Indifferent pumps in parallel

[pumpmann]

Hi
What is the draw backs of installing 2 new pumps in parallel of existing three of different brand and curves not matching 100%. Anything to be check before approving the order?

 

[LittleInch]

Try searching parallel pumps in this forum.

See

http://www.eng-tips.com/viewthread.cfm?qid=348291

for one example.

Basically unless you have some way of"trimming" the pumps your two new ones could either end up doing all the work or no work and actually dead head. Having similar pumps but not quite is probably worse than having a bigger difference.

Key thing is that at the duty point your new pumps are very close to the old ones and it's better to have them as a steeper pump curve than existing.

Even identical pumps from same vendor differ slightly so new pumps have a worse problem.

If you can post pump curves and system curves for both you will get more detailed answers

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[MikeHalloran]

If the pumps don't give a crap, why should I?


Mike Halloran
Pembroke Pines, FL, USA

 

[Artisi]

Mike, very clever.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[GaTechTheron]

Take the pump curves and at each head increment, add the flowrates. Assuming that both pumps pull from the same header (i.e. the suction pressures are identical), the line losses between the pumps and pipe section (where each stream meet) are accounted for, you will have an accurate model of what the pumps' curve will look like while operating together.

At the end of the day, the slope of the curve is what is important (not how similar the curves are). Two pumps with a nearly flat curve operating in parallel will be very difficult to control. Two pumps operating in parallel with a very steep pump curve will have little to no issues of control because the difference in flow (relative to head) means there will be a lower difference in flowrate with a small difference in head. Thus, there is a lower chance than one pump is at dead head and you know it.

Clear as mud, I know.

 

[jimhokie]

To add slightly to what GaTechTheron said, wherever we have pumps in parallel, we specify that the pump curves must be "continually rising to shutoff" for just the reasons he explained.