Pipe Flow Design for Water Distribution 1

[RJB32482]

Hi Everyone,

I am designing a system where there is a tee in the system and two water pipes branch off of the main header. The specs at design flow rates would be..

Q1=100 GPM delta P1=15 psi
Q2=75 GPM delta P2=22 psi

So for this design off the tee, I would need to add a flow orifice or size the piping differently to achieve the design flowrates. If I kept it this way, wouldn't Q1 be higher than 100 GPM and Q2 be lower than 75 gpm since the presure drop would need to be the same to get these rates? I'm assuming pressure in the line before the tee is 22 psi.

Thanks

 

[MikeHalloran]

For balancing multiple flows by sizing pipe or adding orifices, I find it helpful to compute a flow resistance for each line. Okay, this is really a conductance:

Cv = gpm/sqrt(psid)

[ Yes, it's the same formula used for computing Cv for a valve. The _incredibly_stupid_ convention of defining Cv as "gpm at 1 psid" totally obscures the real units of the measure, because 1 is its own square root. ]

From there, the problem is just like sizing a resistor network to get a desired current ratio, except of course for the pesky square-law nature of the resistors.

Use a spreadsheet, and carry the units too, and you'll be okay.



Mike Halloran
Pembroke Pines, FL, USA

 

[RJB32482]

So the big question in this problem is if I don't add a orifice in the line with the least pressure drop (Q1), would I lose the Q2 flow or would Q1 just be higher (that would not be a problem).

Thanks.

 

[MikeHalloran]

Kirchoff's laws apply.

Which means the flows will balance such that the pressure drops are equal.

You may adjust the pressure drop in either line by changing its size, or the number of elbows, or by inserting a restrictor.



Mike Halloran
Pembroke Pines, FL, USA

 

[RJB32482]

Still don't understand the reasoning why the Q2 value would not be 75 GPM even though the branch off of it would have a lower pressure drop. If 175 GPM is flowing in the pipe where the tee breaks off, then at 22 psi into the tee, 75 GPM should flow into line 2 and the rest of the flow (175-75=100) should flow into the branched line. I thought pressure drops only need to be the same if they tie back into the same header at a certain point. These 2 lines are used by atmospheric sources and don't get tied into a common header. Please comment.

Thanks.

 

[bimr]

No sure from your posts if this is some theoretical concepts or you are designing a real system.

If you actually designing a system to split the flows, you should use control valves. Control valves will allow you to maintain constant flows regardless of changes in pressure in the lines.

 

[proinwv2]

Watch out for that Cv equation. It really isn't used to size valves because it only works for certain conditions.

See ISA S75.01

Paul Ostand

http://www.ostand.com

 

[MikeHalloran]

It's convenient to state the flow resistance (conductance, really) of individual pipes in a network as 'flow at pressure drop', which is what I assumed you had done.

That does not imply or guarantee that each pipe will achieve the stated flow when they are connected together; it's just a way of stating their relative size.



Mike Halloran
Pembroke Pines, FL, USA

 

[katmar]

RJB32482,
You can only have one pressure in the tee where the divergence occurs. This is your 22 psi. You stated that both lines discharge to atmosphere. This means the pressures at the discharge points for the two lines are the same. If the two flows come from the same pressure (22 psi), and discharge to the same pressure (0 psig), then the two lines have the same pressure drop.

If the lines, as constructed, will not give the desired flows with this fixed pressure drop then you have to add in valves or orifices to dissipate the additional pressure drop, or you must accept that the flows will be higher than specified (but this may mean upstream problems and changed pressure drops there).

Harvey

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[RJB32482]

Thanks for all of the posts.

So the line with the less pressure drop (15psi) will get more than the design 100 GPM flow, correct? And the line with the design pressure drop (22 psi) will get the design flow (75 GPM). This is assuming that 22 psi is available right before the tee. This condition would not hurt the system, probably help it.

Thanks.

 

[MikeHalloran]

Would you mind divulging the nature of the water source?



Mike Halloran
Pembroke Pines, FL, USA

 

[RJB32482]

The water is coming from a treated water line in the main header. The branch is going to a heat exchanger and the end user is to fill up a water tank. Flows are as stated and water is at 60F. Looks like from measurements plant pressure at tie point is around 30-35 psi.

Thanks.