Increasing induction motor frequency

[Guineafowl]

TL;DR Summary

What are the electrical consequences of running an ac induction motor above design frequency?

This is a question on a welding/machining forum.

The OP has a lathe powered by a 6-pole (c.1000rpm) 3ph, 3hp, 50Hz, inverter-duty motor run in delta at 240V by a VFD capable of 400Hz output. Hobby use.

The lathe is mechanically set to low speed, for maximum torque when parting off, etc. He wants to use frequency control to increase speed for convenience, to save repeatedly swapping belt positions. What is a sensible maximum frequency? The motor specs don’t provide one. As far as I can see, there are are mechanical and electrical considerations:

Mechanical:
- Bearings are rated well above any speed contemplated.
- Rotor balance may become a problem above design speed, but the same motor is offered in a 2-pole (c.3000 rpm) version.

Electrical:
- Torque will drop off steeply above 50Hz, but since higher speeds are used for smaller parts, this may not matter too much.
- Skin effect at higher frequencies will increase effective winding resistance, but this may be offset by higher fan speed and so better cooling.

Given the above, I can’t immediately see why operating at 150Hz (c.3000 rpm for the 6-pole), or even higher, would be a problem. Is there anything we’re missing? Remember, this is hobby use, so perhaps we could push the envelope further than in an industrial setting.

 

[jrmichler]

Not a problem, because the only difference between 2, 4, and 6 pole motors is the stator windings. They all use the same rotor, and those rotors are normally designed for 2 pole 60 Hz motors (3600 RPM) plus a safety factor. Back when I last bought a VFD, I was told that any induction motor was good for 4000 RPM. That might be different if the motor manufacturer only sold motors in 50 Hz countries, in which case the maximum would be about 3300 RPM.

The motor will run at higher speeds, BUT:
1) The rotor has a critical speed, above which it will vibrate until it hits the stator. The bearings will also be overloaded if this happens.
2) Centrifugal forces will cause rotor problems above some speed. That speed is probably very high, but I do not positively know that.
3) The bearings also have a maximum speed.

Try it. If you observe motor vibrations, slow down. Immediately.

Torque will drop off steeply above 50Hz, but since higher speeds are used for smaller parts, this may not matter too much.

First order approximation: Constant torque below 50 Hz, constant power above that frequency. So torque will be 1/2 at 100 Hz, and 1/3 at 150 Hz.

Skin effect at higher frequencies will increase effective winding resistance, but this may be offset by higher fan speed and so better cooling.

I believe that inverter harmonics are a significant source of increased heating. Need some input from a EE on this.

 

[Guineafowl]

Thanks. This is someone else’s machine, but I do have some direct experience.

My own milling machine is powered by a smaller 4-pole 3ph, 50Hz motor and VFD. There is a pot to adjust frequency, set between 10-100 Hz.

The machine’s mechanical speed is set moderately low, such that the grunt work of milling is done at 50Hz, small cutters/drill bits at up to 100Hz, and thread tapping, slitting saw etc at 10-20Hz.

Below 50Hz, the VFD is able to compensate the V/F ratio so the only remaining problem is reduced cooling at the lower speeds. Again, this is hobby use.

100Hz operation (c.3000rpm) doesn’t seem to bother it. The setup has been in place for a year. Milling jobs often involve multiple tool changes in one spot, requiring different speeds, and the reality is, I can’t be bothered to keep changing the belt position when I can just twiddle a pot instead. I guess time will tell.

 

[Tom.G]

Just a history to keep in mind.

In the early days of VFDs there was a problem when driving "Large" motors. The presence of higher frequency harmonics coupled to the rotor, then flowed thru the bearings to the grounded frame of the motor.

End result was the bearings would wear out quite rapidly due to arcing. I don't know what the fix was, but I haven't heard about the problem in several years.

Cheers,
Tom

 

[Guineafowl]

Just a history to keep in mind.

In the early days of VFDs there was a problem when driving "Large" motors. The presence of higher frequency harmonics coupled to the rotor, then flowed thru the bearings to the grounded frame of the motor.

End result was the bearings would wear out quite rapidly due to arcing. I don't know what the fix was, but I haven't heard about the problem in several years.

Cheers,
Tom

That seems to be related to the kHz-range switching frequency. Apparently fixed by installing a shaft grounding ring.

 

[onatirec]

Torque will drop off steeply above 50Hz, but since higher speeds are used for smaller parts, this may not matter too much

Torque should not drop off steeply if you increase the bus voltage (amplitude) along with the frequency.

Skin effect at higher frequencies will increase effective winding resistance, but this may be offset by higher fan speed and so better cooling.

Skin effect is almost certainly not an issue at 150Hz. Probably you will see increased heating from core losses in the lamination steel (eddy, hysteresis, and stray losses from slotting and permeance harmonics), and not an appreciable increase in copper losses at the same current; though these increased losses will push rotor and armature resistances slightly higher.

 

[Baluncore]

The lamination thickness for the induction motor field was selected for the frequency of operation. When you increase the frequency, some of that material becomes inaccessible due to skin effect, so the inductance falls, but the rising frequency needs less inductance to limit the field current. Consider compensating voltage for frequency, to limit the field winding and lamination temperature.

The skin effect in the non-magnetic rotor should not be a major problem, as the frequency of the currents induced in the rotor are proportional to slip %, and the load is light.

 

[Guineafowl]

The lamination thickness for the induction motor field was selected for the frequency of operation. When you increase the frequency, some of that material becomes inaccessible due to skin effect, so the inductance falls, but the rising frequency needs less inductance to limit the field current. Consider compensating voltage for frequency, to limit the field winding and lamination temperature.

The skin effect in the non-magnetic rotor should not be a major problem, as the frequency of the currents induced in the rotor are proportional to slip %, and the load is light.

The OP had mentioned reading that running well above 50Hz is ‘hard on the windings’ but had no further detail. Maybe it was this.

It’s not especially feasible to increase the supply voltage dynamically with the speed - some sort of boost autotransformer that would need to be adjusted along with the speed pot, I guess? He’s pretty much stuck with 240V.

Given that, and the V/Hz compensation, sensorless vector control and torque boost functions of some VFDs, would he be wiser to set the mechanical speed a notch or two higher, then use the VFD at lower frequencies, eg 25-75Hz?

 

[Averagesupernova]

As far as I know, running a 3 phase motor with a VFD allows for reduced frequency with no ill effects as long as the volts/hertz ratio is followed. However, I was always under the impression that raising the frequency is fine as well, but the voltage cannot increase past the motor's rated voltage. So the torque drops off as speed increases above 50/60 hertz.

 

[Guineafowl]

However, I was always under the impression that raising the frequency is fine as well, but the voltage cannot increase past the motor's rated voltage.

This short article by a reasonably well-known supplier of motors/VFDs suggests you can. I think it’s assumed that you’ll be using a modern, inverter duty motor.

https://inverterdrive.com/HowTo/Increase-Motor-Power-Output-and-Speed-by-73-delta/