Altivar 630 Ground & Motor Short Circuit Faults, Motor Trouble Too

[jimjam300]

Howdy.
I have an Altivar 630 drive running a 300hp motor on a vertical turbine water pump. Drive was a replacement for an Altivar 71 that lost an IGBT and output board before my tenure and the new drive was commissioned by Schneider factory techs last year. We've been having constant problems with the whole setup and it's hard to tell if any of the issues are with the drive or if they're all motor/pump. Drive has input and output reactors. Motor connected by VFD cable which is about 100 feet long. Motor termination is by copper split bolts.

First thing we noticed was that the drive doesn't give the typical VFD whirr noise when running, it just sounds off. This persists even when changing carrier frequency from min to max. Someone decided to leave it at 2kHz. The motor and drive do not get hotter than normal as far as we can tell.

Last year one of the motor-side leads melted inside of the motor junction box. I think it's possible the split bolt terminal was loose, but the motor might have been on its way out considering that a couple months later the motor shorted to ground and we had to get it rebuilt. I don't remember if it was the same phase as the melted wire.

Fast forward 6 months, drive tripped on ground short circuit. I changed carrier frequency to 4kHz according to the manual and double checked all parameters. Then I ran it again and it tripped after 5 minutes. I megged the VFD cable on both sides, megged the motor, and used two different meggers to be sure. I turned the motor shaft by hand and let it spin while I megged the motor. Had the guys who rebuilt our motor send out a tech with a fancy tester to check the windings and insulation. Everything checked out. Re-wrapped the split bolts and put it back in service. Ran for 2 weeks and got another ground short fault. Reset, ran a few days then a motor short fault. Megged everything again, no indication of failure. Put it back in service and it faulted again a few days later.

I've done some searching online for others who have had similar issues. I see people talking about electrical noise, pinholes in wire insulation, heat expansion, and motor vibration. The last one sounds convincing, like the pump is unbalanced or something, considering all the previous issues we've been having and I'd love to know a good (and preferably inexpensive) instrument I can use to log vibration, and if anyone has some pointers on a guide for checking, selecting equipment, peak vs rms etc.

In the meantime we just got a Fluke MDA-550 drive analyzer and once I figure out how to use it, I am going to use it to record the drive. I'm super pressed for time with other projects, so would it be fruitful to check with AND without load on the output and DC bus (4 tests total), or can I save time by checking it just one way or another?

 

[Electromatic]

SEWAG: bearing fluting? This is something I've only read about. It seems that the use of a load reactor and VFD cable would prevent that, though. Just throwing it out there.

 

[jimjam300]

The Fluke analyzer we got can do motor shaft readings so I'll check that out.

A couple days ago I did end up using a Fluke 805fc to measure vibration against ISO 10816 standards and everything checked out fine, but it would be nice to catch it when the drive faults.

 

[petersonra]

It might not be in the motor. You could have some kind of intermittent fault inside the drive that is causing the ground fault.

You might want to take a very very close look inside the drive and see if you notice anything unusual. You might see something you might not.

 

[synchro]

Did you test the reactor on the output of the drive with a megger?

 

[Jraef]

If the drive is tripping on Ground Fault, it is doing a "residual current detection" (RCD) method for GF protection, so it is not actually measuring anything except the 3 phase currents, looking for them to sum to zero. If it doesn't, it "assumes" that the reason is because some current is going to ground. But anything that changes the balance significantly will trip it too. One of those things is capacitive coupling current in the conductors. So long as this is the same in all 3 conductors, it is not picked up as a difference, so not a GF situation. But if one of the cables is leaking current to the grounded shield, then it may be high enough impedance to not show up as a dead short, but it MIGHT be enough to fool the RCD. The VFD cable is the best way to avoid that, but it might be that in the previous failures in the motor, some collateral damage to the cable or shielding took place. The reason I suspect that is because you mentioned it getting worse when you increased the carrier frequency. Capacitive coupling current gets worse with increased carrier frequencies. It's tough to find that with standard test equipment though. I think what you would have to do is disconnect the cables at both ends and megger them to the shield at at least 1000V.

 

[petersonra]

Unfortunately these intermittent things that aren't real obvious to find are not real obvious to find and you may have to resort to just changing things until the problem goes away

 

[Jraef]

Unfortunately these intermittent things that aren't real obvious to find are not real obvious to find and you may have to resort to just changing things until the problem goes away

How many times…

The one like that where I almost lost it one day was at a corn hog processing facility in L.A.. I did NOT want to be there for another second, the smell was almost as bad as a rendering plant. But nothing I tried was working. The motor leads were about 150’ through 4” conduit under a driveway, so I didn’t want to have to pull them, but that was the only thing left. Pulled them out, the conduit was full of water even though we were in a very dry area in the middle of summer. Blew the water out, it stank of the same smell in the entire plant. Meggered the cables and they were fine, laid them across the ground and connected the motor for testing, no tripping of the VFD. Apparently the water changed the capacitance and that was what was making the drive trip.

Turned out they had a wastewater leak under that driveway that was filling the conduit. It was also, it turned out, undercutting the road bed, so it would have become a huge sinkhole any day. That means the VFD tripping saved them from suffering the sink hole damage!

 

[petersonra]

Sometimes I also think that you get a multitude of minor problems that build up on you to the point where the thing won't work anymore but it's not obvious why because there's no one thing that's bad enough to cause it to fail.

Usually you can swap things around and prove to yourself that a particular thing is not at fault and pretty much rule out everything. So what the hell does that leave?

It is not very satisfying to fix stuff by randomly changing things until it starts to work, at least to me, but sometimes that's all you are left with.

 

[ptonsparky]

Sometimes I also think that you get a multitude of minor problems that build up on you to the point where the thing won't work anymore but it's not obvious why because there's no one thing that's bad enough to cause it to fail.

Usually you can swap things around and prove to yourself that a particular thing is not at fault and pretty much rule out everything. So what the hell does that leave?

It is not very satisfying to fix stuff by randomly changing things until it starts to work, at least to me, but sometimes that's all you are left with.

The only thing that helps is when finally you hit the right solution and it is identifiable.