Power in Parallel Pumping 1

[impeller1]

Good day All,

A question here..I have 6 vertical pumps in parallel. Now, all 6 were fitted with slightly larger impellers. With this upgrade, their BHP's have gone up to around 600 horsepower. From the attached sketch you can see the arrangement. All 6 discharge into a large 42" header which splits into 2. The header then goes to a bank of filters entering on either side. All the pumps, with the exception of one, are operating at around 595 horsepower with a discharge pressure of 82 psi. Pump C, from the sketch, is operating at 615 horsepower with a discharge pressure of 86 psi. All 6 pumps were sent at the same shop, and all 6 had more or less the same repair work done to it and fitted with the same sized impellers and suction casings. Can anyone explain why this pump is drawing more power? I included the sketch because I'm not sure if its position on the header has anything to do with it??

Thanks!

 

[cvg]

open the valves fully and look at the check valve on pump c to see if it is stuck

 

[BigInch]

Its position on the header would make it have one of the highest system heads against which it would have to discharge, as well as F's. Since F is apparently not affected, you might check the suction draw. Is there vortexing going on there that would reduce the effective suction pressure by 4 psi?

Another possibility. Is the Tee at C too close to the end of the header? It is discharging into a dead end region, so it may be experiencing some additional head due to turbulence in the header. A Tee is not as streamlined as an ell fitting, so extra pressure may be needed there. How close is the end of header at pump C when compared to the distance to the end of the header at pump F?

Other than that, you've only have 1 of 6 pumps with less than a 5% difference, which is well within typical variance. Keep your eye on it, and if it doesn't get worse, don't worry too much for now.

**********************
"The problem isn't finding the solution, its trying to get to the real question." BigInch

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

 

[impeller1]

Thanks for your advice BigInch. I suspected the location had something to do with it, but I'm not too clear on the theory behind it. I checked Pump C and Pump F's location from the end of the header and, as you suspected, Pump F is about 36 inches away from the end of the header, while Pump C is only about 15 inches away. But just one thing, if Pump C's pressure would have to be higher to deal with the extra head due to turbulence, won't it move back on its curve and cause the BHP to decrease??? Also, do you have any literature on how a pump would behave being so close to the end of a header?

I would really appreciate any further help!!!!

Thanks

 

[cvg]

you really need to have flow measurements and a certified pump curve. It would have been nice to have certs for all the pumps

 

[impeller1]

Unfortunately we don't have any way of telling the individual flows apart from checking the power and reading it on the curve which isn't very accurate.

 

[CH5OH]

you could balance discharge header by connecting both dead ends with same size pipe as discharge lines pumps, if difference in operating points bothers you.

 

[cvg]

with proper flow measurements and pump curves plus accurate estimate of the headlosses for each pump discharge line, you could maybe come up with a reasonable explanation. Without such data, you will have a guess. It is very likely that all six pumps have slightly different pump curves as no two pumps are ever the same. I know your schematic is not to scale, but pump c does not appear to have the highest loss of all the pumps. It looks like pumps D, E and F all have a longer distance to pump plus two additional 90 degree bends. I would not expect all of the pumps to be running at the same power level nor would I expect pump C to be the highest.

 

[BigInch]

Pump A & B are pumping into a moving slipstream, C is pumping onto the opposing wall and creating a vortex that crosses around the inflow jet. That activity would raise the head loss of T and increase the discharge pressure of C.

Sorry no text on the subject.

**********************
"The problem isn't finding the solution, its trying to get to the real question." BigInch

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

 

[Artisi]

The problem maybe an over performing impeller which is possible within a 6 pump arrangement. It follows that, the head on pump C will be slightly higher than B and higher again compared to A, and if over performing in terms of flow then a few more HP is not unrealistic. Alternately it could be a mechanical problem, you haven't said what pump style they are, if mixed flow with wear rings then maybe insufficient wear ring clearance, if axial flow then maybe insufficient clearance between the impeller and impeller ring. But as pointed out already, doesn't sound like a big deal but worth monitoring.

 

[impeller1]

Makes sense..but again, I have no idea of determining what my actual flow is off that pump C. Artisi, they're axial flow pumps. Workshop reports state the clearance they used was the same clearance as all other 5 pumps with respect to the impeller and case wear ring.

Also, if pump C is operating at a higher discharge pressure, won't it mean that the pump will operate to the left on its curve therefore reducing the power?? This is what confuses me.

And well here's the other part..The pump manufacturer themselves made these impellers for us, however, they botched up their calculations. They were supposed to design an impeller to allow an increase of flow from 8000 usgpm to 10,000 usgpm at a power of 595 hp. However, when it was all said and done, the actual performance is 9,600 usgpm @ 600 horse. The constraint here is that my motors are rated at 600 horse! This is why I do not want to exceed that 600 horsepower limit.

 

[Artisi]

Seems someone miscalculated somewhere.
The confusion regarding the increased power for increased head is a function of axial flow impellers, unlike low specific speed pumps (centrifugal) where power decreases as head increases.
As we know know that these are axial impellers it seems that it could well be a function of header hydraulic / pump outlet problem which you are stuck with.

The cure for pump C looks to be a reduced diameter impeller and a new impeller wear ring to suit the new diameter. A small diameter change will hardly be noticeable in terms of flow but give a bigger change in power input, remember power chages at the cube of the diameter change.

 

[whammett]

I agree 100% with Artsi. While C would have higher head pressure, so would F, so it negates the fact that it could be due to the position it is in. It COULD be several of a few smaller, negligible problems, but IMO Artsi hit the nail on the head with the oversized impeller idea.

BigInch also had a good point in which if it says within 5% deviation from the other points, why worry? Just keep an eye on it.

"Scientists dream about doing great things. Engineers do them." -James Michener

 

[CH5OH]

Also, if pump C is operating at a higher discharge pressure, won't it mean that the pump will operate to the left on its curve therefore reducing the power?? This is what confuses me.

the horse power, might it be derived from amps measurement?
if so, current might have gone up, together with cos phi.
so actual hp or kw might have gone down

 

[Artisi]

CH5OH.
On axial flow impellers, increase head with reduced flow means increased power input.

 

[CH5OH]

artisi,
I was just quoting the OP, commenting his question

 

[impeller1]

Artisi, So Axial flow impellers do not follow standard pump curves?? If you're saying for increased head and reduced flow, the power increases.

And the thought of a larger impeller crossed my mind..but I'd like to think the manufacturer made all 6 impellers exactly the same.... but then again, they did botch up their calculations so anything's possible.

 

[rshayesteh]

since the pump C is located at the end of line, you may also check if something has not blocked its discharge pipe or tee (or the main header before pump B)leading to higher pressure drop and head.
Do you have any record of pump powers before re-impllering?

Reza

 

[Artisi]

impeller1, have attached info on the increase of power as head increases on axial flow pumps.

Bottom left (sorry not all that clear) shows impeller shape as Specific Speed increases, axial flow impellers are about 9000+, bottom right indicates typical power curves for various Specific Speeds, curve No5 is typical axial flow and shows increasing power as flow reduces.

 

[whammett]

Star for Artisi. He is correct, but then the question arises. why is flow in pump C decreasing?

Can you extend your original rough P&ID to include piping upstream of the said bank of pumps?

"Scientists dream about doing great things. Engineers do them." -James Michener