Ain Shams Engineering Journal (2015) 6, 399402

Ain Shams University

Ain Shams Engineering Journal

www.elsevier.com/locate/asej
www.sciencedirect.com

CIVIL ENGINEERING

Power-law free overfall in subcritical ow regime

Ali R. Vatankhah
Department of Irrigation and Reclamation Engineering, University College of Agriculture and Natural Resources, University
of Tehran, P.O. Box 4111, Karaj 31587-77871, Iran
Received 12 February 2014; revised 5 October 2014; accepted 23 October 2014
Available online 25 November 2014

KEYWORDS
Free overfall;
Power-law section;
End depth;
Discharge relationship

Abstract Free overfall can be used as a discharge measuring structure by a single measurement of
end ow depth. Many theoretical and experimental studies carried out on free overfalls with different cross-section shape, different slope (mild and steep), and different roughness. However there is
not any study for prismatic free overfalls with power-law cross section. This research presents a theoretical end-depth ratio (EDR) and a theoretical end depth-discharge (EDD) relationship for a free
overfall located in a horizontal open channel with power-law section. The power-law section is a
general section which reduces to the commonly used rectangular, parabolic, and triangular sections.
Direct solutions are proposed by simulating the free overfall as a weir without crest in subcritical
ow regime. Experimental and theoretical results are then used to verify the proposed relationships.
The proposed relationships show excellent agreement with the experimental results.
 2014 Faculty of Engineering, Ain Shams University. Production and hosting by Elsevier B.V. This is an
open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

1. Introduction
Sudden drop (free overfall) is usually used as a discharge-measuring device in open prismatic channels with different shapes.
The depth of ow at the sharp edge is known as brink depth or
end depth [1]. The ratio of the end depth to the critical ow
depth (EDR) offers a possibility to estimate the discharge.
Flow discharge prediction is possible using the free overfall
without the need to any extra water structure such as the weir.
Fig. 1 shows a schematic view of ow in a channel with a drop.
Generally, the value of end depth depends on cross sectional
shape of the approach channel, the longitudinal slope and
E-mail address: arvatan@ut.ac.ir
Peer review under responsibility of Ain Shams University.

Production and hosting by Elsevier

roughness of the channel also the characteristics of the nappe.

If the longitudinal slope of the channel is negative, zero or
mild, the ow at upstream of end depth will be critical as
shown in Fig. 1.
Free overfall (brink) is used as a ow measuring structure
for different shapes of the channels with known end ow depth
[2,3]. The end depth-discharge (EDD) relationship for channels with different shapes has been extensively studied, experimentally and theoretically, by various researchers. The main
focus of this study is on direct discharge equation in terms
of end ow depth for subcritical ows in power-law sections.
The power-law section reduces to the commonly used rectangular, parabolic, and triangular sections.
Rouse [4] was the rst to investigate the end depth experimentally in rectangular channels and obtained an end-depth
ratio (EDR = 0.715) in mildly sloping rectangular channels
for the overfall with conned nappe. After Rouse, the
researchers concentrated on the value of the end depth ratio.
Dey [2] reported a comprehensive review of researches on free

http://dx.doi.org/10.1016/j.asej.2014.10.013
2090-4479  2014 Faculty of Engineering, Ain Shams University. Production and hosting by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

400

A.R. Vatankhah

Figure 1

Denition sketch of a power-law free overfall.

overfalls. For rectangular free overfall and according to experimental data; Rajaratnam and Muralidhar [5] reported the
value as 0.715, and Kraijenhoff and Dommerholt [6] gave it
as 0.714 whereas Rajaratnam and Muralidhar [5] reported a
value of 0.705 for unconned free overfall. Among theoretical
studies, only Ferro [7] obtained a value of 0.715; Hager [8] and
Marchi [9] reported 0.696 and 0.706, respectively.
Rajaratnam and Muralidhar [10] experimentally investigated the effect of different shapes of the channel cross section.
They reported EDR = 0.795 and 0.772 for triangular, and
parabolic sections, respectively.
This research uses weir approach to determine the end
depth of a free overfall in horizontal or mildly sloping
power-law channels. The ow over a drop in a power-law
channel is simulated by that over a sharp-crested weir to calculate the EDR and end depth-discharge (EDD) relationship.
This approach eliminates the need of an experimentally determined pressure coefcient. The proposed model is compared
to the experimental data and other existing theoretical
solutions.
2. Geometry of power-law section
The geometry of power-law section is shown in Fig. 2. The
section has two identical sides. Power-law section represents
a class of geometric shapes where the area is related to the ow
depth, y, as follows [11]
A kya

