Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations KootK on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Pressure drop calculation (parallel pipes) 1

Status
Not open for further replies.

Skatesy

Mechanical
May 13, 2005
1
I am working on a project whereby it is planned to increase the flow of water through a manifold which has 24 parallel pipes (and heat exchangers) running from an inlet manifold to the return manifold. It is known what flow is required through each pipe. The flows will be balanced using gate valves. I am trying to establish typical loss values or values of hydraulic resistance which would give those flows through each pipe leg. This is simply for pump selection and only needs to be a rough guide. can anybody recommend a suitable technique for doing this?
 
Replies continue below

Recommended for you

Skatesy:

You should get a copy of the Henry Vogt forged steel valve Catalog F-11 (1980). On page 294 you will find a complete explanation on why it is most advantageous to use the Cv factor when dealing with a Parallel network of fluid piping. In their Table 4 they list the typical Cv factors for gate valves.

The following equations utilize the Cv factor for determination of fluid flow or pressure drop. They are essentially the same equations as those utilizing the "K" factor; however the K factor has been converted to the Cv factor by the conversion equation: Cv =29.9*d^2/(K)^0.5

The equations are:
Cv(eq.) = Cv1 + Cv2 + Cv3 + Cvpipe = Equivalent flow coefficient for the parallel network

Q = Cv(eq.) * (deltaP/S)^0.5
where,
Q = total flow, gpm
deltaP = pressure drop from inlet manifold to outlet manifold, psi
S = liquid specific gravity (relative to water @ 60 oF)

By using Cv(eq.), the total flow rate through the parallel piping network cna be determined.

I hope this helps.




 
Well, you used the phrase "and heat exchangers", which can change the design significantly.

The most detailed approach is: for each of 24 channels with heat exchangers, plot a curve of pressure drop DP vs asumed flow thru the channel. This will involve assuming several flow rates, perhaps 0.4Wa, 0.6Wa, 0.8Wa, 1.0Wa, 1.2Wa, where Wa is the average flow , the calculating the heat transferred for each flow in each channel, and the associated pressure drops for each case.

Next, you construct an ensemble pressure drop vs flow curve, that is for each pressure drop, add up all the flows from each of 24 channels to yield a totla flow vs presure drop curve. This curve is then plotted on the same graph as the pump characteristic curve,and the intersection of these 2 curves is the balanced flow for the system.

Knowing the original channel characteristic curves and determine the expected flow in each channel.
 
The system you described sounds a lot like a closed loop circulation system on a combined cycle power plant. In those cases, each of the heat exchangers has a different design flow and design pressure drop. If they are all in parallel, the one with the most pressure drop will govern, and you would throttle the inlet valves of the other 23 channels to that each leg would have the same pressure drop as that of the HX with the max drop.

During startup , the service engineer would either set the valves' throttling loss for this design pressure drop, or install temporary ultrasonic flow meters and set the valves to provide the required design flowrate for each HX.

If this is the case, there is no need to compute the chraracteristic curves discussed earlier.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor