Pumps in Series - Total Dynamic Head Question

[HouTxMechPE]

Hello everyone. New to the forum and after having searched for the subject I haven't really found what I need. The situation is a gasoline transfer project from a large storage tank to an intermediate pump to a large transfer pump that is already in place. I need to determine the discharge head for the intermediate pump so it can be sized. I know my total headlosses for all pipe/fittings involved and the elevation drop. Its all downhill so that is what's throwing off my numbers, making the pump's HP go down.

I am confused on whether or not I add the NPSHr of the 2nd pump to my total head before figuring out what size the interm. pump needs to be. The pumps are vertical turbine pumps that sit in a can or well. The nozzle is the focal point for elevations and the pump mfr. is told that there is zero NPSHa at that point. From there they spec how deep the can needs to be and how long the pump is etc.

Head loss in all the piping is 54' from the sump of the tank to the nozzle of the 2nd pump. Elevation drop is 17'. Desired flow rate is 38,500 GPM and Spec. Grav. is 0.68. Using that head loss I come up with 420HP. Reducing that by the elevation drop yields 290HP. NPSHa calcs give me -25' at the 2nd pump, which is 200HP. Any pointers are much appreciated!

 

[RWF7437]

Why are you pumping DOWNHILL?

 

[HouTxMechPE]

The suction headers are in the ground and the piping runs pretty much level but there are a few places that jump up 20' and go over a road/piping rack and then back down again. Also, gravity will not move any fluid that fast. From point A (Tank Sump) to Point B (Pump Header) its a elevation drop. Correct me if I'm wrong.

 

[RWF7437]

Your original post said it was "all downhill". If that is so, then no pump is needed. You then only need to provide a very large pipe for this high flow rate ( 38,500 gpm?)

If it is not "all downhill" then you MAY need a pump, or at least an air/ vacuum valve at the high point. To make your question clearer try posting a sketch or scale drawing of the actual conditions.

good luck

 

[MortenA]

NPSHr is just a treshold number - you dont add it to the head - since its not subtracted.

You calculate you pressure at the pump inlet (NPSHa) and the discharge pressure is the NPSHa plus dH (use density to transfer to your preferred pressure unit).

If you a looking for max discharge pressure use max NPSHa and shut-in pressure for the pump.

Best regards

Morten

 

[HouTxMechPE]

Sorry for any confusion. The piping is sloped downhill yes, but there are several areas where the pipe has a "U" in it to go up 20'over a road to allow for trucks to clear and they need to move the gas at a very high rate. Its 36" pipe.

MortenA: Thank you for your reply. dH is the major/minor head loss? Or density head? Vapor Pressure of fluid converted to head. For the NPSHa I am already accounting for that and come up with a negative number.

Most curves have the ShutOff Head listed, is the ShutIn the starting point of the curve?

 

[nationalpump]

What is the output (in feet of head) of the first pump? Are you using this value in your calculations? Are you also using absolute barometric pressure to calculate NPSHa?

Adam

 

[MortenA]

dH=delta head - that is the amount of head added by the pump.

Multiply with the density and gravity acceleration constant and you got delta pressure (make unit match your self ).

Again you use vapour pressure for calculation the NPSHa - but dont subtract it from the total head - its just a value for checking.

NPSHa can be negative - when you consider 0 psig as your reference pressure. In absolute (more commonly used for NPSH) pressure can of course never be negative.

Most smaller normal speed pumps will have a NPSH below 0 psig.

Bestregards

Morten

 

[Artisi]

For simple consideration without getting too academic for series pumping is:-

The inlet pressure at the 2nd pump is the discharge head(pressure) of the first pump less the losses between discharge of the first unit and inlet flange of the second pump. Normally you would select / calculate / position the second pump so as to have a positive pressure at the inlet to the second unit - therefore you can assume that the NPSHa is greater than NPSHr and doesn't require any further consideraton. However, remember that any positive pressure at the second inlet is added to the capabilty of the second pump.
Don't forget to convert all pressures for density etc. as already pointed out by other postings.

 

[HouTxMechPE]

Thank you all for the help. I am using the basic formula dP=SW[dZ+Hl) since the velocity stays the same. One question I have tho is on the lower flow runs I am coming up with a negative pressure drop. I.E. the elevation change is greater than the Headlosses. Also for the total head required of the pump I assume I should be adding the discharge headloss to the suction loss (tank sump to header) in order to size my pump.