Variable Frequency Inverter + single phase motor

[Haenk]

I'm going for a single phase motor (with a 16uF run cap, it seems), about 900W/230V.
But I need it to run only at 50% or even better 30% speed.
I know that a different motor or gear is the far better way to go, but this is a complete machine, which cannot be modified (direct drive and IP rated).
So any opinions on this chinese inverter?
https://www.ebay.de/itm/314802749442

Or is it just a matter of "might work, might not work, need to test"?

[CaptDon]

As long as the inverter is only powering the motor and nothing else. Don't know if the motor will be happy or not? It is a split phase motor with capacitor run and may not like the frequency shift. I suspect the speed will vary largely with even small shifts in the load placed on the motor. At reduced voltage/frequency/PWM the motor may not even start under load or when it reaches 'break-away torque' it will then spin up to nearly normal speed. A lot of 'try it and see' to think about.

[tridac]

There are single phase output inverters around, but as a previous post mentioned, the phase shifting motor start cap may cause problems at other than rated frequency. If the motor has a centrifigal switch on the start winding, there's not much that can be done, other than replacing the motor with a 3 phase type. However,  if the motor is cap start and run, with no switch, it's actually a 2 phase motor, with the cap providing a 90 degree phase shift.  If so, discard the cap and find a 2 phase output inverter, to get a variable speed drive...

[Haenk]

That would only be good for a 120 degree phase shift, which might not work well either? I don't think one could set the phase shift manually on those cheap inverters.
I guess I'll just try it (as expected) - so thanks all

[ace1903]

Friend of mine asked me the same question year ago.  I was sure that it will not work and motor will overheat.
He tried with small ABB inverter drive and it still works. Load is fan wheel, so almost no load at start up and progressive load as speed increases.
He connected the motor between two outputs of the inverter. No changes were made in wiring of the capacitor.
I was surprised that drive did not notice phase current imbalance and did not report any warning or error.
So my advice is if load fan like you can try.

[PSR B1257]

but this is a complete machine,

What type?

At 30% speed you only get 30% power from the motor...

I guess I'll just try it (as expected)

Not worth it.
The Motor might do something, but not necessary what he should (properly). At lower frequency the impedance of the capacitor increases thus the current in the auxiliary winding decreases thus even less torque.

[tridac]

Looking again at the original post- is that really is a 900 watts motor. That's around 1 hp and doubt it that would be a cap *run* motor. Usually cap start and centrifugal switch for that sort of power.

What is the motor driving ?. If the thing is belt drive, why not just fit a smaller pulley to the motor ?. Usually use taper lock bush pulleys here for tasks ike that. Not the cheapest, but no fiddling with keys and keyways. Fit once and forget...

[Siwastaja]

Single-phase run-cap motor is actually a two-phase machine - I mean actual two-phase which implies 90-degree phase shift (sine and cosine, which define a rotating vector). I know there are specialized inverters just for this purpose but can't remember brands to suggest. I have built one myself in past and it did work as intended (was one-off for own use). The second "phase" (i.e. run cap winding) is probably supposed to run at lower voltage (current) which needs to be taken into account in inverter design.

The problem with 2-phase system is that it requires a third neutral wire which carries the sum of phase currents, and from inverter design it requires two full bridges i.e. 4 half-bridges while 3-phase inverter works with one half bridge less; or alternatively, a center-tapped power supply which I opted for in my DIY project.

I don't think a 3-phase inverter could be abused for this but maybe I'm wrong.

[johansen]

I ran a 120v single phase capacitor run 1 ton r22 compressor off of a 3 phase vfd, programmed for 110v 60hz.

You need a 120v to 60 volt transformer, and an additional transformer as needed, to subtract some voltage from thr "start" winding, so you dont burn it up.

In my case, thr capacitor "start" winding, ran good on 130vac and the run winding on 110vac, so i used a 120v to 32v transformer to add voltage to the 3rd phase.

I was using a motor with only 3 terminals, start and run share a common point.

Since you can most likely isolate the two coils, you only need a 120:120 isolation transformer, connected as an auto transformer. And an additional transformer as needed so you get good torque without burning up the motor.

If your motor is a 120/240 motor, you can use thr motor itself as the auto transformer, connect the start winding to the midpoint of the run winding and the other end to the 3rd phase of the vfd. An auto transformer may be needed to buck the 3rd phase down

Anyhow, regardless if you use an auto transformer or the motor as an auto transformer, you get a true 2 phase output from a 3 phase vfd. A second transformer is needed to get the correct voltage for the second phase

[ace1903]

Keeping capacitor connected to the same phase as the motor will keep acceptable phase shift and current balance(between main and starting winding).
Trick is that when inverter is powered by 220v 3phase output has lower voltage so it is you get v/f compensation for free
Torque curve isn't ideal so that is why I think it will work for some loads and for other not.

