedings, Part 2.

Associate Professor of Civil Engineering, University of Riyadh, Saudi Arabia.
685
ABOU-SEIDA AND QURAlSHl
Introduction
Weirs of different shapes are used as control structures in irrigation systems.
The flow overthe weir is preferred to be a clear overfallflow where the equations
relating the discharge passing the weir and the head over the weir are well
established. In practice, it may sometimes be difficult or impossible to prevent
the downstream water level from rising above the sill level of the weir. In this
case the weir is said to be submerged or drowned. Generally it is expected that
this submergence will reduce the flow, yet the precise relations betweenthe up-
stream headH I , the downstream headHz over the weir crest and the dischargeQ
arenotknownformost of weir shapes.Moreover, theshape of theweir
cross-section has a great effect on the relationships between H,, Ha and Q. In
this study a general flow equation is developed to describe the discharge over the
vertical sharp-edged rectangular weir for the free case as well as the submerged
case. This has been tested and the results are presented. Also the effect of the
presence of an upstream sloping face on the discharge over the weir has been
studied in the case of submerged flow. The characteristics of the flow over the
narrow crested weir with edge rounding are also presented for the submerged
case.

Review of previous studies

2. Previousstudieshavebeenconcernedwiththe flow oversharp-edged
weirs. The general profileof the flow overthis type of weir is shown in Fig. l(a).
Vennard and Westod summarized the work of Francis, Fteley and Stearns,
Bazin, Cone and Cox, and showed that all the data presented could be put in a
relationship between HajHl and Q/Q,, where Ql is the discharge a t the head

4 -
Q,- /
~
~,
~~
;/
~~
/l
..
..
..
_
-~ + - / HI 8 2

Qz-

(4 (b)

3,,,A ,; zIy;nf-=/
4
HI
--L-kz-k
,,,,
P
#

B .

(4 (d1
Fig. 1. Weir modelsused in thestudy; (a) verticalweir, (b) weirwith upstream
sloping face equal to P, (c)weirwith upstreamslope less than P, (d) narrow
crested weir
686
 F L O WE Q U A T I O N FOR SUBMERGED
RECTANGULAR WEIRS
Hl computed from the equation for free discharge in the form
Q1 = KH1" . . . . . . . . (1)
Q is the discharge for submerged conditions, Hl is the head on the upstream side
of the weir and H2 is the head on the downstream sideof the weir. Vennard and
Weston found that all the data tended to fall on a single curve relating Q/Q,
against H,/Hl, except for high valuesof Hl/P, where P is the height of the weir
crest above the floor of the channel.
3. Villemonte2 carried out experiments on rectangular, triangular, parabolic,
cusped and proportional vertical weirs. He showed that all his results could be
presented by the equation
Q/Ql [ l - ( H z / H ~ ) n l ~.' ~ .~ ~. . . . ( 2 )
=
where n is the exponent for the free discharge as in equation ( l ) .
4. Mavis3conducted tests onrectangular,triangular,parabolic, circular,
Sutro and cusped vertical weirs. He derived a single equation to express the
results of his tests relating Q/Ql and H,/H,. The data presented by Villemonte2
and Mavis3 were plotted by King and Brater4 who found thatQ/Ql is a function
of (H,/Hl)" for all weir shapes used in their studies.
5. Otherinvestigators like Gibsonsassumed thatthe flow overthesub-
merged sharp-edged weir can be considered as a summation of two flows: the
flow over the upper part of depth H, - H,, which may be considered as a free
discharge over a weir, and that over the lower part of depth H,, which can be
considered as a flow through a submerged orifice, or
Q' = 3C~b(2g)1'2(Hl-Hz)3'a . . . . . (3)
Q" = C,b(2g)1'2H,(H1-H,)1'a . . . . . . (4)
The total Q equals Q'+ Q". Cl and C, are the coefficients of discharge for the
flow over aweir and flow through anorifice respectively, and b is the weir breadth
normal to the direction of flow.
6. Values of C with different values of H,/Hl were given. Gibsons quotes
Redtenbacherwhoassumed Cl=0.57 and C2=0.62 and Pestorlozziwho
assumed Cl =0.62 and C,=0.534-0.566 depending on the ratio Hz/Hl. The
value of C, increases as H2/Hl decreases.
7. The problems of flow over a weir with a sloping faceand flow over narrow
crested weirs have not been studied previously. It is felt that a general equation
can be developed to describe the discharge over sharp-edged rectangular weirs
for the free overfall case and the submerged case.

