Converting a pump curve from water to oil

[Dieter21]

Hello,

My question is if it is possible to convert a pump curve from water to oil. I am currently running a pump where the supplier provided me with a curve that was tested with water. The problem is that I need to pump oil. The oil has a density of 800kg/m3 and a viscosity of 3.94mPas.

Are there any equations that could help me?

Thanks

 

[JJPellin]

Curves for centrifugal pumps are plotted with units of feet or meters on the vertical axis on order to be universal for all fluids. No conversion is needed. You should review a text book on the subject of pump curves in order to fully understand the use of the curves. The only curve provided that should need adjustment is the horsepower curve. The power that your pump will require will be proportional to the specific gravity of the fluid pumped. Most pumps are tested using water. But, at the time of the test, if the manufacturer knows the fluid that the pump will eventually be pumping, they will do the conversion for you and have the horsepower curve plotted for the process fluid that the purchaser specified. Examine the curves closely to see if there is a specific gravity noted for the power curve. If it is plotted for water, then convert the power to your fluid by multiplying the value from the curve by the ratio of the specific gravities of your fluid over the specific gravity of the test water.

If your fluid is extremely viscous, additional corrections could be required for the performance curve. And, if a very high degree of accuracy was required, corrections could be made for the Net Positive Suction Head Required curve. But, for a typical service, these curves can be used as plotted with no correction necessary.


Johnny Pellin

 

[ione]

As clearly pointed out by the good dissertation of JJPellin, viscosity is the parameter which affects the pump’s performance curve when dealing with fluids other than water. There are specific correction factors which take into account the effect of fluid viscosity on performance curves. These coefficients are based on tests carried out by the Hydraulic Institute (

http://www.pumps.org).

Useful stuff at the links below:

http://www.pumpfundamentals.com/help10.html

http://www.gouldspumps.com/pol_0006.html

Density just affects the power of the pump throughout the following formula:

P = (SG * Q * H) / (3960 * ?)

P = power [hp]
SG = fluid specific gravity
Q = flow [gpm]
H = head [ft]
? = pump efficiency

Note that viscosity also affects power but very slightly (you need very thick fluids to appreciate sensible variations).
For thick fluids PD pumps are the best choice, anyway the fluid you’ve described is not particularly thick (cinematic viscosity approx 5 cSt).

 

[chemjock]

Your discharge pressure is going down. Take a look at the downstream hydraulics to see where your new operating point is.

 

[Dieter21]

Thanks for all the answers. I have a pump curve with Flowrate (liters/min) on the x-axis and Pressure (bar) on the y-axis.

I did not have a chance to have a look at the links, but will do so tomorrow.

 

[JohnGP]

Well, if your curve is plotted with Pressure in Barg and not Metres Head, then you can replot with pressures reduced to 80% of the values shown now for water at each value of flow - should give you a reasonable idea of performance on your oil.

But this means that you have to work everything in terms of gauge pressures. It may be easier to convert the units of Pressure on the curve back to metres head, i.e multiply your Barg units by 10 is close enough, then as JJPellin points out, the curve applies to oil as well.

Cheers,
John

 

[Dieter21]

Thanks to all of you!

Regards

 

[BigInch]

A pump with a "pressure" curve. OMG.

"We have a leadership style that is too directive and doesn't listen sufficiently well. The top of the organisation doesn't listen sufficiently to what the bottom is saying." Tony Hayward CEO BP
"Being GREEN isn't easy." Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://virtualpipeline.spaces.liv