VFD High Voltage/Torque Output Reading

[naejharas]

I have a crane/hoist application where two identical hoists/motors/VFDs on the same crane appear to be acting differently. One drive (I'll call it Drive A) shows expected output current, voltage, and torque readings at no hook load, read off the drive's keypad. The other drive (Drive B) shows expected output current, but high voltage and torque at no load. Drive B was swapped for a spare, and the spare had very similar readings at no load, leading me to believe there is nothing wrong with the drive.

As expected, the voltages of both drives increase when we increase the frequency/speed, but what is not expected is that Drive B reports about 60% torque at no load, and increases to nearly 150% at rated load/speed, while Drive A reports about 12% torque at no load and close to 100% at rated load/speed.

The drives' output currents were measured to be fairly close to what the drives reported, with no significant imbalancing.

The manufacturer's field technician told me that actual output voltage could not be measured for comparison because it is a DC pulse and that the drive is not actually controlling the voltage, only the current, so the variation in voltage, and therefore torque, is not anything to worry about necessarily (huh?).

Hmmm... here's my concern: two identical drives, motors, encoders, gearboxes, nearly identical lead lengths, and different torque readings. We've ruled out mechanical drag.

Would the drive act this way if the motor had a damaged winding? Or would we see increased or imbalanced currents? Of course the motor supplier is not willing to admit their motor could be damaged, and we're really not in a position to send it back, not knowing if it's the problem.

The drives are 480v, 50hp, Electromotive Impulse VG+ Series 2, connected closed loop flux vector with feedback from a motor-shaft-mounted encoder. Voltage drop is minimal over the full distance of the crane's travel. Each drive is supplied by its own drive isolating transformer. Braking is not connected for regeneration, excess power is dissipated through dynamic braking units/resistors. Motors are 460v, 60Hz, 50hp, inverter duty, 1000:1 CT, NEMA Design B. The motor lead lengths are about 50ft.

Thanks for any insight you may have.

 

[Skogsgurra]

Snip:"The manufacturer's field technician told me that actual output voltage could not be measured for comparison because it is a DC pulse and that the drive is not actually controlling the voltage, only the current, so the variation in voltage, and therefore torque, is not anything to worry about necessarily (huh?)"

Yes, it is not meaningful to measure motor voltage. At least not with a DMM. Not even an old-fashioned moving iron instrument will give the right answer. The reason is (as the FT told you) that you deal with a PWM voltage that contains lots of harmonics. One way to measure that voltage would be to use a meter that filters anything but the fundamental voltage away (one such beast is the m&w MX210, but there are surely others that can be used). If you cannot find such an instrument, you can easily build a filter yourself. Best results if you build it symmetrical and with rather low impedance. I have used two 10 kohm resistors that connect to the motor voltage and on the other side a 7.96 uF capacitor. It may be easier to use a 10 uF capacitor and tweak the resistor values to suit (12.56 kohms).

PLEASE NOTE THAT THIS NEEDS A VERY GOOD UNDERSTANDING OF WHAT YOU ARE DOING. SIZE THE RESISTORS TO WITHSTAND VOLTAGE AND POWER. MAKE SURE THAT YOU DO NOT TOUCH ANY POTENTIAL. IT IS LETHAL.

Connect the DMM across the capacitor.

Such a filter will have a 1 Hz corner frequency and will show the V/Hz figure. Test against a DOL motor to "calibrate" the device and get some confidence in it. A 480 V 60 Hz motor has a V/Hz value that is 8.0 V/Hz.

You can use this filter to measure the motor V/Hz. It should be close to 8 across the whole frequency range (if you are running scalar - or V/Hz - mode). In vector control mode, there is a less constant voltage. It depends a lot on motor loading. But it should not be above 8.0 V/Hz in any situation. Lower is OK at partial load - but never higher than 8.0 V/Hz.

If it is higher, then you have a problem in the drive.

 

[Skogsgurra]

Correction to the above post: V/Hz = 460/60 = 7.67 V/Hz Not 8.00 as I said.

 

[DickDV]

skogsgurra, why go to all that trouble when the drive will, in most cases, display output voltage on the keypad?

I my view, far better than trying to measure it and safer too!

 

[buzzp]

Skogs,
Your claims about rms meters not reading voltage correctly from a PWM drive are bothering me. I have started a new thread to talk about measuring the voltage and current out of a PWM drive. Please join. Thanks.

 

[naejharas]

Thanks for the insight. Is it possible that I'm having trouble with the motor and not the drive (the drive was swapped with a spare and showed the same readings)? In the past, I've pretty much depended on the drive manufacturer to explain things, but on this, they've taken the attitude of "it's weird, but it works."

 

[Skogsgurra]

Swapping drives usually means transferring all parameters from the old drive to the new one. If the parameters are wrongly set, it doesn't help to switch drive.

Have you checked key parameters like nominal frequency and nominal motor voltage? Are they the same on both drive A and B? Sometimes the drives need to know the number of pole(pairs). Check that it is correctly entered. Even the number of pulses per revolution (encoder PPR) can have an influence. It depends on the drive.

 

[niallnz]

Have you checked that the mechanical brake on hoist B is completely disengaging, the motor/drive may simply be driving against the holding brake. Its worth a look if all the other things are normal.

 

[jraef]

I say "BINGO" to niallnz's and skogsgurra's comments.

I have worked on the Yaskawa/Electromotive crane drive in the past. It has a "torque proving" feature that is used for hoisting applications. This circuit applies power to the motor while the mechanical holding brake is still engaged, ensuring that the motor is at full torque before the brake is released. This feature was embedded into their special EPROM used to reprogram the default capabilities from Yaskawa (that was the exclusivity maintained by Electromotive). If one of the drives has that special EPROM and the other doesn't, they will not act the same.

If they are the same, there is still a way to explain the diferent outcomes. You did not mention if you have holding brakes or even if these were hoist drives or not, leading me to believe they are not. If however you have the torque proving feature turned on the one drive (and not the other), AND the motor shaft is not being held with a brake, it may explain the difference. Especially in light of skogsgurra's insight as to repeating the same programming mistake on the replacement unit.

Go over the programmed functions one by one carefully between the differring drives and make sure even the tiniest of details are covered and identical.

"Venditori de oleum-vipera non vigere excordis populi"

 

[jraef]

Oops, just saw that you did say it was the hoist, but still, check that feature out. They may be programmed differently.

"Venditori de oleum-vipera non vigere excordis populi"