The generalized flow equation

8. With reference to Fig. l(a) and Gibsons it may be seen that the discharge
over the weir in the submerged condition is the sumof two discharges: Q', which
representsthe free flow over a sharp-edgedrectangular weir under a head
Hl - H,, and Q",which represents the discharge through an orifice of depth H,
and under a head H , - H,. In this case the discharge passing overthe weir can
be written as
Q = Q'+ Q" . . . . . . . . (5)
or
Q = +Clb(2g)1/a(Hl- +
C2bH2(2g)1'2(H1 - . . (6)
687
A B O U - S E I D AA N DQ U R A l S H l

Head over &/weir height, HI/ P

Fig. 2. Effect of sloping face of height less than P on C

9. In equation (6) the effect of the velocity of approach is neglected. If it is

assumed that Cl= C, = C, after rearranging the terms equation(6) can be put in
the form
Q = 3Cb(2g)1'2H13'2[(1-p)1/2(l+p/2)] . . . * (7)
where p= H2/Hl,or in general
Q = 3c+b(2gyw13/2 . . . . . . . (8)
where
+
= (l-py(l+/3/2) . . . . . . . (9)
for the free conditionH2/H1= O and += 1. Equation (8) for the free condition is
Q = +Cb(2g)1'2H13'2 . . . . . . (10)
$ represents the ratiobetween Q for thesubmerged condition and Qlfor the free
flow condition under the same headHl.
10. It is believed that equation (7) represents a general equation describing
the flow over rectangular sharp-edged weirs.
11. In equations(8) and (10) the value of Ccan be calculated using Rehbock's
formula3
C = 0*611+0*08H1/P . . . . . .
(11)

Presence of an upstream sloping face

12. It has beenshown6thatthepresence of an upstreamslopingface of
height P increases the discharge. Fig. l(b) shows the model of the weir in this
case. The relationship between the value of C, calculated from equation (lO),
and H J P for the free condition and fordifferent slopes is shown in Fig. 2. The
slope 1: 1 gives the maximum value of C and consequently the maximum value
of Q .
13. For submerged flow it is expected that equation (7) will be applicable
after introducing a coefficient a to account for the modifiedflow field caused by
the presence of the slope. Equation (7) is then put in the form
Q= . . . . . (12)
If the height of the slope is less than the full height of the weir, as in Fig. l(c),
equation (12) can be applied but with a different value of a. The coefficient a is
expected to be a function of H2/Hl and the angle of slope 8. This function can
be obtained experimentally.
688
 FLOW
EQUATION
FOR
SUBMERGED
RECTANGULAR
WEIRS
/

0.70-
c,
U
P

----- Weir with upstream slope

055 I I I I l
0.2 0-4 0.6 ae I.o
Head over welr/helght of welr, HI/P

Fig. 3. Effect of sloping face of height P on C

14. The value of C in equation (12) may be taken from Fig. 2 depending on
is lessthan P
H l / P and theangle of slope 8. For the caseof a slope whose height
(Fig. l(c)) the value of C is taken from Fig. 3.

Narrow crested weir withedge rounding

15. Sharp-crestedweirsused for irrigationprojectsneedcontinuousmain-
tenance. The crests become rusted or may be damaged by sand particles moving
with water or by floating debris. The use of a weir of a certain width in the
direction of flow may provide the solutionfor such problems. Previous studies6
for the free flow condition proved that when rounding the upstream edgeof the
weir, the discharge is increased. Fig. l(d) shows the shape of this weir with the
variables involved in its study. Fig. 4 shows the relation between Hla/Pr and C
for different values of r/B. It can be seen that the maximumvalue of C is
obtained when rlBZ0.75, where r is the radius of the curve and B is the weir
breadth in the direction of flow. The submerged case where the downstream
depth is higher than the height P of the weir was studied for rlB30.75. It was
expected that equation (7) could be written in the form
Q = +c+~b(2~)1~~3'a. . . . . (13)

./B= 0.75- 1.00,

Fig. 4. Variation of C with HIa/Pr for narrow crested weir

689
A B O U - S E I D AA N DQ U R A l S H l
where X is a coefficient expected to be a function of H,, Hl and B. This function
was obtained experimentally.

