Understand pump curves

Aug 2019
7
0
Hi,

I am in a situation where I need to estimate the flow rate of air [CFM] through a compressed air fixture. These fixtures have pump curves using water, but to my knowledge this will not help me estimate the CFM capabilities for air instead of water.
Which brings me to the question, how can a water pump curve be useful for different fluids? If I am wrong, how can this curve be used to understand flow behavior of a different fluid? Could this information be converted for fluids of different extremes such as air and water?

I understand how a pump curve can be useful if the pump curve is made using the same fluid, but I'm confused how it can be used elsewhere.

Any information would be helpful. A book or write-up on the topic would be helpful as well.

Experiment information on the water pump curves: The fixture is mounted inline with a system. Water is pumped through the system and the pressure drop between the inlet and outlet is plotted against flow rate.
 
Jun 2016
1,198
565
England
My first thought is that you need very different pumps for water and air.
you may be able to pump air with a water pump but the efficiency is likely to be woeful.
 
Aug 2019
7
0
Yes I understand that. In my situation I'm wanting to understand the head loss in connectors, fixtures, regulators, etc. and to do so I'll be plotting head loss against flow rate.

The means of flow transport will vary, but that's not what I am interested in.

My questions is if I have a head loss chart for say water, can that information be converted and or useful for different fluids such as oil or anti-freeze.
For some of the connectors that I will be running test on they have the ability to work with both air and water.

Thanks for the response!
 
Jun 2016
1,198
565
England
It Depends...

I would suggest that it depends on how similar the fluids are to water.
I would guess that the head loss will be related to the density and viscosity of the fluid.
It is possible that you could find a connection with Reynolds Number.

However I must stress that I am extrapolating from other knowledge about fluid flows,
rather than direct knowledge of the subject of your original post.
 
Oct 2017
578
297
Glasgow
My questions is if I have a head loss chart for say water, can that information be converted and or useful for different fluids such as oil or anti-freeze.
For some of the connectors that I will be running test on they have the ability to work with both air and water.

Thanks for the response!
Yes, but I don't know how much work would be required to do perform the conversion. The level of detail you'll want to go to might also depend on the particular situation you are trying to engineer. It might be better to just start from scratch with a new curve.

Different fluids have different functions for viscosity, heat capacity and density with temperature, so the head loss curves will look very different for different working fluids. The main differences manifest as different head loss gradients in pipes due to friction.

You could therefore make some progress by recalculating the friction head losses in pipes using the Darcy-Weisbach or Hazen-Williams equations and substituting the variables relevant for your new working fluid.

You'll also want to re-evaluate the Reynolds numbers for flow in your channels when computing your hydraulic problem to ensure that you're using the correct friction factors for laminar/turbulent flow.

If you care about heat transfer in the fluid, you'll also need to recalculate how the supplies/loads are performed because the change in heat capacity of the working fluid affects those devices. You'll also need to watch out for boiling because some working fluids have lower boiling points than water (i.e. the boiling-limit minimum head curve will change with the different fluid).

The presence of polymers or other media in the fluid can also change the friction behaviour ("Tom's effect"), but for most engineering situations this is not recommended, so you probably don't need to worry about that.