Effect of second pump in the line on NPSHr

[Pipeline1972]

Hello,
I have a water system with a huge NPSHr, compared to the NPSHa. I can't do much about the NPSHa, so i need to somehow reduce my NPSHr. I'm wondering what the effect would be on the NPSHr of adding a second pump in the line, in serie with the first one, somewhere halfway the end.
Any thoughts?

Thanks,
M.

Process - Piping
ing - EiT

 

[GaTechTheron]

We are going to need more information to give you an accurate answer.

If I were you I would add another pump, but make sure you run it at 1750rpm. Keep in mind: In series the flow stays the same, and the TDH stacks.

I would say you are better off replaceing what you have with a 1750rpm multistage pump, or a vertical can pump if you are really hurting for NPSHa.

 

[Artisi]

More info please.
A question for you, how can you have a water system with a huge NPSHr? - NPSHr is a pump requirement not a system requirement, do you mean that the pump NPSHr exceeds the NPSHa, if this is the case then you need to select a pump with a NPSHr lower than the NPSHa.

 

[MortenA]

except for replacing the pump i dont think you can change the NPSHr for a given pump.

Best regards

Morten

 

[BigInch]

Reduce the specific speed & suction specific speed
by reducing flow, or speed or increasing head.

http://www.mcnallyinstitute.com/07-html/7-03.html

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

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

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25% to 50% of the total electrical energy usage in certain industrial facilities." - DOE statistic (Note: Make that 99.99% for pipeline companies)

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

 

[bk19702]

NPSHr is a function of the pump design and little can be done to alter it beyond the design stage. NPSHa is system related and usually there are some measures to take to improve the value. Pumps in series won't solve the issue, like GaTech said. Running pumps in parallel can solve NPSH issues in a particular system since you are utilizing half of the available value, more or less. Setting that configuration up correctly so it operates reliably is a whole new ball of wax and could be a separate thread on it's own.

 

[BigInch]

"Running pumps in parallel can solve NPSH issues in a particular system since you are utilizing half of the available value, more or less."

Its not clear how you would be using half of "the available value". What available value, NPSHa? Would you then be running 2 pumps at half flow?

Then why just not run one pump with a control valve to cut back on flow and increase the NPSHa in the suction line?

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25% to 50% of the total electrical energy usage in certain industrial facilities." - DOE statistic (Note: Make that 99.99% for pipeline companies)

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

 

[GaTechTheron]

"Then why just not run one pump with a control valve to cut back on flow and increase the NPSHa in the suction line?" - BigInch

This sounds possible, but ineffecient, correct? If you were to replace the current pump, you might as well put in something that will do the job.

 

[BigInch]

A little inefficency is better than NPSH cavitation or adding a second pump so they can both run at half speed.

Of course replacing the pump with the proper model is best, but that option wasn't part of the question.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25% to 50% of the total electrical energy usage in certain industrial facilities." - DOE statistic (Note: Make that 99.99% for pipeline companies)

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

 

[Pipeline1972]

thanks for all the replies. I think the formulation of my problem was unclear.

I'm trying to model a system in Pipeflo. Ultra Pure Water system, closed loop. I have a source tank with water level at 10m. System requires a 80m3/h flow. Piping is PVDF, 225 mm. Larger pipes are not ok. Required head is about 15 m.

The program tells me the NSPHa is not sufficient. For a good NSPHa, i need to raise the level in the tank to 35 m. For obvious reasons, the level is limited to 10 m.
Fluid velocity: 1.5 m/s (which is too low actually)

At this point, i don't have a pump model. That's what i'm trying ot find out with this program, but it get stuck because of the NPSHa. Maybe a 2nd pump in serie would help, but Pipeflo doesn't like that option..


Process - Piping
ing - EiT

 

[TenPenny]

I don't use pipeflo, but I'm curious about how it knows you have insufficient NPSH, if you don't have a pump model yet.

You should be able to do that easily with a 125-80-200 pump, needing approx 1.7m NPSH.

Perhaps you have too much suction piping/losses, and you run out of NPSH before the pump....where is your pump located in relation to the supply tank? What temperature is the water?

 

[Pipeline1972]

Thanks Tenpenny.

When i launch a calculation, i get an error message at the pump, that won't go away until i raise the level in the tank to 35m. Tank and pump are close together, so the solution is not there. Normal conditons for water.

How did you come up with 1.7m NPSH?

Process - Piping
ing - EiT

 

[BigInch]

A second pump in series IS NOT the answer. It will do nothing except raise your head delivered to the system. How will that help your suction??????


Reduce the specific speed & suction specific speed
by reducing flow, or speed or increasing head.


Stop using pipeflo for awhile and start reading the references I gave you.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25% to 50% of the total electrical energy usage in certain industrial facilities." - DOE statistic (Note: Make that 99.99% for pipeline companies)

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

 

[GaTechTheron]

I agree with BigInch. Obviously there is something wrong with the software, or how your setting up the problem. It sounds to me like you need to add the pump model first because your software is not taking the pump into account. If you lift the tank 35 meters, then you wouldnt need a pump. You can just Gravity Feed!

Do it the old school way.

 

[TenPenny]

Pipeline:

I ran your 80m3/h at 15 m head through a pump sizing program, it came up with lots of options that require 1 to 2 m NPSH.

There's obviously something a bit wacky with the way the problem is set up, or the way the system is calculating.

 

[BigInch]

Maybe I misunderstood and you want to add a booster pump into the suction line? Is that what you were trying to say?

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25% to 50% of the total electrical energy usage in certain industrial facilities." - DOE statistic (Note: Make that 99.99% for pipeline companies)

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

 

[Artisi]

Go back to basics and calculate the system loss/NPSHa with a sheet of paper, a penci, a set of friction loss tables and NPSH = Ha-Hvpa (+/-) Hst - Hfs - no error messages if you do.

 

[smckennz]

A larger, second pump, running at a slower speed will have a lower NPSHR and should be selected to satisfy your available suction conditions. If you install this upstream of the existing pump, and perhaps adjust the existing pump impeller diameter or speed to preserve the combined TDH , you should be able to achieve your objective.
However it sounds as if your Pipeflo program or the defaults or inputs are bad. Suggest you calculate by hand, with simplifications if necessary, this might help some insight into whats going wrong.

Cheers

Steve

 

[Pipeline1972]

Thanks you all. There's gotta be something wrong, not with the program, i'm not that arrogant lol, but with my input.

I did a test with a very short line, and still had the same issue. So i figure something's wrong in the design settings, material specs, estimated pressure loss in equipment etc...

Biginch, i did check your references. Thanks. And yes, a booster pump is a possibility. I'm not familiar with these, tho, so i'll need to investigate that as well.

I realise this post is probably not in the right forum... sorry about that..

Process - Piping
ing - EiT

 

[Pipeline1972]

yup, there it is: i picked the wrong fluid for my system: vinyl chloride instead of water!! It was in front of me the whole time!


Process - Piping
ing - EiT