where k and a are parameters for which the channel takes

different shapes. As noted ow area, A, is a function of the vertical coordinate y measured above the channel bed. The value
of a equals 1, 1.5, and 2 for rectangular, parabolic, and triangular channels, respectively, as shown in Fig. 2. In a triangular
channel, k is side slope of the channel, k = z (1 vertical to z
horizontal), and in a rectangular channel k represents the
bottom width of the channel, k = b.

Figure 2

Geometry of a power-law section.

3. End depth ratio (EDR) and end-depth discharge (EDD)

relation
The ow of a free overfall in an open channel can be assumed
to be similar to the ow over a sharp-crested weir of zero
height [7,8,1218]. Hence, theoretical procedure applied to
compute the discharge over a sharp-crested weir can also be
used to a free overfall to get the end-depth-discharge (EDD)
relationship. In this study, it is assumed that the emerging jet
is undeected (free ow) and a coefcient of contraction, Cc
is considered to account for the curvature of streamlines.
Moreover, zero pressure is assumed at the brink and a subcritical approach ow to the brink is considered. It is notable that
the discharge coefcient, Cd (=CvCfCc), of the weirs may be
considered of three components; velocity head correction coefcient, Cv, head loss correction coefcient, Cf, and contraction
coefcient, Cc. Cv is considered for neglecting the velocity head
in the approach channel, and Cf is considered for neglecting
the frictional head losses through the weir. The discharge coefcient of the free overfall (a weir with zero height) can be considered to be equal to its contraction coefcient Cc under two
conditions: the specic energy, H, at the section with critical
ow depth yc is used for the discharge computation through
an horizontal elemental strip Cv @ 1, and the frictional head
losses along the short and horizontal free overfall are negligible
Cf @ 1.
The discharge through an elemental strip of thickness dy at
a height y above the brink is given by the following:
dQ Cc 2g1=2 H  y1=2 dA

Figure 3
overfall.

End-depth ratio EDR in mildly sloping power-law free

Power-law free overfall in subcritical ow regime

401

where Q is the discharge, g is the acceleration due gravity, and

H is the specic energy at the section with critical ow depth
yc. Note that, if the slope of channel is negative, zero or mild,
the ow at upstream of end section will be critical and end ow
depth, ye, depends only on the cross sectional shape of the
approach channel and its discharge.
Using dA = akya1dy and integrating both sides of Eq. (2),
the total discharge of the channel is computed by the
following:
Z yc
Q akCc 2g1=2
H  y1=2 ya1 dy
3
0

Let s = y/yc then Eq. (3) can be written as:

Z 1
Q akCc 2g1=2 yca1=2
H  s1=2 sa1 ds

in which s is a dummy variable dened on the interval [0, 1] and

H* = H/yc = 1 + 1/(2a). The coefcient of contraction Cc is
considered to account for the curvature of streamlines and is
equal to ratio of ow area at brink to the section with critical
ow depth and is determined as Cc = (ye/yc)a. substituting H*
and Cc into Eq. (4) yields.
1=2
Z 1
1
a
Q ak2g1=2 y1=2
y
1


s
sa1 ds
5
c
e
2a
0
The discharge, Q, for critical ow condition (Froude number is
unity) can also be written in terms of critical ow depth, yc, as
follows:
g1=2
Q
kya1=2
6
c
a
Eliminating Q from Eqs. (5) and (6) yields
!1=a
1=2
Z 1
ye
1
1=2 3=2
a1
2 a
1 s
s ds
2a
yc
0
Table 1

In which subscript c denotes the condition of the critical

state of the ow at upstream and subscript e denotes the
condition of the ow at brink. The integral term in Eq. (7)
can be determined as follows:
Z 1
0

1
s
2a

1=2

sa1 ds

1
1

a2 2a3

1=2 


1
2a
F  ; a ; 1 a;
2
1 2a
8

where F is the hypergeometric function [19].

