Course: Advanced Hydraulics Module 5: Flow through Non-Prismatic Channel
Section
Lecture 5
Channel Junction
CHANNEL JUNCTION
In order to compute the flow characteristics in a channel network often computer
programs that compute the water surface profile are employed.
Course: Advanced Hydraulics Module 5: Flow through Non-Prismatic Channel
Section
Most of the models employed use Newton-Raphson method in order to solve the
simultaneous non linear equation generated for every reach of channel simultaneously.
Flow through a channel network is a phenomenon that involves numerous variables, such
as the number of the adjoining channels, the angles of intersection, the shape and slope of
the channels, the directions and discharges of flow, rounding of the corners at junction
etc.
The problem is so complicated that only a few simple and specific cases have been
studied.
While analyzing the subcritical flow through channel networks the flow depth and
discharge at every channel is not known a priori.
Consider a channel network having N channels connected. Each channel has a different
cross section, bottom slope and Manning's n.
Consider a simple looped channel network having two parallel channels.
Figure. 5.9: A Looped Channel Network (Chaudhry, 1994)
Course: Advanced Hydraulics Module 5: Flow through Non-Prismatic Channel
Section
The energy equation between the channel reach between section j and j+1 for a i th
channel can be written as
Qi2, j Qi2, j 1
zi, j yi, j i, j z i , j 1 y i , j 1 i , j 1 hf j , j 1
2 gAi2, j 2 gAi2, j 1
Where y i , j 1 and yi , j is the flow depth at section j and j+1 respectively for the channel i.
hf j , j 1
is the head loss between section j and j+1.
Qi2, j Qi2, j 1 Qi2, j 1 ni2, j 1 Qi2, j ni2, j
xi , j 1 xi , j 2 2
1
zi, j yi, j i, j z i , j 1 y i , j 1 i , j 1 2 2 1.33
2 gAi2, j 2 gAi2, j 1 2 C Ai , j 1 Ri1,.33 C 0 Ai , j Ri , j
0 j 1
Course: Advanced Hydraulics Module 5: Flow through Non-Prismatic Channel
Section
Figure 5.10 A typical channel reach (Chaudhry, 1994)
Assuming i , j i , j 1 therefore the equation can be now written as
Qi2, j 1 Qi2, j 1 Qi2, j 1 ni2, j 1 Qi2, j ni2, j
Fi , 2 j 1 z i , j 1 z i , j y i , j 1 y i , j i xi , j 1 xi , j 2 2
2 2 1.33
2 gA 2 2 gA 2
2 1.33
i , j 1 i, j 0 i , j 1 i , j 1 C 0 Ai , j Ri , j
C A R
A assuming that there is no lateral inflow, the continuity equation between section j and j+1
is written as
Fi , 2 j Qi , j 1 Qi , j
Each channel is subdivided into Mi reach where i is the channel number and i varies from
1 to N. For each reach two equation, namely energy equation and continuity equations
are available. Therefore total we have 2M i M i 1 M i 2 ...M N equations. However,
we have 2M i M i 1 M i 3 4 unknowns therefore; we need 8 additional equations to
solve these unknowns. These equations are obtained by using continuity of flow and
water depth at each internal junction and at the end junction, we have boundary
conditions.
At the downstream end junction of channel i+3 for the sub-critical flow the flow depth yd
and discharge Qd are to be specified. Therefore we get following equation
Fi 3, 2 M i 3 1 Qi 3, M i 3 1 Qd
Fi 3, 2 M i 3 2 y i 3, M i 3 1 y d
Additional six equations are evaluated by considering energy and continuity equation at
junction Jn1 and Jn2.