Experiments
16. The experiments were conducted in a rectangular flume, 30.48 cm wide,
30.48 cm deep and 9.10 m long. The flume had a smooth bed and Plexiglass side
walls. The weir models were made of brass for the vertical weirs with upstream
sloping face, and of wood for the narrow crestedweir models. Water is pumped
from the main laboratory sump through a 15 cm dia. pipe to the channel en-
trance. A valve in the delivery pipe is used to regulate the rate of flow. The
head over the weir is measured by a point gauge fitted to a travelling bridge.
The downstream depth is measured by a piezometer connected to the channel
bed. The discharge leaving the channel is measured by weight in a weighing
tank.Theupstreamdepth wasmeasured at a distance 3H1 fromthe weir
location. The downstream depth was measured at a distance five times the weir
height P from weir location. The downstream distance chosen was just beyond
the turbulence caused by the nappe. Fig. 5 shows the general arrangement of
the testing flume.
17. Three groups of tests were carried out to test the validity of equation (7),
and tofind the values of a in equation (12) and X in equation (1 3).
18. The first group dealtwith the submerged flow overasharp-edged
rectangularweir.Two weir heightswereused: 15.27 cm and 11.37 cm. The
second group of tests dealt with model weirs with an upstream sloping face.
Two sets of experiments were used: those with a slope whose height was equal
to the full height of the weir and those with a slope whose height was less than
the height of the weir. In thefirst case three heights and three slopes were tested.
The heights were 15.27 cm, 11.70 cm and 8.18 cm and theslopes 30", 45" and 60"
to the vertical. For the second case two heights-P= 15.27 cm and 11.37 cm-
were used with upstream slopes of 60°, 45" and 30°, and covering the rangey/P=
0 . 2 4 4 4 8 5 (seeFig. l(c)). The third group of tests dealtwith the flow over
narrow crested weirs withupstreamedgerounding.Four weir modelswere
used with widths of 2.54 cm, 5.08 cm, 7.62 cm and 10.16 cm. Two heights were
investigated: 15.80 cm and 10.15 cm. All weirs had r/B=0.75. Oneexperiment
was carried out for thesubmerged condition in the case of a broad crested weir
for comparison.

Results
Verification of the general flow equation
19. The testing of the validity of the general flow equation is shown in Fig. 6.
The experimental points obtained fromthe tests using sharp-edged vertical weirs
are showntogetherwiththevaluesobtainedfromequation (9). Thecurve
shows the relationship betweent,h and H 2 / H l . It can be seen that there is a good
agreement between the experiment and the derived equation. Thus the hypoth-
esis used to derive equation ( 8 ) is fairly well established.

Eflect of upstream slope whose height is less than P

20. Threedifferentslopeswereused for different values of y / P . It was
found that thepresence of these slopes does not have anyeffect on theflow over
690
 FLOW EQUATION F O R SUBMERGED
RECTANGULAR WEIRS

Measuring tank

Fig. 5. Experimentalflume

Fig. 6. Testing the validity of the weir equation

Fig. 7. Change of $ with HJH, for all values of 8, y and P

691
ABOU-SEIDA AND QURAlSHl

0 0.1 02 0.3 0.4 0.5 0.6 0.7 0.8 09

H2 /=l
Fig. 8. Change of a with HdH1 for weirs of sloping face

the sharp-edgedweirin the submerged case. Fig. 7 showsequation (9) for

vertical sharpedged weirs without slopes and the experimental points obtained
for the different slopes of height less than P. For the free fall condition the
presence of this slope changes the value of C and consequently the value of Q,as
shown in Fig. 3.

Efect of sloping upstream faceof height P

21. The effect of the threeslopes used, namely30", 45" and 60" to thevertical,
can be seen in Fig. 6. The 30" slope has slight or noeffect on the discharge, and
the experimentalpoints lie on thecurve for vertical weirs. For 45" and 60"
slopes an increase in the discharge was observed. This increase depends on the
ratio H2/H1. To obtain the value of a in equation (12) experimentally, an en-
larged chart was drawn to find the difference betweenthe experimental points for
the 45" and 60" slopes. a is the ratio between # when there is a slope and # for
the vertical case. The changeof a with H2/H, for different slopes used is shown
in Fig. 8. It can be concluded that using equation (9) together with equation
(12) and Fig. 8 gives the discharge for the case of the submerged flow condition
for a vertical weir with an upstream sloping face of height P.

Narrow crested weir with edge rounding

22. Previous studiess showed that the edge rounding of the narrow crested
weir increases the discharge. The maximum value of discharge was obtained
when the ratiobetween the radiusof rounding to theweir breadth in the direction
of flow was equal to or greater than 0.75. The models used in this study with
r/B=O-75 showed an increase of the discharge withthe increase of the breadthB
in the direction of flow. The increase depends on the ratioH2/Hl. Fig. 9 shows
a plot of Q/Ql against Ha/Hl for the vertical sharp crested weir represented by
692
 F L O W EQUATION F O R S U B M E R G E D RECTANGULAR WEIRS
Broad crest

B i 254 cm \\
oBt598cm

X B..
m B I7.62 cm
10.16 cm

=l

Fig. 9. Q/Q, against H J H , for narrow crested Weirs

equation (9) together with the results obtained from the four models used for
narrow crested weirs. The curve from the test of the broad crested weir is also
shown for comparison. Fig. 10 shows # against H2/Hl for different valuesof B.
The increase of B increases the # value and consequently the discharge. The
value of X in equation (13) is shown in Fig. 11. The value of X is a function of
Hz/H, andBlb. It is believed that equation (9) together with Figs6 and 11 gives
the discharge in the submerged conditionfor a narrow crested weir with theratio
r/B= 0.75.
693
A B O U - S E I D AA N DQ U R A l S H l

