VFD Ground fault

Status
Not open for further replies.
ptonsparky

ptonsparky

Tom
Location
NE (9.06 miles @5.9 Degrees from Winged Horses)
Occupation
EC - retired
Customer with a VFD and an older 200 hp motor. Motor failed showing a ground fault and was. replaced with a similar loaner motor which ran for about a hundred hours when the VFD started showing a ground fault on that one. We ended up going to just South of The Golden Spike to test the second motor. The first took a shot of lightning. No doubt about it from the motor shop. How the VFD survived IDK.
We removed the VFD load wires and tested the load reactor, wires and motor all at one time. We wiggled the wires at the PH, with interesting changes but not a fault. Did both the 1 minute and 10 minute megger checks with the values climbing and dropping repeatedly until it stopped at 1200 mega ohm at the ten minutes. Doing nothing else we wiped the thin layer of dust, dirt, sand from the exposed terminals of the load reactor and along the wires. Did the tests again with both settling in at over 4G within seconds. I expect some change but nothing like that.
 
ptonsparky said:
Customer with a VFD and an older 200 hp motor. Motor failed showing a ground fault and was. replaced with a similar loaner motor which ran for about a hundred hours when the VFD started showing a ground fault on that one. We ended up going to just South of The Golden Spike to test the second motor. The first took a shot of lightning. No doubt about it from the motor shop. How the VFD survived IDK.
We removed the VFD load wires and tested the load reactor, wires and motor all at one time. We wiggled the wires at the PH, with interesting changes but not a fault. Did both the 1 minute and 10 minute megger checks with the values climbing and dropping repeatedly until it stopped at 1200 mega ohm at the ten minutes. Doing nothing else we wiped the thin layer of dust, dirt, sand from the exposed terminals of the load reactor and along the wires. Did the tests again with both settling in at over 4G within seconds. I expect some change but nothing like that.
Click to expand...

When you say interesting results when wiggling wires at motor, can you expand on this? Were there any signs of insulation damage where the conductors enter the PH? Correct me if I'm wrong but a megger puts out high voltage/low current onto the wires. However, when starting a motor, the starting current surge (not as likely with VFD) can cause the wires to move which could cause them to rub and wear down which could be the issue now. If it were the VFD being the issue, I'm not sure why the fault would go away then come back in this particular case so I would rule that out.
 
ptonsparky said:
Interesting results were that when I touched the wires in the PH the Megger readings would drop briefly then start climbing again. (Yes, I had my gloves on).

I changed the size of the insulator when I touched the wire.
Click to expand...

Perhaps more accurately you changed the capacitance between the wires and between each wire and ground (earth).
 
I am assuming this is a LV system (480V?)
What are the GF settings set at on the VFD controller?
Since the wiring between the VFD and motor are the suspected culprit (damage caused by the strike) then what harm
would it do to raise the GF settings (PU & TD) on the VFD and run the motor?
When the Poco out here runs into an intermittent GF they can't locate they put a "Thumper" on it that raises the voltage
till the fault blows and shows it's ugly head!
Either that or hook up a Hipot tester if it's MV gear to break down the faulty insulation.
Another thought: What if the GF detection circuit on the VFD was damaged by the strike and it is showing a false indication of an actual GF.
 
ATSman said:
I am assuming this is a LV system (480V?)
What are the GF settings set at on the VFD controller?
Since the wiring between the VFD and motor are the suspected culprit (damage caused by the strike) then what harm
would it do to raise the GF settings (PU & TD) on the VFD and run the motor?
When the Poco out here runs into an intermittent GF they can't locate they put a "Thumper" on it that raises the voltage
till the fault blows and shows it's ugly head!
Either that or hook up a Hipot tester if it's MV gear to break down the faulty insulation.
Another thought: What if the GF detection circuit on the VFD was damaged by the strike and it is showing a false indication of an actual GF.
Click to expand...
All good questions but until someone tells me it’s broke again, I’m happy. Right now, it’s running, AFAIK.

I hate to remove the GFP. The few drives I’ve had that show up on they have always been correct. I only work on two time’s a few anyway.

there might be 12 feet distance from the VFD to the motor.
 
It would be surprising to see the GF trip settings in a VFD as adjustable, they are typically a fixed percent of the rated load, i.e. 25% using a residual current method taken from the internal current sensors inside (often Hall Effect transducers piped into the mP of the VFD). Usually all you can do is disable it or enable it, maybe. In A-B drives you can't adjust it or disable it, but you can set it up as a "minor fault" and then add that to the list of minor faults that will allow the drive to continue running in that condition until it is shut down, at which time restarting will be disabled until it's cleared. You can also just program a lower level as an alert, i.e. 5% of rated current, before it ultimately shuts down at 25%.

