Welcome! Here are the website rules, as well as some tips for using this forum.

# Looking for document or Lit about overspeeding pumps

Options
Member Posts: 205
Anyone ever seen an journal, article, document or other speaking to any pro's and con's of overspeeding pumps past base speed.
Basically anyone reputable have documentation about using VFD to spin 60hz pump at say 72 or 84hz, affects on torque - pump curve - gains or losses.

anyone have some insight from past experience?

• Member Posts: 23,322
edited March 2019
Options
Flow changes linearly with pump speed (rpm). Head changes with the square of pump speed. Therefore -- no surprise -- power required changes with the cube of the pump speed. The whole pump curve will shift accordingly.

If we take a pump operating on 60 hertz, two pole motor -- so a nominal rpm of 3450 or thereabouts, at 72 hertz it will turn at a bit over 4,100 rpm. Flow will increase by a factor of 1.2, head by a factor of 1.44, and power required by a factor of 1.73/ At 84 hertz, over the base, the factors are 1.4, 1.96 and 2.74. Power -- current draw or heat dissipation in the motor -- is most likely to be the limiting factor.

The relevant equations are the pump affinity laws for centrifugal pumps.

Can't speak to the pros or cons -- except for a general aversion to operating devices beyond where they are intended to operate (whether it's a pump or a Chevy V8!).
Br. Jamie, osb
Building superintendent/caretaker, 7200 sq. ft. historic house museum with dependencies in New England
• Member Posts: 27
Options
No experience but here are the basics of running an A/C motor at a higher frequency.

If you run at a higher frequency than design RPMs increase (20% in the case of 72 hz for a 60 hz; 40% in the case of 84 hz for a 60 hz) while torque remains essentially unchanged. I'm not sure what happens with power delivered but logic tells me that it should increase. Besides the increase in RPM there is an increased current draw--typically about the same as the speed increase but in the case of a centrifugal load (like most? every? hydronic circulator) the power requirement increase is closer to the square of the speed increase! While I suspect that flow rate (like delivered power) should increase I have no idea how much. I do know however that any well-designed centrifugal impeller/fan/etc. is carefully matched to the speed and power of the motor driving it lest it either destroy the motor or waste a great deal of energy.

If that potential of doubling the current requirement to maintain the new speed with just the 12hz increase doesn't make you say, "No way", then hook one up and check the amperage while running and circulating. If it's above the full-load amp (FLA) rating you can be certain it will quickly die. If for some reason the windings are so oversized that you're below the FLA rating I'll say that the circulator will die far sooner than it would when operated at the proper line frequency.

Unless there is some new sort of variable frequency drive (VFD) available such are designed to decrease motor speed, not increase it. They do this by decreasing frequency and voltage.
• Member Posts: 7,569
Options

Flow changes linearly with pump speed (rpm). Head changes with the square of pump speed. Therefore -- no surprise -- power required changes with the cube of the pump speed. The whole pump curve will shift accordingly.

If we take a pump operating on 60 hertz, two pole motor -- so a nominal rpm of 3450 or thereabouts, at 72 hertz it will turn at a bit over 4,100 rpm. Flow will increase by a factor of 1.2, head by a factor of 1.44, and power required by a factor of 1.73/ At 84 hertz, over the base, the factors are 1.4, 1.96 and 2.74. Power -- current draw or heat dissipation in the motor -- is most likely to be the limiting factor.

The relevant equations are the pump affinity laws for centrifugal pumps.

Can't speak to the pros or cons -- except for a general aversion to operating devices beyond where they are intended to operate (whether it's a pump or a Chevy V8!).

That's what I was gonna say
"If you can't explain it simply, you don't understand it well enough"
Albert Einstein
• Member Posts: 7,569
edited March 2019
Options
I would say that heat dissipation will be the limiting factor. When sizing big motors at high altitudes, for ski lifts and snowmaking pumps, they intentionally oversize the motors and VFD's to compensate for the lower air density and associated reduction in cooling ability of the air. They then run them at lower frequencies to get only the power they need.
"If you can't explain it simply, you don't understand it well enough"
Albert Einstein
• Member Posts: 3,627
Options
The last VFD I just ordered (To fix a mis-quote by a vendor; thanks, \$vendor!) has a param to set the base frequency. If I understand it correctly, it's the frequency that the drive calls 100%, it's supposed to be set per the motor nameplate but the range is 0-400 Hz. There's nothing to prohibit one from setting that param to 72 or 84 Hz.

I am mildly interested in watching what happens to a device run hotter than it's design; but I'd want it instrumented so as to observe task performance & not just motor performance.

• Member Posts: 23,322
Options
ratio said:

The last VFD I just ordered (To fix a mis-quote by a vendor; thanks, \$vendor!) has a param to set the base frequency. If I understand it correctly, it's the frequency that the drive calls 100%, it's supposed to be set per the motor nameplate but the range is 0-400 Hz. There's nothing to prohibit one from setting that param to 72 or 84 Hz.

I am mildly interested in watching what happens to a device run hotter than it's design; but I'd want it instrumented so as to observe task performance & not just motor performance.