Bfb = 0.25 - 0.33

PP B'?o'oa3

Fig. 11. h against Ha/H1for narrow crested weir

Conclusion
23. Aflow equationhas beendeveloped forthe dischargeover a sharp
crested rectangular weir in such waya that it can be used for both free and sub-
merged flows. Theequationhas been verified experimentally. The presence
of an upstream sloping face whose heightis less than the full height of the weir
does not haveany effect on the discharge inthe submerged condition, butit does
have an effect in the free discharge case. The presence of a slope of height P
increases the discharge and the increase depends on the ratio H2/Hl and on the
angle of slope 0 but increases as the angle increases. For a slope of 30" to the
vertical there is no change in the valueof discharge for the sameH2/Hl and the
experimental points followed the curve obtainedfor vertical weir., This increase
is caused by the new component of the velocity inthe direction of flow resulting
from the presence of the slope. This component gives rise to additional dis-
charge.
24. For narrow c'rested weirs with rounded upstream corner, the discharge is
greater than thatresulting from the sharpedgedweirs. The increase of the width
of the weir in thedirection of flow increases the dischargeuntil the broadcrested
weir condition is reached. The value of the discharge in the free overfall con-
dition and in the submerged condition canbe obtained as shownin the examples
(S 26-28).
25. In previous experiments1+ the water surface profile downstream of the
weir was characterized by the formation of waves which persist to the channel
outlet. At small submergence ratios the nappe,plunges below the surface and
returns to the surface. For high submergence ratios the nappe plunges below
694
 F L O WE Q U A T I O N F O R S U B M E R G E DR E C T A N G U L A RW E I R S
the surface but remains above the weir crest level. In the two cases of sub-
mergence a standing wave is formed whose height and length depends on the
submerged ratio H2/H1. Downstream of this standing wave a series of small
waves forms which graduallyreduce in size downstream.Thisaspectneeds
further study.

Examples
26. To find Q for a sharp crested weir with P= 15.14 cm,Hl = 9.24 cm and
H2=3.71 cm one mayproceed as follows. For H1/P=0.61, from equation (11)
C = 0.6598
Ql/unit width = 3(0.6598)(2g)1'2(9.24)3/2= 0.5473 I/s per cm
H2/H1 = 0.401
and from equation (9)
' $I = 0.929.
Then using equation (8)
Q = 0.5473 X 0.929 = 0.508 I/s per cm
QmeQSured = 0.5026 I/s per cm.
27. In the presenceof a sloping faceof 45", the dischargecan be calculated as
Hl = 8.45 cm
P = 15.39 cm
H2 = 4.4 m.
so
HJP = 0.55.
Using Fig. 2 for the 45" slope
C = 0.712
&/unit width = 3(0.712)(2g)1/2(8-45)3/2= 0.5165 I/s per cm.
For H2/H1=0-529, from Fig. 6 $=0.87, and from Fig. 8 Q = 1.044. Then
Q = 0 8 7 X 1.044X 0.5165 = 0.467 l/s per cm
Qmepsured = 0.466 I/s per cm.

28. For a narrow crested weir with r/B=O.75, H,= 8.53 cm, P= 15.93 cm,
B= 10.16 cm and H2=4.61 cm
r = 7.62 cm
(Hl/P)(HJr) = (8~53)~/(15.93X 7.62) = 0.6.

Using Fig. 4
C = 0.7
Ql/unit width = +(0.7)(2g)1'2(853)3'a= 0515 l/s per cm.
From Fig. 6 for a vertical weir for H2/H,= 0.637
$ = 0.8.
Using Fig. 11, for H2/H1= 0.637 and B/b=0.33
h = 1.16.
695
A B O U - S E I D AA N DQ U R A l S H l
Equation (13) gives
Q = 0-8 X 1.16 X 0.515 = 0.478 l/s per cm
Qmeasured = 0.483 l/s per cm.

References
1. VENNARD J. K. and WESTONR. F. Submergenceeffect on sharp crestedweirs.
Engng News Rec., 1943, June, 818.
2. VILLEMONTE J. R. Submergedweirdischargestudies. Engng News Rec., 1947,
Dec., 866.
3. MAWSF.T. How to calculate flow over submerged thin-plate weirs. Engng News
Rec., 1949, July, 65.
4. KINGH. W. and BRATER E. F. Handbook of hydraulics, 5th edn. McGraw-Hill,
New York, 1963, 5-20.
5. GIBSONA. H. Hydraulics ant1 its application, 5th edn. Constable, London, 1952,
166.
6. ABOU-SEIDA M. M. Effect of sloping face and edge rounding on flow over weirs.
Bull. CON.Engng Cairo Uniu., i975.

696