Case well water pump efficiency in variable speed applications

[fastline12]

We are looking to build a variable speed, constant pressure water well system for use in a geothermal grid. We would like to use a 1.0-1.5HP system to flow between 10-21GPM. This pump would also provide drinking water.

I have contact OEMs regarding pump efficiency other than 60hz synchronous operation but not much info. I would really like to know if a pump that is optimized for max eff at 33GPM, what would the eff look like at half speed? These would be multi stage cased well 4" pumps. I would like to take advantage of the service factor in a 1HP pump and overspeed the pump slightly so we can get the best possible performance at the mostly required 10-15GPM.

 

[stanier]

Search this website for discussion of VFDs and how energy inefficient they are. Also got to

http://www.pipingdesign.com

for some technical papers and presentation of VFDs.

"Sharing knowledge is the way to immortality"
His Holiness the Dalai Lama.

http://waterhammer.hopout.com.au/

 

[fastline12]

Thank you. I have a decent handle on the efficiency of VFDs, I was more or less concerned about the pump side performance. The pumps in question are usually specifically optimized for 1750 rpm IIRC. The OEMs did not have any charting for performance at alt speeds. We would roughly need to half the typical pump speed.

 

[Pumpsonly]

Unless your operation process variation calls for a VFD ,how much energy can you save with a 1.5HP using VFD? You will probably end up with more headache.

 

[fastline12]

Well, the numbers seem to be leading to near a linear relation of water flow and power requirement and since much of the requirement would be in the 10-12gpm range, a 1/2hp pump should be more than enough. So in a perfect world, we might see a 66% savings. Realistically, I am still not sure but I guess if the pump responds is fairly linear in relation to efficiency, it might be close.

 

[stanier]

Use the Laws of Affinity to estimate the performance of the pump at reduced speed. Refer attached paper. The power is reduced by the speed change cubed. Hydraulic power for fluid is Q*H*rho/6116*60. Q m3/hr, H m, rho kg/m3.

Divide hydraulic power by pump power for efficiency.

Then add all the inefficencies of the VFD, including its embedded energy.

"Sharing knowledge is the way to immortality"
His Holiness the Dalai Lama.

http://waterhammer.hopout.com.au/

 

[Artisi]

I can understand not being able to get much info' on efficiency at reduced speed - pumps of that size are very inefficient anyway and knowing reduced speed inefficiencies is really academic, especially if connected to a VFD. By the time you connect a VFD the power input might well exceed the full load current at its full load speed. Just take a guess for the pump efficiency, if it's 45% at full speed not connected to VFD, at 1/2 speed it might be 25/30% and at the end of the day does it really matter seeing as how you are talking about a few amps here and there.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[BigInch]

Halving the speed cuts flow by 1/2, but head by 75%. Will that work in your geothermal piping system?

You want to vary power from 1 to 1.5 HP, 50%, but flow by 100%
The "linear relation of water flow and power requirement" is more like cubed.

"OEMs did not have any charting for performance at alt speeds", should tell you something. Nobody runs them at alt speeds. They are already inefficient enough.

You might have a handle on something, but its not pump flow. Pumps this small don't use enough energy to save anything worthwhile, nor are they efficient enough to warrant use of a VFD in such a low efficiency system. You're going to try to save 20% of the energy use by adding a VFD consuming 5-10%

You've already spent all the money you might have saved with a VFD, if you would have saved any at all and that's highly unlikely, on design time for this problem. Don't lose any more.




We will design everything from now on using only S.I. units ... except for the pipe diameter. Unk. British engineer