So if one wants to find proper inverter v/f drive is keyword to search for.

[PSR B1257]

when inverter is powered by 220v 3phase output has lower voltage

The maximum output voltage is always equal to the input voltage independent of 1 or 3 phase system, and of course measured between phases.

[Haenk]

Looking again at the original post- is that really is a 900 watts motor. That's around 1 hp and doubt it that would be a cap *run* motor. Usually cap start and centrifugal switch for that sort of power.

What is the motor driving ?. If the thing is belt drive, why not just fit a smaller pulley to the motor ?. Usually use taper lock bush pulleys here for tasks ike that. Not the cheapest, but no fiddling with keys and keyways. Fit once and forget...

Thanks for your thoughts. Unfortunately, I don't have the machine in hands (yet); and part infos come only from explosion drawings. I doubt it's a start cap, at 900W (I think they state about 1.2hp) it's just 4A (plus some extra at startup) which can easily be handled by 16A fuses. But for running on single phase, it will certainly require a running cap - so that's my assumption for now.

As said, it's an IP approved machine (i.e. a tile cutting saw with water contact), the axle directly drives the saw blade. No gears, no mechanics, no driving wheels, no belts involved. So actually going with an inverter really might be an issue, if the motor power is greatly reduced.

Oh, and I don't want to rewire the motor. But replacing the motor is certainly an option when the cheapo method fails.

I'll let you know how this works out, when the machine is here.

[duak]

Fun Fact: The angle between phases in a 3 phase system is 120 degrees when referenced to neutral, ie., the common connection point at the center junction of a wye arrangement.  When connecting a typical VFD that doesn't really have a neutral, one of the phases becomes the reference and the electrical angle between the remaining two phases becomes 45 degrees.  This is sufficient to operate a reversible capacitor run (not capacitor start) motor without the run capacitor if the motor has identical windings.  I don't know if the OP's motor has identical windings, but i would try running the motor off the line and measure the winding currents.  When connected to the VFD, I would measure the currents and if the run winding current was higher than before, use a transformer to reduce the voltage to the run winding.  If the VFD is programmed in Volts/Hz mode, it should scale the voltage as the frequency is reduced to maintain the correct current and thus output torque as the speed is reduced.  Cooling might be an issue, and the motor could run hotter than usual.

Cheers,
 

[PSR B1257]

Fun Fact: The angle between phases in a 3 phase system is 120 degrees when referenced to neutral, ie., the common connection point at the center junction of a wye arrangement.

The angle between phases is always 120°, regardless of a neutral conductor and regardless of Delta- or Star-configuration. A phase angle refereed to neutral does not even exist.

So actually going with an inverter really might be an issue, if the motor power is greatly reduced.

This is true for every motor. Less speed, less power.

[Siwastaja]

The angle between phases is always 120°, regardless of a neutral conductor and regardless of Delta- or Star-configuration. A phase angle refereed to neutral does not even exist.

Yeah, remember though that inversion, aka 180° phase shift, can be done, and is done in every motor, by simple passive arrangement: another winding wired in the opposite direction.

This is how a 2-phase motor (this is what "1-phase capacitor run" motor actually is; this is what most stepper motors are; and this is also the imaginary model used in 3-phase motor control algorithms!) actually has 4 poles capable of generating magnetic field waveforms in 0, 90, 180, and 270 degrees, when supplied with 0 and 90 degree electrical waveforms.

Similarly, when a 3-phase motor is supplied with three phases 120deg out of phase, the simplest possible so-called "1 pole-pair" or "2-pole" (which means 2 poles PER PHASE) motor creates, with its 6 poles, waveforms of 0, 60, 120, 180, 240 and 300 degrees, when supplied with 0, 120 and 240 degree waveforms. And this underlines the mathematical beauty of 3-phase distribution (the thing Nikola Tesla figured out): the angular "resolution" doubles from what you supply. This does not happen with 2 phases, where you create two signals 90 deg out of phase and get four out which again are 90 deg out of phase.