Course: Advanced Hydraulics Module 5: Flow through Non-Prismatic Channel
Section
F jn1 ,1 Qi , M i 1 Qi 1,1 Qi 2 ,1
Qi2,M i 1 Qi21,1
F jn1 , 2 z i ,M i 1 z i 1,1 yi ,M i 1 yi 1,1 1 k
2 gAi2,M i 1 2 gAi21,1
Qi2, M i 1 Qi2 2,1
F jn1 ,3 z i , M i 1 z i 2,1 y i , M i 1 y i 2,1 1 k
2 gAi2, M i 1 2 gAi2 2,1
Figure 5.11 A Typical Channel Junction (Chaudhry, 1994)
Similarly, the equation at junction 2 can be written as
F jn 2,1 Qi 3,1 Qi 1, M i 1 1 Qi 2, M i 2 1
Q i21, M i 1 1 Qi23,1
F jn2 , 2 zi 1,M i 1 1 zi 3,1 yi 1,M i 1 1 yi 3,1 1 k
2 gA 2i 1, M 2 gAi23,1
i 1 1
Qi2 2 ,M i 2 1 Qi23,1
F jn2 ,3 zi2,M i 2 1 zi3,1 yi2,M i 2 1 yi3,1 1 k
2 gAi2 2 ,M i 2 1 2 gAi23,1
Course: Advanced Hydraulics Module 5: Flow through Non-Prismatic Channel
Section
The sets of equations are then solved simultaneously, firstly the initial values of discharge
and depth at each of the channel is assumed. In case of complex channels network where
flow direction opposite to that assumed can exists the term Qi2, j can be written as Qi , j Qi , j
These sets of non-linear equations are solved using Newton Raphson method. We
determine the corrections y such that y i1, j y i0, j y i , j . Where yi1, j is the better estimate
of the flow depth/discharge at the section and yi , j is the flow depth/discharge at previous
0
iteration.
Fi ,1 Fi ,1 Fi ,1
...
y i ,1 y i , 2 y i , M 1
y i ,1 Fi ,1
Fi , 2 Fi , 2 Fi , 2
y .. ... y i , 2 Fi , 2
y i , 2 y i , M 1
i ,1 . .
. .
. . . .
. .
. . .y .F
Fi , M 1 Fi , M 1 Fi , M 1 i , M 1 i , M 1
. ... .
y i ,1 y i , 2 y i , M 1
The first term on the right hand side is known as Jacobian matrix. This matrix contains
the partial derivative of the function with respect to flow depth and discharge.
For the energy equation the partial derivative are written as
Fi ,l i Bi , j 2ni2 xi , j 1 xi , j Ri , j ni2 Bi , j xi , j 1 xi , j
1 Q 2
y i , j gA 3
i, j
3C 2 2
A R 2..33
y C 2 3
A R 1..33
i, j 0 i, j i, j i, j 0 i, j i, j
Fi ,l i ni2 xi , j 1 xi , j
2Qi , j
Qi , j 2 gA 2 2 C 2 2
A R 1..33
i, j 0 i, j i, j
Course: Advanced Hydraulics Module 5: Flow through Non-Prismatic Channel
Section
Fi ,l i Bi , j 1 2ni2 xi , j 1 xi , j Ri , j 1 ni2 Bi , j 1 xi , j 1 xi , j
1 Q 2
y i , j 1
i , j 1
gA 3 3C 2 2
A R 2..33
y C 2 3
A R 1..33
i , j 1 0 i , j 1 i , j 1 i , j 1 0 i , j 1 i , j 1
Fi ,l i ni2 xi , j 1 xi , j
2Qi , j 1
Qi , j 1 2 gA 2 2C 2 2
A R 1..33
i , j 1 0 i , j 1 i , j 1
Here subscript l is the equation number.
For the continuity equation the partial derivatives are written as
Fi ,l 1
1
Qi , j
Fi ,l 1
1
Qi , j 1
Similarly, the partial derivatives of the equation at the junction is written as
F jn Bi , j
1 Qi2, j 3
y i , j gA
i, j
F jn 1 k Bi 1,1
1 Qi21,1
y i 1,1 gA 3
i 1,1
F jn Qi , j
2
Qi , j gA
i, j
F jn 1 k Qi 1,1
Qi 1,1 gAi21,1
Course: Advanced Hydraulics Module 5: Flow through Non-Prismatic Channel
Section
For a complex channel network, the terms of the Jacobian matrix may not lie nearby the
principal diagonal. Therefore, it increases the memory/ storage requirement, as all the
terms of the matrix is needed to be stored. There are a number of algorithm which try to
minimize the bandwidth of the Jacobian matrix. However, there is no generalized
procedure for arranging the governing equations to obtain a Jacobean matrix of minimum
bandwidth.
Course: Advanced Hydraulics Module 5: Flow through Non-Prismatic Channel
Section
Objective Questions:-
1 Examples of channel transitions are
a) contractions
b) expansions
c) bends
d) All of the above
2 Flow developed due to sudden transition is
a) Gradually varied flow
b) Spatially varied flow
c) Rapidly varied flow
d) Uniform flow
3 Following factors affect the flow through culvert
a) Geometry
b) Bottom slope
c) Tail water conditions
d) All of the above
4 Flow through culvert is
a) Upstream controlled
b) Downstream controlled
c) Both
d) None of the above
5 When the culvert is partially full, the flow will be
a) Subcritical
b) Supercritical
c) Critical
d) Uniform
Course: Advanced Hydraulics Module 5: Flow through Non-Prismatic Channel
Section
6 At a bridge site the flow will not choke if
a) z BB E BB z BB E crit
b) z BB E BB z BB E crit
c) z BB E BB z BB E crit
d) All of the above
Where BB is the bridge section.
Answers:-
1(d) 2(c) 3(d) 4(c) 5(a) 6(c)
Course: Advanced Hydraulics Module 5: Flow through Non-Prismatic Channel
Section
Subjective Questions
1 What do you mean by sudden transition in an open channel explain with sketch.
2 What are Culverts. What they are used for.
3 Differentiate between inlet controlled and outlet controlled culverts.
4 Write down the difference between the flow through obstructions and flow through
bridge.
5 A 2m wide rectangular channel carries a discharge of 15 cumec at a depth of 2.5m.
There is a step rise of 0.1m in the channel bottom. Assuming there is no losses in
the transition; determine the flow depth downstream of the bottom step. (Ans.
1.41m and 2.31m).
6 A discharge of 2.0 cumec flows through a rectangular box culvert having D=1.5m,
b=1.0m, L=40m, n=0.013, and S=0.002. Outlet of the culvert is submerged with
the tail water head of 1.5m. Determine the headwater depth. Take ke =0.5. (Ans.
1.795m)
Course: Advanced Hydraulics Module 5: Flow through Non-Prismatic Channel
Section
References:-
1. Akan, A.O. 2006, Open channel hydraulics, Elsevier, Canada
2. Chaudhry, M.H., 1994, Open-Channel Flow, Prentice Hall of India Pvt. Ltd. New
Delhi.
3. Chow, V.T., 1959. Open-channel hydraulics, McGraw Hill, New York.
4. Formica G. 1955. Esperienze preliminari sulle perdite di carico nei canali dovute a
cambiamenti di sezoine (Preliminary tests on head losses in channels due to cross
sectional changes), L'Energia Elletria, Milan 32 no. 7 (July):554.
5. Nagaratnam,S., 1976. Fluid mechanics, Khanna publishers, Delhi.
6. Normann, J.M., Houghtalen, R.J. and Johnston, W.J. (1985). Hydraulic Design of
Highway Culverts. Federal Highway Administration, Hydraulic Design Series No. 5,
McLean VA.
7. Ojha, C.S.P., Berndtsoon, R., and Chandramouli, P.N., 2010. Fluid Mechanics and
Machinery, Oxford University Press, India.
8. Subramanya, K., 1991. Flow in open channels, Tata McGraw-Hill New Delhi.
9. US Army Corps of Engineers (2002). HEC-RAS river analysis system, Hydraulic
Reference Manual. Hydrologic Engineering Centre, Davis, C.A
10. Yarnell, D.L., (1934). Bridge piers as channel obstructions, Technical Bulletin no
422, U.S. Department of Agriculture, Nov.