Cable capacitive current can fool them for sure and if the dust was even slightly conductive, I've seen that happen on the newer VFDs that have the latest generation of IIGBTs, because these new transistors have a much faster rise time that before, which makes cable capacitance issues worse. It's rare, but it happens.
 
We were happy for about a week in this ongoing saga. Going on the assumption that the present motor has not failed. (Not even sure of that now. Actual Motor rewind shop called with indication that it was most likely high voltage spikes on non inverter duty wiring. Typical VFD damage. Motor ran for a couple years before we got load reactor on it. Long story.)

First, this is an AB 753. I found out last week the demise of the previous drive was Dust, Dirt and General Crud. Would have been nice to know sooner, although I did make the suggestion years ago to cover this drive in the off season.

Obviously the factory inlet filter is a waste of time for this installation. We trimmed and rolled a section of fiberglass filter to fit behind the factory one, but how do I know what is to much? Guessing if I still get dirt in the cabinet it's not enough.

We will probably be making another trip up there because they will be putting the rebuilt original motor back on tomorrow, and do the Autotune, etc.

Next question, seeing as this is an intermittent problem at the moment is there any cleaning that I could do in the field that may correct a DD&GC problem?
 
ptonsparky said:
We were happy for about a week in this ongoing saga. Going on the assumption that the present motor has not failed. (Not even sure of that now. Actual Motor rewind shop called with indication that it was most likely high voltage spikes on non inverter duty wiring. Typical VFD damage. Motor ran for a couple years before we got load reactor on it. Long story.)
Click to expand...
It is usually high rate of change of voltage (dv/dt) rather than high voltage spikes that do older motors in.
 
ptonsparky said:
My phrasing was is wrong again. Thank you.
Click to expand...
Just my experience. We had one project where the motors were random wound rather than mesh wound. And failed after a few weeks. The motor manufacturer took the pain. As one guy commented at the time you would not normally expect 780 kW motors to be disposable items.
 
In the PF753, there are parameters (941 through 944) that monitors the actual heat sink temperature. If you have a SCADA system, you can monitor those parameters and trend the data so as your filter loads up and the drive heats up, you can generate a PM call.

As to cleaning, I recommend the "canned dry air" you can get at office supply stores for PC, and a decent natural bristle paint brush to knock crud off of the PC boards and out of the heat sinks.
https://ibuyofficesupply.com/perfec...MIja6Bpcqk3AIVF7jACh38uwyREAQYBSABEgKFvvD_BwE
 
Last edited:
Jraef said:
In the PF753, there are parameters (941 through 944) that monitors the actual heat sink temperature. If you have a SCADA system, you can monitor those parameters and trend the data so as your filter loads up and the drive heats up, you can generate a PM call.

As to cleaning, I recommend the "canned dry air" you can get at office supply stores for PC, and a decent natural bristle paint brush to knock crud off of the PC boards and out of the heat sinks.
https://ibuyofficesupply.com/perfec...MIja6Bpcqk3AIVF7jACh38uwyREAQYBSABEgKFvvD_BwE
Click to expand...
No such luck on the SCADA.

Thank you again.
 
Jraef said:
In the PF753, there are parameters (941 through 944) that monitors the actual heat sink temperature. If you have a SCADA system, you can monitor those parameters and trend the data so as your filter loads up and the drive heats up, you can generate a PM call.

As to cleaning, I recommend the "canned dry air" you can get at office supply stores for PC, and a decent natural bristle paint brush to knock crud off of the PC boards and out of the heat sinks.
https://ibuyofficesupply.com/perfec...MIja6Bpcqk3AIVF7jACh38uwyREAQYBSABEgKFvvD_BwE
Click to expand...

I don't know if it's the same product that we have here called Air Duster. If so, a word of caution. It is flammable. My daughter got a bit of a fright when she got flames coming out of her laptop. The lesson is don't use it on live equipment. I just thought ptonsparky ought to made aware of that.
 
Jraef said:
You could map one of those parameters into DeviceLogix and have it turn on an output to light a warning light on the box with a label that says “Change The Filter You Dummy!”
Click to expand...
Most of the places we put drives had routine maintenance programmes that involved changing the filters regularly - once a month IIRC. It's inexpensive and doesn't involve a shutdown. The filter holders were on the outside of the enclosure.
 
Status
Not open for further replies.
Top