0% found this document useful (0 votes)
4 views

PipeFlow_3

Uploaded by

bhavins22
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
4 views

PipeFlow_3

Uploaded by

bhavins22
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 8

Examples: Flow through a single pipe

1. Finding discharge: The head loss for flow of oil (kinematic viscosity, /= 10−5 m2/s) through a
galvanized iron pipe (D= 100 mm, = 0.15 mm) is 60 m per km length. What is the flow rate?
Solution:
; m/s and Q=0.0159 m3/s

2. Finding diameter: What size of asphalted cast-iron pipe ( = 0.12 mm) is required to carry
water at a discharge of 0.1 m3/s with a maximum head loss of 20 m per km?
Solution:

Assume f = 0.02, then D = 0.242 m => Re=5.27 x 105 => f = 0.0177


which gives D = 0.236 m => Re=5.40 x 105 => f = 0.0178; D = 0.236 m => Re=5.39 x 105 => f = 0.0178.

Direct solution: = 0.241 m

Provide the next larger available diameter


STEADY pipe flow: Flow through multiple pipe
• Pipes in series: Discharge is same, Head loss is sum of the head losses
• Finding Head loss- Simple. Re and /D are known, find f and then hl for each pipe and add
• If minor losses (at entrance, exit, expansion/contraction) are significant, these should be added also
• Finding Discharge- Iterative solution. Assume large Re, /D is known, find f, compute Q. Repeat till
convergence.
Example: Two reservoirs with a water-level difference of 50 m, are connected through two pipes in series. The first pipe
has a length of 1 km, diameter of 20 cm, and roughness height of 0.1 mm; and the second pipe has L=2 km, D=50 cm,
and  =0.2 mm. Estimate the discharge through the pipes. Ignore all minor losses.
Solution: From the Darcy-Weisbach eqn,
Assume large Re, then m3/s
Next iteration: Re1=678957, Re2=271582 m3/s
Next iteration: Re1=661916, Re2=264767 m3/s
Next iteration: Re1=661570, Re2=264628 m3/s
STEADY pipe flow: Flow through multiple pipe
• Pipes in parallel: Head loss is same, Discharge is sum of the discharges
• Finding Discharge- Simple. Similar to single pipe problem of finding Q for given hf
• Finding Head loss- Iterative solution. Assume large Re, find f, distribute Q. Repeat till convergence.
Example: A pipe carrying 0.5 m3/s of water is split into two parallel pipes. The first pipe has a length of 1 km, diameter
of 20 cm, and roughness height of 0.1 mm; and the second pipe has L=2 km, D=50 cm, and  =0.2 mm. Estimate the
head loss and discharge through each pipe. Ignore all minor losses.
Solution: Head loss is same, m3/s
Assume large Re, then m3/s
Next iteration: Re1=389955, Re2=1117257 m3/s
Next iteration: Re1=383602, Re2=1119799 m3/s
Next iteration: Re1=383424, Re2=1119870 m3/s
Therefore, discharges in the pipes are 60.23 litre/s and 439.77 lit/s. Head loss is 16.89 m.
STEADY pipe flow: Flow through multiple pipe
• Branching pipes (three reservoir problem): Direction of flow (into/out of middle reservoir) is
not known (NOTE: only direction-finding is in syllabus. Rest is for information)
• Use sign conventions as follows
• For head: Take datum at level of middle reservoir (B); hA (+ive), hC (−ive), hJ will have some value between hC and hA
• For discharge: flow towards the junction positive; QA(+ive), QC (−ive), the sign of QB opposite to that of hJ
• hJ will adjust such that the continuity at the junction is satisfied:
• Iterative solution:
• First determine if hJ is positive or negative- Assume hJ equal to zero, find QA0 and QC0. Then, Sgn(hJ)=Sgn(QA0+QC0)
• If QA0+QC0 is zero, no iterations are needed. There will be no flow into or out of the middle reservoir
• Start with a reasonable guess value of hJ, find QA , QB , and QC , and iterate till continuity is satisfied at the junction
• Using the expressions for obtaining discharge for a given head loss, we may write
• , where and
• Continuity equation:
• May be solved by the Newton method: An approximate derivative could be written as

Example: The water level in three reservoirs are at 30 m, 18 m, and 9 m, respectively. The reservoirs are connected
to a junction through pipes having the following properties: L=1 km, D=30 cm,  =0.1 mm; L=500 m, D=15 cm, 
=0.2 mm; L=300 m, D=20 cm,  =0.3 mm. Estimate head at the junction and discharge through pipes. Ignore minor
losses.
Solution: We have hA=12 m, hC =−9 m. The coefficients are . Similarly, ; . With hJ=0, we get m3/s and m3/s => hJ is +ive.
Assume starting value as half of hA, i.e., 6 m. It gives m3/s and m2/s, resulting in hJ=5.041 m. Next few iterations result in
hJ=5.056, 5.055, 5.056, 5.056 m. The discharges are 0.11093, −0.02036, −0.09057 m3/s.
Notes: 1. Even if we start with a negative value for the initial guess, it will converge. Finding sign of hj is not
necessary!
2. Another type of the three-reservoir problem involves a demand at the junction, QJ (−ive). The solution procedure
is similar with continuity equation replaced by and the direction of flow for C also unknown.

and
STEADY pipe flow: Pipe Networks
• Pipe networks: Most complicated, generally requires computer-solution
• Small networks may be solved by hand-computation – Hardy Cross method
• Large networks- set of nonlinear equations
• Head loss by simpler Hazen-Williams (H-W) formula (not Darcy-Weisbach equation )
• In SI units: ; C is H-W coeff., typical values 50 to 150 (Higher C for smoother pipe)
• We will use a general form , with r indicating “pipe resistance”, which is not dependent on Q
• For hf in m and Q in m3/s, a 1 km long pipe of diameter 1 m will have r of the order of 1.
• Typical information given for a network would be the supply and demand at “nodes” (or
“junctions”) and the resistance (or, length, diameter, and roughness) of all pipes. Some nodes
may have a water-tank, therefore, head at that node is fixed. The solution provides head at all
nodes and discharge in all pipes.
• The continuity and head-loss equations are used for the solution
• Continuity implies that the algebraic sum of discharges at each node is zero
• The algebraic sum of the head-loss in all pipes in a closed “loop” should be zero
• Hardy Cross method solves the network loop-by-loop using the Newton iteration
• Software (e.g., EPANET) to solve entire network together by forming a set of nonlinear
equations

You might also like