question on manifold pipe 2

[jorgefherrero]

I have a question regarding manifold piping, I would like to know the how head and flow behave in branches of a manifold.
It is a reverse osmose plant, fed with a 6" ANSI Sch40S pipe, with five branches, 2 1/2" each. Each branch feeds reverse osmose vessels. The total flow is 190 cubic meters per hour. (Pressure is 70 bar, around 1000 psi) After the last branch the manifold ends with a welded cap.

I suppose pressure increases along the manifold, as the flow through it gets smaller, so the last branch would have the highest flow. Do you know the formula or similar to estimate this, as well as the head losses involved?

Regards

 

[Latexman]

jorgefherrero,

In my Chemical Engineers' Handbook it's under the topic of "Fluid Distribution". Basically, you assume a design or in your case you calculate the flows and pressure drops for an existing system due to friction and momentum recovery. Then you calculate the % maldistribution between the first and last branch. Make modifications until the maldistribution < 5%. There's several equations and my book is at work, so I'll leave it to you to find a copy or a different reference.

Good luck,
Latexman

 

[katmar]

Latexman is correct in the general sense, but in your particular case where the pressure is so high because of the high pressure drop across the membranes any pressure drop differences between the branches of the manifold become negligible. If the membrane vessels are identical you can safely assume equal flow through each.

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[BigInch]

True as Katmar says, your pressures are high, so you probably won't notice the small losses along the manifold too much. But if you want to calculate it, do like this;

Velocities will decrease with reduced flow, if pipe x-sectional area remains the same and increase if x-sectional area reduces, both in accordance with V=Q/A.

Pressures will not increase along the manifold unless you have a reduction in x-sectional area such that the change in velocity head, v^2/2/g, is greater than the energy lost to flow resistance through the reduction. In a header of uniform x-section lessening resistance to lesser flow will decrease the corresponding pressure loss per unit length and pressures at each branch will decrease.

Choose a pipeflow pressure loss equation suitable for your flow, Darcy, Churchill, etc. to calculate these pressure drops in each segment of the header and determine the pressures at each inlet to a branch off the header.

Additionally I'd suggest you balance the manifold with the inlet in the center, spreading to two caps on each end of the manifold. Uniform flows to each branch will be easier to maintain. Also goes for the outlet manifold on the other side of the RO equipment, if they also recombine into an outlet manifold. A lot of the time it is easier to fabricate a header of uniform pipe diameter rather than have lots of reductions to weld in and inspect, but that depends on your configuration, so you decide.

http://virtualpipeline.spaces.msn.com

"What gets us into trouble is not what we don't know, its what we know for sure" - Mark Twain