Duak's claim of 45 degrees from a 3-phase inverter I don't buy, but reserve the right to be wrong and corrected here, maybe I'm missing something; I'm only a motor control semi-expert. But 60 degrees is actually a better approximation of the desired 90 degrees than 45 is, so maybe duak meant that one

[duak]

I haven't been clear in what I mean by the angle between the phases above and by the term reference.  Last first - I should have said Ground Reference; ie., the common potential or connection to which all voltages are measured or compared to.

First case: An isolated or battery powered oscilloscope with three single ended inputs will show the 120 degree sequence of the three phases if the ground clips (signal returns) of the oscilloscope are connected to neutral which is the common point of the wye.

Second case: Now disconnect the ground clips from the neutral (with circuit de-energized) and connect them to Phase U.  What does this do?  Phase U will now show 0 Volts.  Phases V and W will now show a voltage of 1.73 or sqrt(3) x greater than the phase to neutral voltage.  ie., this configuration sums the three phases into two phases of greater amplitude with a different phase relationship.  (Siwastaja is correct, the angle between these phases is 60 degrees, not 45 as I said above. D'oh!!, memory fault!)  I've attached a phasor diagram of what I mean.

The included angle between phases V & W is 60 degrees if phase U (not neutral) is the reference.  This is important because the OP's single phase PSC motor will have to be connected so that the two windings share a common connection to one of the VFD output phases.  For purposes of analysis with a single phase motor it is conceptually easier to view  the output of the VFD as a two phase system because the motor is connected only to the U, V & W terminals and the neutral is not used.

I used this to drive a existing two phase motor (with identical windings) to make a variable speed reversible drive.  It is also in part the basis of the Scott-T transformer connection that requires only two transformers instead of three for the classical 3 phase configuration.

I hope this is clearer and makes more sense.

Cheers,

[PSR B1257]

Second case: Now disconnect the ground clips from the neutral (with circuit de-energized) and connect them to Phase U.  What does this do?  Phase U will now show 0 Volts.  Phases V and W will now show a voltage of 1.73 or sqrt(3) x greater than the phase to neutral voltage.  ie., this configuration sums the three phases into two phases of greater amplitude with a different phase relationship.

That's correct, now I know what you meant.

because the motor is connected only to the U, V & W terminals and the neutral is not used.

Fortunately, there is no neutral terminal at the VFD in the first place 

I used this to drive a existing two phase motor (with identical windings) to make a variable speed reversible drive.

Would be interesting to have the motor hooked up to a dynamometer and measure its performance and compare it to a properly driven motor.

As mentioned above, the motor will certainly do something...

In @Haenks case, he might get away with it, since a tile-cutter not necessarily need the full motor power if the cutting blade is feed gently into a slim tile. Hogging through a thick plate however is most likely a problem with only 30% (or even less) of the rated power or takes forever, if you have to feed the blade too gently.

[Zero999]

What sort of load does the motor drive?

Is it acceptable for it to temporarily run up to near full speed, during start-up before being reduced? That would enable it to still work, if it has a centrifugal switch.

[johansen]

Second case: Now disconnect the ground clips from the neutral (with circuit de-energized) and connect them to Phase U.  What does this do?  Phase U will now show 0 Volts.  Phases V and W will now show a voltage of 1.73 or sqrt(3) x greater than the phase to neutral voltage.  ie., this configuration sums the three phases into two phases of greater amplitude with a different phase relationship.

That's correct, now I know what you meant.

because the motor is connected only to the U, V & W terminals and the neutral is not used.

Fortunately, there is no neutral terminal at the VFD in the first place 

I used this to drive a existing two phase motor (with identical windings) to make a variable speed reversible drive.

Would be interesting to have the motor hooked up to a dynamometer and measure its performance and compare it to a properly driven motor.

As mentioned above, the motor will certainly do something...

In @Haenks case, he might get away with it, since a tile-cutter not necessarily need the full motor power if the cutting blade is feed gently into a slim tile. Hogging through a thick plate however is most likely a problem with only 30% (or even less) of the rated power or takes forever, if you have to feed the blade too gently.

Usually capacitor run  motors need a higher voltage than the main run winding. Capacitor start only motors typically have a high resistance lower voltage winding. You can still drive these motors from a vfd, but you have to be very careful not to burn up the start winding.

So what you want to do is measure the voltage across the capacitor winding under load. Some things like 120v refrigerant compressors will have 180v on the cap winding, 120 on the run winding, in my case the motor was more efficient at 130/110.

Driving a cap run motor from 2 phases will work, use an auto transformer as needed to change the ratio. The motor will still have a rotating field but it will be about half as effective, as using the methods i have described.

Schematic is attached