re: PRV Selection

[TurbineBlade]

Hi,

I am performing a temporary modification to create a tie-in to an existing cooling water system (150 psig, 100C) involving adding a small booster pump (operating point at 20 psi, 11 L/s) in a branch circuit (3" line). I plan to use the existing system design pressure and temperature for the modification. The normal system operating pressure is 100 psig.

1) If the existing system operates say (144 psig) due to an upset, which is just below the setpoint 145 psig of the existing RV. My modification with the booster pump operates at 144+20 = 164 psig, which exceeds the design pressure. I have no place to divert the water to a drain from the pump discharge. I put a PRV in the pump re-circulation, which set to open at 145 psig and divert the fluid back to the pump suction at 5 L/s + the operating path (11 L/s) = 16 L/s @ 5 psi. The pump will be at run off condition. This will keep the pressure within the design pressure.

Since the back pressure is higher than 50%, the vendor suggests using a pilot operating PRV (1.5"x3"). I would like to ask if there is other way to solve this problem as I read putting a RV in the pump re-circulation line which is not a common practice.

Turbineblade

 

[georgeverghese]

a) At relief, this pump is only seeing 1psi dp (145 -144), so is the flow for this pump = 16 l/sec at 1psi dp ?

b) You say the duty for this PRV is 16 l/sec at 5psi dp, but this 5psi dp is not consistent with the pump dp of 1 psi at this relief condition

c) From the limited information here, it looks like you've got no headroom to make a PRV work if you are looking at a recirculation scheme for the PRV. It is not advisable to drop a relief stream on closed loop cooling water to drain. Can we not return this PRV dicharge back to the cooling water expansion drum ?

 

[e43u8]

TurbineBlade,

Please upload the pump curve if it's centrifugal...
Is 5 psi differential pressure of pump at 16 L/s? If so, then the pump suction pressure or PRV back pressure at relieving conditions should be 140 psig. Did you consider the effect of such huge back pressure on the PRV capacity even to be pilot operated type? Did you obtain the 5 psi dp around PRV by trial and error? Please clarify mentioned points to get proper and precise response...

 

[TurbineBlade]

Hi all,

Thanks for all the responses.

Georgeverghese,

The pump with the re-circulation will run at 16l/s at 5 psid.

e43u8,

Yes, the pump is centrifugal and the back pressure is high. I re-run the hydraulics by trial and error to get the 5 psi differential pressure, but the pump is at the run out condition. I will re-check if the upstream system if it can ever get to the high pressure of 144 psig.

Turbineblade

 

[georgeverghese]

So the relief flow should be much higher at pump dp of 1.0psi

Also, running this pilot operated PSV with only 1psi across it is impossible - since you havent included frictional losses in the PSV discharge?

As suggested earlier, the only solution I can see is to return this relief stream back to cooling water expansion drum - maybe you have some nearby line which is going back to this expansion drum ?

 

[georgeverghese]

There is an error in my previous advice

Since system pressure is 145psig, we need to compute pump capacity only when pressure downstream is at relieving conditions, which is in this case = 145 x 1.1 = 159.5 psig.

So the differential to be used for computing booster pump capacity at relief = 159.5 - max developed pressure in upstream section, which may be upsteam pump dead head pressure.