Using existing pump for an application the pump is not designed for 1

[icesmith]

Good day! In a certain application, the recommended centrifugal pump has to have head of 220 ft at flow of 1100 gpm. Given the pump is not yet available, we are looking into using existing pumps for the mean time. We have an existing pump with head of 374 ft at 485 gpm. Can we use the existing pump for the said application? Another, existing pumps suction and discharge sizes are at 6" and 4" respectively. And the recommended pump was designed for 14" suction and 12" discharge. Does this difference in suction and discharge sizes make the existing pump not usable for the said application?

 

[LittleInch]

You need to develop a system curve from the pump to the end user. This gives you the head required for a variety of flow rates and is a curve which rises as flow increases (flow on x axis, head on the y axis)

Then overlay the pump curves with total developed head 9inlet head plus differential head) for the various flowrates noting the point at which the curve stops, which is the paractical limit for that pump.

Given the discrepancy between the two pumps, if you only use one pump, it is likely that the pump will go off the end of its curve, vibrate badly and probably trip on excess amps.

You might be able to get 600gpm out of the existing pump by adding a throttling valve or orifice plate to add some resistance which that pump needs so it doesn't go too fast.

Think of this like a car - you have designed it to go at a speed of say 100 mph and an engine to suit. If the car is stuck in second gear and the accelerator is stuck hard down, then the engine rev beyond its limit as there is insufficient air resistance. The engine will make a lot of noise and eventually blow up. However if you then make the car with a big sail on it (adding resistance), the engine will slow down to a more normal speed, but you will still only be doing 45 mph.

You need to use proper data and not just guess here otherwise you will seriously damage your equipment and it won't work.

Provide the information above (system curve and pump curves, and we might be able to figure it out, but this really needs someone who understands this to do it properly.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Artisi]

It seems LittleInch has covered most of the field however, if you are not capable of undertaking the review, post a copy of the pump curve and a detailed description of the system you intend pumping in to - this way we can make some recommendations.

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.)

 

[LittleInch]

The only thing I forgot to say was that in normal operation, where the pump curve crosses the system curve, that is your anticipated flow rate. if the pump curve stops ABOVE the system curve, then you are in trouble unless you make up the difference in head by adding a fixed head difference equal to the difference between the two curves at that point.

If you could use two pumps in true parallel operation, then the parallel pump curve becomes elongated and might intersect the system curve before the end point.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[bimr]

It does not appear that your pumps are a match in head and flow capacity.

Have you considered renting a pump?