Run 575 volt motors on 480 ?

[PowerQualityDoctor]

Measuring Harmonics

Measuring Harmonics

There is difference between the real harmonic pollution and the measured one. The standard requires measurement according to IEC 61000-4-30, which is up to 49th harmony. This means that if you create pollution in frequency higher than 5 kHz (50Hz) or 6kHz (60Hz), the harmonic meter will not see all the pollution.

I had a customer which suffered from unexplained failures. He used Nexus 1252 meter which can measure up to 255 harmonics and everything was OK. I have used Elspec G4430 to measure the same site, which measures up to 5122 harmonies, and found harmonic pollution at 333rd and 334th harmonics. It was 60Hz network, which means he had pollution at 20kHz. Surprisingly, the switching frequency of his inverters was 10kHz. When he changed the frequency - the problems were solved.

 

[Besoeker]

There is difference between the real harmonic pollution and the measured one. The standard requires measurement according to IEC 61000-4-30, which is up to 49th harmony. This means that if you create pollution in frequency higher than 5 kHz (50Hz) or 6kHz (60Hz), the harmonic meter will not see all the pollution.

I generally use a digital oscilloscope with storage capability for harmonic measurements.
This is a detail of the one currently in my office at home:

Capture at that rate doesn't exclude much that is likely to be present on the power system.
It has a FFT facility too, but I usually download the *.csv to a spreadsheet format as well and do analysis from that.
And I carry around a more modern TDS 220 100 MHz with an RS232 comms module in my car boot/trunk.
I like to view waveforms in real time. Not just to get harmonic measurements. But to get a closer look at anything unusual and to be able to look at transient dv/dt and overshoot that might have rise times in the micro second range. It helps in determining the probability of partial discharge failure and the number of pulses to get to the point of probable failure.

I had a customer which suffered from unexplained failures.

But what failed?
And what problem was solved?
Supply impedance is mostly inductive which attenuates harmonic currents. Particularly the higher orders.
Maybe your problem was related to radiated rather than conducted?
Good installation methods generally obviate such problems.

 

[PowerQualityDoctor]

What is the current sensor? Usually, this is what limits the harmonic spectrum.

I don't know what was the particular problem, as it was one of my channels overseas. It was something related to some system got "crazy". The solution was to change the switching frequency. I assume the installation was good, as it is very reputable equipment manufacturer and it was inside their machines.

 

[philly]

Well, there is no simple answer that I know of.
It depends on loading, the extent to which the voltage is reduced and, of course, motor design.

For one specific motor...

Take the 3% reduction in voltage you suggested. To get full load torque (FLT), the stator current would increase by a little under 3%. For 0.5 FLT, the increase is about 1.3%. At a quarter load, the reduced voltage makes the current lower than at full voltage.

That is interesting. So for a motor that is a quarter loaded the reduced voltage actually lowers the operating current? I guess it depends on loading like you said.

I guess the same would apply for overvoltage, and supplying a 460V motor with 480V?

Boiled down to its basic elements, motor torque is essentially a product of stator flux and rotor current.
Reduce the supply voltage and you reduce the stator flux.

By reducing the supply voltage wouldn't you in turn be reducing the rotor voltage for any given speed. A reduced rotor voltage would result in a reduced rotor current if everything stayed the same. I guess however since there is not less avaliable torque the slip increases and therefore the rotor resistance changes lowering the rotor resistance and increasing the current.

 

[Besoeker]

What is the current sensor? Usually, this is what limits the harmonic spectrum.

Perceptive as ever.
As has been posted by others, voltage distortion usually matters more than harmonic current. The voltage measurements are made directly using passive 'scope probes that have a bandwidth of 250MHz and a rise and fall time of 1.4ns.

I don't have a bandwidth specification for the current sensor. But I know it can faithfully replicate a 10us current pulse at a 1000:1 ratio.

Standard inverters have an uncontrolled front end (B6U configuration) and commutation from one phase device to the next occurs at a very small voltage difference. As standard, we fit line chokes except on the larger drives which have their own unit transformer. Supply reactance adds to that so it is not unusual for the reactive component to be several tens of uH.
With the relatively small voltage, the di/dt = e/L won't be large. What falls out from that is that the highest frequency component isn't actually very high in either magnitude or frequency.

I assume the installation was good, as it is very reputable equipment manufacturer and it was inside their machines.

We have just installed a couple of drives on a plastics injection machine.
The wiring inside the machine is iinappropriate for the application.

 

[weressl]

In the mining industry I have encountered several 575 volt motors connected to 480 volt systems. I was wandering what precautions should be observed in this practice? I have not seen any adverse effects. No burned up motors or overload issues at all. How would the actual output of the motor horsepower and current load compare to its nameplate rating?

The motor should be dual rated for both voltages. This is especially important if the work area is classified where the motor needs to be labaled for the classified area. I am not aware that such motor exist or not.

The manufacturer would be able to tell you what horsepower a motor designed for 575V would deliver connected to a 480V system, but doubt that they would give you this as an official information.

 

[Besoeker]

By reducing the supply voltage wouldn't you in turn be reducing the rotor voltage for any given speed. A reduced rotor voltage would result in a reduced rotor current if everything stayed the same.

But, as you noted, not everything stays the same.
If you reduce the stator voltage, you reduce the stator flux. For the same torque you then need greater rotor current, greater rotor voltage, so a bit more slip.