How to use a 1.0 SG curve for fluids of different SG?

[KernOily]

You know how you think you know something, and then you try to explain it to someone else, and then you find out maybe you don't have it completely straight, after all...

A colleague was checking a sizing calculation for a pump in 25% saturated brine service. The curve used for the pump was of course developed for 1.0 SG (water at 60° F). The question I was asked was: how do I use a pump curve developed for a 1.0 SG fluid for a fluid of 1.15 SG (or any other, for that matter)?

I will now shut up and listen, and see if I have it straight or not.


Thanks!
Pete

 

[bingopin]

Hint...That's why the Y-axis is shown in feet of liquid rather than PSI(G).

 

[BigInch]

Pete, No change to the pump Head curve.

dP changes though. P (psig) = dH(from pump curve) * SG * 62.4 pcf/144

So does the Power curve. P = (Power @ SG = 1) * (SG of the brine).



BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[MortenA]

The neat thing about head is that the head in liquid column is allways the same no matter what liquid you are pumping.

This does however mean that the discharge pressure changes since dP and required power is calculated as biginch has written above.

Best regards

Morten

 

[bulkhandling]

That's something I was confused when started working on pumps years ago.

A pump was used to pump a heavy liquid, say SG=1.5. The liquid was joined by a lighter liquid somewhere at the discharge piping, so the liquid SG became 1.1. We found the total head generated by the pump was much bigger than the pump curve operating point...

For a pump at a fixed operating point, pump speed rpm is fixed, tip speed is fixed. Pump develops head (or pressure) by throwing the liquid away from the impeller and generates momentum. For heavier liquid, the liquid velocity leaving the impeller tips is the same as lighter liquid but momentum is bigger. So the head is the same (velocity leaving the impeller) but pressure is different.

 

[MortenA]

Bulkhandling: You write that the observed HEAD was bigger for heavier liquids. This is contraty to theory - but some factors may not be aken accout for such as the motor being able to deliver the duty - this would however result in a lower head i guess. Or did you mean "discharge pressure"?

Best regards

Morten

 

[Artisi]

Here is a link that might be of use to you.

http://www.fluidedesign.com/yahoo/The influence of specific gravity on total head.pdf

Phitsanulok
Thailand

 

[bulkhandling]

Morten,

I did not mean that the head was bigger for heavier liquid. It was heavier liquid pushing lighter liquid that aquired bigger total head than the operating point shown on the curve (heavier fluid in the pump but lighter in the piping). I think this can be easily explained by general pumping principle.

 

[BigInch]

Well, here's the way it works in multiple product pipelines.

LIGHT BEING REPLACED BY HEAVY

Upstream head losses increase significantly, as the heavy oil is introduced into the pipeline at the initially high light oil flowrate.

As more heavy oil is introduced, the pipeline's flowrate drops rapidly. As pipeline flowrate drops, head losses all along the pipeline, but especially in the downstream segments still containing light oil, become less and suction and discharge pressures no longer fall.

With the addition of more heavy oil upstream, the pipeline flowrate continues to fall, but more slowly. The pipeline's flowrate is now so low that head losses, even in the heavy oil segments become less and suction and discharge pressures there begin to rise, but do not change much in the light oil segments, where the light is soon to be replaced with the heavy oil. The decreasing head losses, from the decreasing flow, balance the increasing head losses from the heavy oil now arriving at those segments.

As the last of the light oil is forced out, head loss tends to increase in that segment with the replacement of the light by the heavy oil, but pipeline flowrate reduces even further, almost cancelling that tendency, but not quite, and the as the last of the light oil leaves the pipeline and the heavy finally exits, the net delivery head settles at the steady state delivery head of the heavy oil.

HEAVY BEING REPLACED BY LIGHT

As you replace a heavy oil with a lighter oil, the head losses of the pipeline segments containing the light oil become less and suction pressures correspondingly increase with the lighter fluid upstream's lesser head loss.

As the lighter fluid enters the pumps, the total discharge head increases rapidly with the pump's suction pressure rise, but the pump differential head will remain the same, as long as the viscosities of the light oil and heavy oil are similar.

The continually decreasing head losses in the upstream segments eventually start to bring up the flow along the entire pipeline, but as the flow is brought up, the head losses all along the pipeline increase until a new equilibruim point is reached.

As the heavy oil continues to moves out, flow of both the light and heavy oil in the entire pipeline begin to increase rapidly, as do the head losses due to increasing velocity begin to overpower the decreasing head loss due to the replacement of the heavy denser oil in the downstream segments, as the last of the heavy oil is forced out.

When the light oil finally exits the pipeline, the last inceases in head due to the heavy oil finally leaving the pipeline are slightly larger than the decrease in head due to the increasing velocity and the ultimate equilibrium point is established at the steady state delivery head of the light oil.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[TenPenny]

Many useful comments, but consider: what type of pump is it, centrifugal or PD? What units is the curve in?