8
\$\begingroup\$

I want to power a three-phase 5hp well pump with a VFD. I want the VFD both to serve as a soft start, and to allow variable speed operation of the motor.

Two different pump salesmen have told me that running a VFD off a generator might hurt the VFD. Is this so? And if so, what is the mechanism by which generator power might hurt a VFD, but grid power does not?

EDIT: If the problem is variation in the generator's frequency or voltage, to what extent can that be mitigated by using an inverter-generator?

\$\endgroup\$
5
  • \$\begingroup\$ I very much doubt you will hurt the VFD. But I also will not be reimbursing you if you do. LOL. My pool pump runs on VFD (only 1HP). It runs find on our standby generator, but that is an 8kW standby generator running our whole house. So the 1000W going into the VFD is only a small part of the load. I have used small VFD's for dynomometers also. They seem to be pretty robust. (I am talking 1-5 HP only). \$\endgroup\$
    – mkeith
    Mar 24 at 5:21
  • \$\begingroup\$ I have a 5HP single phase well pump. I think I will probably replace it with a 3HP 3-phase pump and run it from a VFD if the pump ever breaks. I think 3HP is a lot better than 5HP if you can manage it just because there are more VFD's available with single-phase input at 3HP than at 5HP. Also, for a given size generator, I am sure the smaller VFD will be easier to drive and more likely to be successful. \$\endgroup\$
    – mkeith
    Mar 24 at 5:29
  • \$\begingroup\$ Unfortunately I need 5HP for the quantity of water I'm pumping. Since I posted this, I've purchased a 10HP Toshiba VFD which output more than 5HP if fed with single phase power. Is it hopeless to run this thing off a 6500W Honda generator? \$\endgroup\$ Mar 25 at 16:42
  • \$\begingroup\$ I don't think it will run at full speed. Not enough power. But I would try it and see what happens. I personally don't think you will break anything. Just be ready hit the stop button on the VFD if anything should go wrong. If it is possible, I would try to get the VFD running with utility power first, just to remove one variable. Once it is running the way you want, then try the generator. My 5 HP single phase well pump uses 28 Amps at 240V once it is up and running. I don't know what the power factor is. But that would be too much for a 6500 W generator, even ignoring startup surge. \$\endgroup\$
    – mkeith
    Mar 25 at 16:54
  • \$\begingroup\$ With the VFD, there is no startup surge. Assuming you program it correctly it can spin up slowly. So you just need to limit the top speed of the motor to something that the generator can supply without being overloaded. Then verify that the water flow rate is acceptable. If the flow rate requirement can only be met by running full speed, then you will need a bigger generator for sure, just from a power perspective, and ignoring stability, etc. \$\endgroup\$
    – mkeith
    Mar 25 at 16:56
7
\$\begingroup\$

Two different pump salesmen have told me that running a VFD off a generator might hurt the VFD.

It's not that simple. But I can understand why the saleman says it. It often causes problems because the VFD is too big for the generator or is misconfigured, or the only load on the generator.
This can cause severe harmonics, or instability (governor and voltage regulator) if the VFD is not running simple UF mode.

Sometimes the VFD caps blow up. Sometimes it errors mains overvoltage/undervoltage. Sometimes other components blow up. Potential for damage is high.

So, when you do this, make sure you pay attention to:

  • Sizing. Try to get the generator 2 times larger that the total VFD load. This is a simple rule of thumb, more specific calculations are possible taking the subtransient reactance into account. Talk to your generator supplier. Ask them for numbers.
  • Drive switching frequency. As high as possible considering drive derating.
  • Drive control algorithm. Simple V/F mode, no vector control algoritms.
  • Drive ramp times. No sudden speed changes, the engine is slow and can't give you instant kilowatts.
  • Use a braking resistor for braking, not 4-quadrant operation.

There is no cost effective filtering available to correct for a harmonics problem here. That's why the salesman just says you can't, because it could become an expensive job. Lot's of pointing fingers. I've seen it many times.
The frequencies you want to filter are 3rd to 20th harmonics, where normal mains filters work beyond the 200th harmonic.


I wouldn't even consider using a VFD on an inverter generator. Those are so small, recipe for disaster.

\$\endgroup\$
3
  • \$\begingroup\$ What does UF mode mean? Is that the same thing as what I would call "V/f"? \$\endgroup\$ Mar 11 at 7:14
  • \$\begingroup\$ @AndrewCone Yes, that's the same thing. \$\endgroup\$
    – Jeroen3
    Mar 11 at 7:29
  • \$\begingroup\$ Is a "mains filter" the same thing as a "line reactor"? \$\endgroup\$ Mar 13 at 7:15
3
\$\begingroup\$

This really depends on the VFD and generator combination. If the VFD can handle variable input voltage and frequency then I would say it can run on a generator. If the input voltage requirements are tight, the maybe it can't. The problem is that, in general, generators are not very stable in either voltage or frequency. This is of course dependent on the generator and load. Under varying load conditions the generator may under or over-voltage your VFD input and it may produce 70 Hz for short periods or 40 Hz depending on how the load varies. A more specific answer could be given with information about the specific equipment you are working with.

\$\endgroup\$
3
  • 1
    \$\begingroup\$ The two properties the generator needs specifically are good speed regulation to control output frequency and regulated field coils to control output voltage. If the VFD is built in such a way that it pulls non-line frequency current surges off the supply line that could cause problems with smaller generators as well. \$\endgroup\$
    – K H
    Mar 11 at 6:34
  • \$\begingroup\$ I updated the question to ask if it helps to use an inverter-generator \$\endgroup\$ Mar 11 at 6:40
  • 1
    \$\begingroup\$ @AndrewCone I don't think it would help to add yet another frequency sensitive feedback control system to the mix. \$\endgroup\$ Mar 11 at 8:09
3
\$\begingroup\$

The problem is the harmonics produced by the VFD itself. And the problem will be a mutual adverse to both VFD and generator operations.

In fact, VFD works on chopping the waveform which produces harmonics and these harmonics affect the generator waveform which is not the case for the grid supposed to have infinite inertia.

Harmonics may also affect sensitive electronic devices used also to control the generator like the AVR regulating its voltage for instance which may create instability in the system operation.

\$\endgroup\$
1
  • \$\begingroup\$ To what extent does it help to use a larger generator? I'm thinking the bigger the windings in the generator, and the more heavy-duty the AVR, the more it should be able to absorb the harmonics. Is this so? \$\endgroup\$ Mar 11 at 6:52
1
\$\begingroup\$

The problems with AVRs and non-linear loads was very real, 25 years ago. But MOST generator mfrs have addressed and fixed those issue, usually with filtering on the voltage sensing for the AVR, so this has not been a problem for decades. Yet, the perception remains in the industry, because a LOT of people got burned by these issues in the 80s and 90s and generator salesmen do not want to get burned again, so they just warn everyone.. There are however two caveats to that:

  1. The more recent advent of "inverter" generators may have screwed that up again, because if you use an inverter generator, you are feeding an converter with an inverter, and the potential problems are as yet unaddressed as far as I know, so yes, heed the advice to not try that.
  2. New Active Front End (AFE) VFDs are being released for "Low Harmonics" and some designs have issues with the higher impedance of a generator compared to a utility source.
\$\endgroup\$
1
  • \$\begingroup\$ My two options for feeding the VFD are a 6,500 watt rated Honda generator with no inverter, and an equivalent Honda generator with an inverter rated for 7,000 watts. I'm guessing it would be better to use the 6,500, yes? \$\endgroup\$ Mar 25 at 16:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.