To provide the user with a simple equation, an approximation is proposed using the curve tting regression technique for
any value of a between 1 and 2 as follows:
1=2
Z 1
1
0:766 0:895a1:45
1 s
sa1 ds
9
2a
1 0:68a1:5
0
The maximum percentage error of Eq. (9) compared to
exact value of Eq. (8), is less than 0.04% for practical ranges
of 1 < a < 2.
Substituting integral term from Eq. (9) into Eq. (7), the end
depth ratio of the power-law free overfall becomes as follows:

1=a
ye
0:766 0:895a1:45
21=2 a3=2 
10
yc
1 0:68a1:5
Substituting yc from Eq. (10) into Eq. (6) the following EDD
relationship for subcritical ow in a power-law free overfall
can be obtained.

 1 1
0:766 0:895a1:45 2a 1=2 a1=2
Q a1=2 21=2 a3=2 
kg ye
1 0:68a1:5
11
Eqs. (10) and (11) present solutions for the power-law free
overfall and can be used for any value of a between 1 and 2.
According to Eq. (10) EDR = 0.715, 0.777 and 0.817 for rectangular, parabolic, and triangular sections, respectively, as

Comparison of proposed model and other research results (power-law free overfall in subcritical ow regime).

Cross section

Experimental and theoretical results

EDR = ye/yc

Dierence (%)

Discharge equation (Q)

Dierence (%)

Rectangular

Rousea [4]
Rajaratnam and Muralidhara [10]
Kraijenho and Dommerholta [6]
Hager [8]
Marchi [9]
Murty Bhallamudi [21]
Ferro [7]
Beirami et al. [20]
This study

0.715b
0.715b
0.705
0.714
0.696
0.706
0.705
0.715
0.702
0.715

0.00a
0.00a
1.40a
0.14
2.66
1.26
1.40
0.00
1.82
Average = 0.96

1.654g1/2bye3/2
1.654g1/2bye3/2
1.689g1/2bye3/2
1.657g1/2bye3/2
1.722g1/2bye3/2
1.686g1/2bye3/2
1.689g1/2bye3/2
1.654g1/2bye3/2
1.702g1/2bye3/2
1.654g1/2bye3/2

0.00a
0.00a
2.12a
0.18
4.11
1.93
2.12
0.00
2.90
Average = 1.48

Triangular

Rajaratnam and Muralidhara [10]

Ali and Sykes [22]
Murty Bhallamudi [21]
Ahmad [23]
Beirami et al. [20]
This study

0.795
0.798
0.795
0.802
0.805
0.817

2.69a
2.33
2.69
1.84
1.47
Average = 2.03

1.255g1/2zye5/2
1.243g1/2zye5/2
1.255g1/2zye5/2
1.228g1/2zye5/2
1.216g1/2zye5/2
1.171g1/2zye5/2

7.17a
6.15
7.17
4.87
3.84
Average = 5.84

Parabolic

Rajaratnam and Muralidhara [10]

Ali and Sykes [22]
Murty Bhallamudi [21]
Beirami et al. [20]
This study

0.772
0.747
0.758
0.764
0.777

0.64a
3.86
2.45
1.67
Average = 2.16

1.370g1/2kye2
1.463g1/2kye2
1.421g1/2kye2
1.399g1/2kye2
1.351g1/2kye2

1.41a
8.29
5.18
3.55
Average = 4.61

a
b

Experimental result.
Conned nappe.

