Pumping in Series

[dg47]

What are the concerns with two pumps in series if they are not identical pump sizes? Looking into the feasibility of removing a sump from the system and just pumping from one pump to the suction of the next. The pumps are also about 2000 feet apart. Thanks

 

[BigInch]

be sure the inlet pressure rating of the 2nd pump is higher than the maximum discharge pressure of the first.

Both pumps should have reasonable efficiencies at the same flowrate. That characteristic should extend over your entire operating flowrate range.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[BigInch]

And of course that the new discharge pressure (including water hammner effect) of pump 2 does not exceed the pressure capacity of the downstream pipe.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[Artisi]

The second pump should be operating at or near to its BEP inline with the anticipated flow and head. High pressure at the shaft /casing interface on the second pump needs to be considered as does the starting sequence for the second pump.

 

[itsmoked]

Yeah especially that case pressure of the second pump. I watched in horror as I did exactly that to a brand new ABS bodied pump. Kerrr-split. DOH!

Keith Cress
kcress -

http://www.flaminsystems.com

 

[BigInch]

Ouch!

Had 2 x 5000 HP 7 stage axial w/ outlet flange ratings 1500# .... inlet flanges were 300# ... better not to assume they're equal. Needed to install a suction PSV on #2 for upstream deadhead pressures.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[SNORGY]

You folks probably commonly deal with much bigger pumps than I do.

For smaller pumps at least, I wouldn't put two pumps in series if one or both had a flat characteristic. This, in my mind, would give rise to the potential for flowrate surges and probable pulsations or oscillations in loads. This is probably more a concern in single stage pumps. In any event, if they had to be configured in series, I would try to put the downstream pump on flow control, and I would consider running the upstream pump somewhat (not too much) left of BEP if it could take it.

On shut-down sequence, I would include a timer in the control logic to shut down the downstream pump first, and a timer on start up to start the upstream pump first. This avoids the potential for the line pressure between pumps to drop to a point where you would get a big "bang" (hammer) on start-up. Of course you would need to look at all the system components to make sure that they could take the short duration pressures involved, and safeguard accordingly.

I was involved in a failure cause analysis for a system consisting of a centrifugal booster in front of a high pressure plunger pump for water injection (cavern wash). When initially put into service, on shutdown, both motors cut out but, whereas the centrifugal stopped pretty much right away, the plunger pump went through 11 full rotations before reaching full stop. There was some pressure recovery in the interconnecting piping by sucking water through the booster, but on start-up the booster had to start first. The result sounded like someone was taking a sledgehammer to the piping. It wasn't long before I was in the field looking at pieces of plungers and broken crankshafts from a disintegrated plunger pump. There were other issues at play, but the "Big Bang" certainly played a part.

Putting start-up and shutdown timers in place eliminated that part of the problem.

For the record...I was not involved in the original design.

Regards,

SNORGY.

 

[BigInch]

Flat curves can be a control problem for any type installation, if your system curve isn't close to vertical. A flat pump curve and a flat system curve don't give much of a stable control signal, as a large variation in flow produces little or no variation in pressure, or v/v.

Downstream pumps work better IMO, if they are suction and discharge pressure controlled and allowed to float around whatever flowrate the first pump is sending it. A start signal or interlock when adequate suction pressure is available works better than a pure timer control. In fact in most cases they can be started and running, on recycle (if necessary) until suction pressure builds up enough to start closing the recycle valve... using no timer at all.

Shutdown the same but in reverse. Just shut down on lo-lo suction pressure (a value higher than NPSHr) and of course on hi-hi discharge pressure too.

Yes, positive displacement pumps can be highly suceptible to NPSH problems and hammers, as they basically are just a series of big pressure pulses in both suction and discharge lines anyway. Surge analysis for start and stop isn't optional on a big installation.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/