Calculating NPSHa for booster pumps 3

[davincigee]

Hi,
I have been tasked to perform a fluid mechanics analysis of the possibility of transferring gasoil and gasoline from our tank farm to another tank farm hundreds of km away via a booster pump. We are certain that the booster pump can send gasoil or gasoline to our desired destination. However what we are uncertain is whether our pumps (rated head is 62m) can actually provide enough suction pressure (npsha) to the booster pumps. Can anyone please help with an appropriate formula to crack this? I tried the usual formula of NPSHa= Ha+Hs-Hvp-Hf-Hi. (where Ha=atm head, Hs=static head, Hvp=vapor pressure, Hi=suction inlet head (usually with a safety factor of 2ft)) . But I am pretty not sure if it'll work. Please advise me if i am on the right track and if i am not what is the right approach.
Thanks

 

[BigInch]

NPSH_R is an "absolute head".

To compare NPSH_A to NPSH_R, obviously both then need to be in terms of atmospheric pressure.

NPSHA = (atm absolute pressure, OR absolute tank pressure) + elevation head - vapor pressure - friction loss + velocity head

When the fluid is in contact with the atmospheric, or same as being pumped from a tank at atmospheric pressure, then add atm [absolute] pressure.

If the fluid is being pumped from a closed tank or closed pipe, then add the tank's or pipe's absolute pressure, which = tank or pipe press psig + atm pressure.

Where,
Elevation head is + when the pump Centerline is below the fluid level.
and
Velocity head is customarily ignored.

"People will work for you with blood and sweat and tears if they work for what they believe in......" - Simon Sinek

 

[BigInch]

Sorry. To compare NPSH_A to NPSH_R, obviously both then need to be in terms of absolute pressure.



"People will work for you with blood and sweat and tears if they work for what they believe in......" - Simon Sinek

 

[Artisi]

Trust you can stand a candid comment, which I will make anyway -- I think you are going over the top with this NPSHa/r calculation.

If it were me, I would initially draw a system curve for the pump discharge ie flow against head ( head consisting of friction loss for the pipeline from your farm to the booster pumps plus any ancillary valves and a bit of fudge factor -to this I would add the NPSHr of the booster pump/s or the NPSHa of 7M if that is all you have available + any elevation change) ignore all the other add-on's and deductions at this point. Draw this across the pump performance curve, if it intersect the pump curve and a meaningful flow rate then this will indicate that NPSH should be ok - if the head is above the HQ curve it will indicate that you could have insufficient NPSHa at the booster inlet.

From here it becomes more complicated and you now need to look at series pumping and the effect of the pumps working together in series which will entail further calculations etc. However, what is not clear and it is not possible to make any comment is the apparent difference in capacity of "your" pumps and the booster pumps.

Of course I and others have assumed that you are pumping direct into the booster pumps - Correct?

Take this for what it is worth, if you don't agree let me know one way or the other - either way is ok with me.

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.)

 

[davincigee]

Wow.... I never thought of the analysis that way. That was really clever. Thank you Artisi. I'm going to try it and give you an anwer. To answer your questions; the rated capacity of our pumps are 145m3/hr and the rated capacity of the booster pumps are 372m3/hr. ....and yes you assumed right, we are pumping directly to the booster pumps. Thanks a million...!

 

[BigInch]

With the mismatch in flowrates, you will need to have a very detailed look at that. As it is now, with no real information, it is difficult to predict what will happen, other than both pumps will try to pump the same flowrate, no matter what their actual ratings are.

For example, say the pumps ranges are

372 m3/h rating has a range of 200 to 400, and the
145 m3/h rated pump has an actual range of 100 to 200 this system won't work very well, except at the 200 m3/h point, at which you will get the combined head of both pumps when they are flowing 200 m3/h. If that's at the far right end of the curve of the 145m3 pumps, then head from that pump is likely to be relatively low. Maybe even very low, thus they may not contribute to the effort much at all. Possibly nothing, while the 372 m3/h pumps are pumping away at 200 m3/h but at a very high discharge pressure. No load sharing, a complete mismatch in fact. Turn the 145m3/h pumps off and bypass them to achieve nearly the same effect. And the 372 m3/h pumps wind up operating at 200 m3/h, relatively far to the left on the curve, in a possibly very inefficient region of that curve, no matter what you do with the 145 m3/h pumps. That might amount to a very costly pumping exercise.

You must also know if you can control the flowrates, or if you will have to just let this system flat at the intersection of the natural pump and system curve. Probably not a good idea with this kind of a mismatch. Both pumps are likely to operate in very inefficient regions of their respective curves.

"People will work for you with blood and sweat and tears if they work for what they believe in......" - Simon Sinek

 

[Artisi]

BigInch spot-on, correct. Next problem to review and remedy.
TD2K mentioned this way back suggesting 2 or 3 pumps in parallel to feed 1 of the booster pumps.


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.)

 

[BigInch]

Yes, but it seemed like you didn't wanted to ignore that topic back then.

"People will work for you with blood and sweat and tears if they work for what they believe in......" - Simon Sinek

 

[Artisi]

Always on the back-burner, but 1 problem at a time, plus the original question was about NPSH - so left it alone other than to suggest early in the discussion of getting someone to review the overall pumping system - but sigh - to no avail.

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.)

 

[BigInch]

Exactly. It does go that way more often than not. They think they have a problem, only to find that it is nothing next to the in the room.

"People will work for you with blood and sweat and tears if they work for what they believe in......" - Simon Sinek

 

[davincigee]

Come on Artisi! I thought we were over this. Anyway, I took TD2K's advice concerning the . I wish I could give more information especially that pertaining to the booster pumps but the organisation won't provide those info's for some wierd reasons. I hear they have some disputes with my company. I was lucky enough to obtain the NPSHa from test result document during the commisssioning of the booster pumps. Nonetheless, everyone's help has enlightened me more... Hey Artisi, like I admitted, I made an error and I apologise. All the same thanks. I am still trying to obtain the full pump data from the booster pumps so I can use your alternative procedure to obtain the results of my analysis i.e. if I manage to get it. But I will get back to you if I get some results...Thanks guys

 

[Artisi]

The performance of the booster pumps is really required, otherwise you have no way of knowing what flow rate you need to supply to the booster pumps. Series pumping is not a black art but you must have detail of both pump/s sets - the pumps interact with each and cannot be treated as seperate identities.

The info from BigInch re the difference in flow rates is the crux of a successful operation.

A simple discussion is attached to help in the understanding of series pumping.

http://www.doylepump.com.au/graph/calc/pump_school_10.html

A rough and ready graphic of what will / could happen with the mis-match in flows.

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.)