Minimum flow issue 4

[wfjer]

Hi,

I have to give my approval for a proposal for a centrifugal pump, it should work at a rated flow of 38750 gpm @ 157 ft TDH, BUT it must be able to work with a flow of 15000 gpm (minimum). I have received a proposal for a pump that meets these conditions except that its minimum continuous stable flow is 21787.9 gpm.

What are the risks of having a pump working with such a less-than-minimum flow? Can I accept this proposal?

 

[BigInch]

It really depends on the type of centrifugal you have and the robustness of the pump design. If it has nice bearings and a small L/D shaft ratio and unbalanced forces and thrusts are low, you might be able to get away with it for awhile.

**********************
"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/

 

[itsmoked]

BigInch; Do you have that larger? I can't read half the stuff on it even blown up 300%.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[BigInch]

Its in your mailbox.

**********************
"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/

 

[raronen]

Before making a selection, I would recommend calculating the suction specific speed of the pump in question.

S = (N*(GPM^0.5))/(NPSHR^0.75)

*(N speed in RPM)
*(GPM at BEP, for double suction impeller divide total flow by 2)
*(NPSHR in feet)

Studies have shown that pumps with higher suction specific speeds have a much smaller window of reliable operation around their BEP. Depending on the guideline you are looking at, typical industry sources suggest S should be less than 11000 or 9000. Because you are looking at such a wide range of flows, I'd shoot for a pump with S less than 9000.

For a really high flow application like the one you are working on, it is quite difficult to find a pump with S < 9000 because you are fighting 2 parameters.

First, you need the NPSHR low enough so the pump does not cavitate. If you are restricted on suction head, your pump selection will be very difficult.

Second, the GPM is fixed, so you can vary the speed. A lower speed pump will be more reliable, but it will also be much larger, more expensive, take up more space, and more expensive to install.

The third factor you might be able to adjust is selecting a pump with a double suction impeller. This cuts the flow rate in half in the equation.

In any case, if you have a pump with a very high specific speed, at low flows it will likely have higher vibration, more internal recirculation (and the potential for recirculation cavitation damage), and long term it will have higher maintenance costs.

I've worked on a project with high suction specific speed pumps in the past (crude oil offloading pumps to a refinery). In this application, the high suction specific speed values definitely translated into sub-standard reliability.

Here is one reference.

http://www.gouldspumps.com/cpf_0008.html

Type "suction specific speed" into Google and you'll find plenty of others.

Also note, this should not be confused with "specific speed" which is a different topic entirely.

 

[wfjer]

raronen I'm not completely sure if I can apply your formula, it could serve as a reference, but like bigInch says, it depends of the robustness of the pump.

I was looking for a similar pump in the Golds pump selection web page (Goulds is the manufacturer of the pump of the proposal) and I found they have pumps that can handle the conditions I require.

I'm going to ask the vendor for a better selection of the pump, to meet more close the pump conditions.

Thanks for your replies.

 

[macmet]

BigInch. I'd also like to see that picture larger. Could you post a link to it?

Regards.

 

[BigInch]

I can't figure out why I just didn't just attach it here before, like I did now.

**********************
"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/

 

[JJPellin]

I think I must be pickier about this issue than most. I would never consider purchasing a new pump with the intention of running it below the published minimum stable continuous flow. In fact, we have our own method of calculating minimum flow that is more conservative than the minimum stable continuous flow. We require that a new pump meets this higher minimum flow criteria that is based on suction specific speed as called out by raronen above. I would suggest that you need to add a spillback to allow the pump to operate above minimum flow during these times of low demand. Otherwise, you could do a number of other things to get out of this bind. You could buy three smaller pumps and run two out of three during high demand and only one during low demand. Depending on your system, you could use a variable speed driver and slow it down during low demand. All of these options would have to be carefully engineered.

If you choose to run this pump below minimum stable continuous flow, you are likely to experience a number of problems, some of which are called out on the diagram that BigInch provided. Suction recirculation cavitation, shaft deflection, reduced seal and bearing life, increased vibration and noise, inefficiency (wasted energy), and other problems could occur. Depending on the configuration of the pump, the resulting reliability might be acceptable or it might be absolutely horrible.