Fascinating. A mil-spec control, perhaps? 400 hz is the operating frequency of most aircraft AC power (they're nice and convert it to 60 hz for the folks in the seats in back).
Br. Jamie, osb
Building superintendent/caretaker, 7200 sq. ft. historic house museum with dependencies in New England
• Member Posts: 3,627
Options
@Jamie Hall, nah, cheapest thing I could find to run a 3 phase exhaust fan (power exhaust on a packaged unit, FWIW) on single phase. \$Vendor didn't check to see if single phase exhaust fans were available, even though we switched to them because the basis of design didn't offer single phase exhaust! To top it off, the durn PM sent my back-of-the-napkin cost off to said vendor. Apparently, my way sounds like the easiest fix to them. Go figure.

I noted the 400 Hz, too. I think it's just a quality part, but I suppose it could have a mil spec version. The cost is right in line with all the other small commodity VFDs I looked at. The mfgr's Fujitsu IIRC.

• Member Posts: 2,646
Options
I believe torque falls off sharply when you are over 60hz. How much I dont know. Saturation of the rotor laminations determines the torque of a given motor design.

This is why it is never a good practice to operate 50hz motors on our 60hz grid, but they can run our 60hz motors at 50hz, just derated. Motor school was a long time ago....

As standard practice we dont operate a motor over 65hz for any prolonged time.
Serving Northern Maine HVAC & Controls. I burn wood, it smells good!
• Member Posts: 1,194
Options
Still need ot extrapolate the pump curve. Could run out of HP on the motor itself. Also need to look at suction piping and overall pipe size.
• Member Posts: 23,322
Options
mikeg2015 said:

Still need ot extrapolate the pump curve. Could run out of HP on the motor itself. Also need to look at suction piping and overall pipe size.

The whole pump curve shifts... what you will run out of, most likely, is cooling in the motor. If you feed a motor enough current, you can get the power out of it -- but it will overheat.
Br. Jamie, osb
Building superintendent/caretaker, 7200 sq. ft. historic house museum with dependencies in New England
• Member Posts: 2,646
edited March 2019
Options
Yes, we sparkies say that motors are dumb! They are so dumb that they will destroy themselves trying to do the work you ask of them. Give a 5HP motor a 10HP load and it will do it! At around twice the nameplate current mind you. This will result in a severe overheat and BOOM.

This is why we have to size overloads correctly.... 125% is rule of thumb. Gotta protect those motors from overworking themselves to death.
Serving Northern Maine HVAC & Controls. I burn wood, it smells good!
• Member Posts: 205
Options
Thank you all,
I agree the affinity/fan laws apply, never could figure out how to shift the pump curve.
Occasionally pump is undersized on some jobs, yes I say match the impellar/pump to the load, just curious if going up to 64,72, 84hz derates the life span of the equipment.
If at 60hz base speed the motor still has amps/HP left, then it sounds like it can still be increased in speed providing the motor manufacture allows for increased RPM of the motor that wouldnt determent motor/bearing life.
I like JamieHall's numbers here, "Flow will increase by a factor of 1.2, head by a factor of 1.44, and power required by a factor of 1.73/ At 84 hertz, over the base, the factors are 1.4, 1.96 and 2.74", looks like I need to reread the affinity laws and pop some numbers out on my own.

Thanks again for all the feedback!
• Member Posts: 903
edited March 2019
Options
Life ..... for 60 --> 80 hz .....I'ld estimate no noticeable effect....... Theoretically I'ld say life is ~ proportional to total number of revolutions over it's life. Basically wear of water seal and bushings. Also old age drying up of rubber that holds carbon/ceramic seals

Wear in general is also somewhat proportional to temp ( though not linearly) , but I suspect increase in heat dissipated in seal is negligible since in my book going from 60 hz to 80 is minor, besides it's "water cooled"

Maybe increased heat in motor windings, but my experience with MY old home circulator ( B&G 100) at 60 hz , motors outlast water seals by 2-3 X life .

• Member Posts: 23,322
Options
So long as the current in the motor windings isn't sufficient to overheat them, I'd pretty much agree with you, @Leonard . The joker in the deck, however, is the potential to overheat the windings, if the power drawn by the pump is enough. Motors are stupid. They don't have any defence against excess current The resulting degradation may be relatively slow -- and it's possible that something else will fail first. However, it can also be quick enough to be legitimately catastrophic -- failure occurring with minutes or even seconds of an overload.

This can be prevented, however, by a very simple measure: find the motor manufacturer's sort term and long term maximum current ratings (this may take some digging) and tailor your fuse or circuit breaker to be just below that curve.
Br. Jamie, osb
Building superintendent/caretaker, 7200 sq. ft. historic house museum with dependencies in New England
• Member Posts: 903
edited March 2019
Options
Yes always have to check overheating, also becomes an issue at low speeds ( unless has separate cooling fan motor ). I read a pretty good artical on VFD for motors, and selecting the motor, was an engineering magazine. I'll see if I can find it.
• Member Posts: 903
edited March 2019
Options
Found it.

This one talks about gear motors but the VFD basics are same. Useful as a primer on what motor issues you 'll come up against when changing motor speed with a variable frequency drive. (Assumes you'll be selecting the motor.)

https://www.machinedesign.com/motion-control/how-maximize-gearmotor-speed-range-0