402
shown in Fig. 3. It should be noted that EDR = 0.715 and
0.777 (rectangular and parabolic free overfalls) are very near
to experimental data reported by [10].
4. Model verication
Experimental data and theoretical results presented in the literature were used to verify the proposed relationships, i.e., Eqs.
(10) and (11). The very good agreement is evident from the
analysis of Table 1. This table shows the percentage difference
in computed EDR via Eq. (10) compared to experimental data
and other theoretical results. According to Table 1, the values
of EDR found by other researchers have a slight deviation
(average percentage difference of 0.98% up to 2.16%) with
the proposed model. Similarly, the values of EDD found by
other researchers have a slight deviation (average percentage
difference of 1.48% up to 5.84%) with the proposed model.
5. Conclusion
The free overfalls of the horizontal power-law channels have
been analyzed using the weir approach. General solutions
are proposed for EDR and EDD for subcritical ows. Proposed solutions are valid over the entire practical range of
end depth and show excellent agreement with the experimental
data and other theoretical results. The proposed relationships
for discharge are useful computational tools for the estimation
of discharge in horizontal power-law open channels.
Acknowledgement
The author gratefully acknowledges the support provided by
the Center of Excellence for Evaluation and Rehabilitation
of Irrigation and Drainage Networks, University of Tehran.
References
[1] Nabavi SV, Beirami MK, Chamani MR, Sterling M. Free
overfalls in at-based circular and U-shaped channels. Flow
Meas Instrum 2011;22:1724.
[2] Dey S. Free over fall in open channels: state-of-the-art review.
Flow Meas Instrum 2002;13(56):24764.
[3] Rashwan IMH, Idress MI. Brink as a device for measurement
discharge for partially lled circular channel. Ain Shams Eng J
2013;4(2):13341.
[4] Rouse H. Discharge characteristics of the free overfall. Civil Eng.,
ASCE 1936;6(4):25760.
[5] Rajaratnam N, Muralidhar D. Characteristics of the rectangular
free overfall. J Hydraul Res 1968;6(3):23358.
[6] Kraijenhoff DA, Dommerholt A. Brink depth method in rectangular channel. J Irr Drain Div 1977;103(2):1717.

A.R. Vatankhah
[7] Ferro V. Theoretical end-depth-discharge relationship for free
overfall. J Irr Drain Eng 1999;125(1):404.
[8] Hager W. Hydraulics of plane free overfall. J Hydraul Eng
1983;109(12):168397.
[9] Marchi E. On the free overfall. J Hydraul Res 1993;31(6):77790.
[10] Rajaratnam N, Muralidhar D. End depth for exponential
channels. J Irr Drain Div 1964;90(1):1739.
[11] Subramanya K. Flow in open channels. New Delhi, India: Tata
McGraw-Hill; 1986.
[12] Dey S. EDR in circular channels. J Irr Drain Eng
2001;127(2):1102.
[13] Ahmad Z. Flow measurement using free overfall in inverted semicircular channel. Flow Meas Instrum 2005;16(1):216.
[14] Dey S. End depth in U-Shaped channels: a simplied approach. J
Hydraul Eng 2005;131(6):5136.
[15] Ahmad Z, Azamathulla HMd. Quasi-theoretical end-depth-discharge relationship for trapezoidal channels. J Hydrol 2012;456
457:1515.
[16] Vatankhah Ali R. Comment on Direct solution for discharge in
circular free overfall by Ahmad Z and Azamathulla HMd. J
Hydrol 2012;466467:1857.
[17] Vatankhah Ali R. Direct solution for discharge in generalized
trapezoidal
free
overfall.
Flow
Measure
Instrument
2013;29(1):614.
[18] Vatankhah Ali R. Comment on Quasi-theoretical end-depthdischarge relationship for trapezoidal channels. J Hydrol
2013;477:2614 [16 January].
[19] Lebedev NN. Special functions and their applications. New York:
Dover; 1972.
[20] Beirami MK, Nabavi SV, Chamani MR. Free overfall in channels
with different cross sections and sub-critical ow. Iran J Sci
Technol 2006;30(B1):97105.
[21] Murty Bhallamudi S. End depth in trapezoidal and exponential
channels. J Hydraul Res 1994;32(2):21932.
[22] Ali KHM, Sykes A. Free-vortex theory applied to free overfall. J
Hydraul Div, ASCE 1972;98(5):9739.
[23] Ahmad Z. Free overfall as measuring device in triangular
channels. Conf Hydraul, Water Resources Ocean Eng 2002:1159.

Ali R. Vatankhah earned his M.Eng. (1999)

and Ph.D. (2008) from University of Tehran.
He is currently an associate professor in the
Department of Irrigation and Reclamation
Engineering, University of Tehran, Tehran,
Iran. He has authored more than 100 international refereed journal articles. His research
interests include investigation of hydraulic
behavior of irrigation structures and their
optimum design, and mathematical modeling
in open channel hydraulics.