Johnny Pellin

 

[BigInch]

Wfjer,

Please listen very closely to JJP's advice. I meant to say the same when I said "for awhile" and posted that diagram as evidence, but he said it so much better.

**********************
"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/

 

[raronen]

raronen I'm not completely sure if I can apply your formula, it could serve as a reference, but like bigInch says, it depends of the robustness of the pump.

Suction specific speed is only a reference, however, it is extremely relevant to the pump selection you are facing.

Hydraulically, the lower suction specific speed pump is the more robust design. The stoutness of the rotor and casing are secondary when the impeller is being destroyed by recirculation cavitation.

If the service truly needs to accommodate flows between 15000 and 38750 gpm, JJPellin's suggestion of installing 3 pumps is an excellent idea. Two pumps run in parallel, one spare. The pumps would be rated with BEP of about 19500, so when you were running at 15000, the single pump would only be about 25% below BEP. For a high-flow pump, it's much easier to find something that can meet that requirement, than to find a pump that can operate at 60% below BEP.

For my money, if I'm buying new pumps, reliability is extremely important. Someone at the plant will have to operate and maintain the pumps for the next 20-50 years. That lifetime cost will be much greater than the additional work on the front end to get the design right.

 

[BigInch]

I'm always surprized by the lack of attention paid to selecting the proper configuration of a pump station with regard to reliability, choosing good flow operating ranges and optimizing the number of pumps. Its really not that hard to get it right, but so many times you see operating expectations like this one for one pump.

**********************
"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]

wfjer (Mechanical)

Why are you not taking the advice of the pump proposer regarding the minimum continuous flow or are you hoping that someone in this forum will say it's ok to run at a lower flow rate and that will make it all OK.

You have been advised a minimum flowrate for a good reason, if it's a problem take it up with the pump manufacturer.

 

[JLSeagull]

Once the pump is selected it is common to apply override controls to maintain a minimum and maximum flow rate such as minimum flow recycle and clamping the downstream load. Was a variable speed driver considered (I know that it is a big pump)?

 

[wfjer]

Well, this pump is part of a bid, so in this stage is not posible to change the configuration to use two or three pumps, although it's a good question why the designers of the system choose to use only one big pump.

As I said in my second post, I'm not going to aprove the proposal, but instead to ask for a new model/size. I agree that reliability in a pump of such dimensions is extremely important and that's why I posted in this forum, to see your opinions about my concern to take the more appropriate decision.

 

[BigInch]

Part of a BID?

RIGHT NOW THAT'S THE ABSOLUTE BEST TIME TO SOLVE THIS PROBLEM.[.B]

Really it hasn't even begun to be a problem, as problems go.
This problem is only the size of the project manager's ball bearings. Tiny, tiny, tiny.

JLS, Please don't suggest a VFD that isn't needed except only trying to solve a problem that isn't even a problem yet.

**********************
"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/

 

[wfjer]

BigInch, I was misunderstood, it's a bid, but basic engineering has been completed, and our scope its to supply part of the equipment, and part of detailed engineering. So I'm not considering the posibility to change the pump arrangement.

 

[BigInch]

Thought it was a bid made to you quys as owners. Please accept my apologies to you. No apology to their project manager.

**********************
"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/

 

[HEC]

wfjer,
Sounds like a highly paid consultant engineer has provided a lazy solution. If you are responsible for providing the equipment and detail engineering as part of your bid then provide the detail engineering. Give thema price for the system as proposed and provide the better detailed engineering solution, be it with parallel pumps or other solutions. As BigInch states NOW is the time to solve the issues. It may swing the job your way dy showing you have considered the problem and have a solution!

Mark Hutton

 

[BigInch]

Mark, An excellent suggestion. It appears to be the best possible approach for wfjer to take in trying to turn around that impending train wreck.

**********************
"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/