Temperature affects on pump performance

[dooka]

Guys,

I am chasing a theory on a dead head pump test. I have a multi stage centrifugal pump that's manufactured deadhead pressure @60hz is 808 PSI (20 degrees Celcius). I am currently testing this pump in a closed loop which runs through a choke setup which is causing the liquid in the tank to heat up. I tested the pump deadhead pressure at different liquid temperatures and got these results:

Degrees (C)- Deadhead pressure
42 - 801.9
37.5 - 803.6
36.2 - 805.2

The trend is obvious but i just wanted to verify if this is a fluid loss or a mechanical loss? Is there a direct relation between the temperature and the pump output?

I am thinking it is to do with the total NPSH of the system and as i increase the temperature i am reducing the vapour pressure causing a drop in head pressure.

Please any advise would be greatly appreciated.

Dooka

 

[Compositepro]

pump pressure increases with liquid density.

 

[dooka]

Based on?

I am no expert but as far as i am aware pump performance is not affected by density. An increase in viscosity can reduce your output but not density.

 

[georgeverghese]

For a minor change in dead head pressure such as this, tank temp should stay constant, provided ambient air temp and wind speed ( which dictate tank heat loss) and tank level also are constant.

 

[dooka]

4 PSI change in pressure over a 10 min test to me is not a minor change in head pressure.I adjusted the temperature after seeing irregular data values at different temperatures. I am currently running a 60 day test to evaluate wear on the pump at a specific solid water mixture. I am seeing gradual reduction in my deadhead pressure over my test period but am getting head pressure spikes when the temperature is cooler. Why? I want to know whether there is a relationship between temperature and pump performance.

 

[itdepends]

Pump performance is measured in metres of liquid- pressure is convenient- but as noted above- if the density changes the pressure will be different for a given pump.
The mixture density may be increasing as you go down in temperature (the water will certainly increase in temperature)- so the measured dead head pressure should rise.

What is the specific solid/water mixture? Are the solids water soluble and if so- how does that solubility vary with temperature.
You need accurate density measurements of the fluid you are pumping to calculate the pump dead head in metres of fluid.
I assume you're measuring the differential pressure across the pump (i.e. subtracting the suction pressure) for all tests?

As a chem eng/metallurgist the first part of any answer I give starts with "It Depends"

 

[georgeverghese]

Okay, didnt realise you could vary tank temps here. As suggested, for a centrif. pump, discharge press would vary with pumping temp since Pd is directly influenced by liquid density.

Still, for an electronic pressure transmitter with an operating span of say 1000psig, a 3-4psi change in Pd is 0.4 % of span - still within accuracy limits for this PT.

 

[LittleInch]

dooka,

Allow me to summarise what the other posters are saying.

1) centrifugal pumps with a fixed inlet head running at a fixed speed with a fixed flow rate produces at its outlet fluid with a fixed HEAD.

2) PRESSURE from that fixed head varies with DENSITY.

3) DENSITY of water changes with temperature. This change is widely available on the interweb.

4) Therefore with everything else remaining the same changes in TEMPERATURE affect DENSITY which affects PRESSURE. Higher temperatures = less density = less pressure.

Has that helped?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[dooka]

Completely understood.

So there is no performance loss, just a change in the conversion between meters of Head to Pressure.

Thank you all for your help.