Extreme Motor Vibration and Now Motor Failure on VFD in Desperate Need of Help! 2

[chpumped]

I am working on a 250 HP 3 phase 480 V 1800 rpm motor on a vertical turbine pump that is run on a saftronic vfd, it had some vibration to it that led to the bearings being replaced and re-installed. Once installed it ran smoothly and we put the pump back in service, a week or two later it began shaking very violently at about 50+hz. So we pulled the motor and brought it back into the repair shop, once hooked up to their power it ran VERY smoothly at 60 hz on their test power.

I connected a small 7.5 HP 1800 RPM motor to the vfd and it started shaking violently and we shut it off. At that point we determined that the VFD was the issue and had a vfd specialist come in, he checked voltage, amperage etc. and determined that the drive was good.

Then a 3600 RPM 150 HP motor was hooked up to the drive and it runs smoothly with very little vibration. So we hooked the original 250 HP 1800 RPM motor back up to the drive and it now will hardly spin with a loud whine at a rough 10-15 RPMs. The motor has had a surge test, meger test done on it, and it passes with flying colors. For all testing it was uncoupled from the pump, the motor still spins freely and smoothly when spun by hand.

Any ideas/suggestions would be greatly appreciated!

 

[jraef]

1) bearing damage is a very well known phenomenon now, you might want to think about mitigation. Do a search on "VFD bearing EDM damage" to learn more.

2) If your VFD was programmed to be in "SVC" mode, hooking the small motor to it would likely give very strange results. Otherwise (ie it is in "V per Hz mode") it should not have cared.

3) Saftronics has been gone as a name now for long enough that the likelihood that this is an old drive is high. That means you may have lost or are losing a capacitor or two, or more. With a lighter load, the bus ripple may not be too bad, but at designed load, the drive can't handle it. A good "VFD specialist" would have known how to check that. If you watched him and all he did was read voltages and currents, he's not that good. That's just one possibility, there are others, such as a failed or failing transistor, firing circuit etc. etc.

"Will work for (the memory of) salami"

 

[chpumped]

jraef,

Thank you very much for your comments.

1) That was my first concern, the previous bearings that were on the motor showed zero signs of EDM damage.
2) We did go through all drive settings initially to ensure it was not a setting and it is on V per Hz mode.
3) I was not there when the VFD specialist inspected, but according to the electrician that was with him it did sound very basic to me.

 

[mikekilroy]

Saftronic is old isnt it? Pre vector? If so, no idea, other than it is intermittently broke?

But all your experiments seem to me to say it is in sensorless vector mode and the parameters are wrong. If so, verify by changing mode back to simple skalar (aka open loop, volts/hz mode) and bet it runs fine. If so, KISS, leave it, or go back and retune it to the motor properly.

http://www.KilroyWasHere<dot>com

 

[chpumped]

The thing that seems strange to me is that the issue seems to have gotten worse over a short period of time with the original motor. It has quickly gone from a smooth running motor to bad vibrations (2x running speed), and now to essentially non-functional. While in the mean time we have connected the motor across the line and ran very well. I guess I am wondering if a setting such as SVM would degrade over time all of a sudden?

 

[jraef]

Maybe a cracked rotor bar or end ring connection? That does not show up on stator tests, and it almost always shows up as vibrations. Still doesn't explain the shaking when the 7.5HP motor was connected though, unless that was an outlier because of a bad connection or something.

Saftronics could be Yaskawa or Fuji, depending on age, but both of those were offering SVC drives at the time the brand-label deals were still in place. Still could be associated to that issue, you would have to check the programming to be sure. As to why it would have started getting worse? A combination of issues? If it was in SVC** mode, the cracked rotor bar/end ring would change the inductance values, which changes the motor model, which could account for worsening erratic behavior.

SVM is Space Vector Modulation, that is different from SVC, Sensorless Vector Control.

"Will work for (the memory of) salami"

 

[ozmosis]

I've referenced this article before and read Allan Budris' reports numerous times, it might be worth a read and then re-think the VFD as the cause:

http://www.waterworld.com/articles/...-before-applying-a-vertical-turbine-pump.html

 

[John2025]

Ozmo,
A star for a really interesting article. Doesn't explain the small motor, though of course.
John

 

[itsmoked]

Another possibility. We're talking an old VFD. A lot of non-volatile memory devices start forgetting in a surprisingly short time period. It could be one or even several of the drive's parameters have 'self-changed'. Any one of many could cause problems. You should walk thru every last one and make each one makes sense.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[ozmosis]

Thanks John. It's really a star to the author.
Maybe you should look at the acceleration time of the drive and reduce it, to run on the (correct) 4 pole motor.
For VTP's, it is important to get the system up to speed as quickly as possible as the thrust on the bearings is a result of the fluid going the pump. Your tests on a 2 pole motor seem to back this up in a way: i.e, the drive will have a set period to get from 0hz to maximum Hz, but using a 2 pole motor instead of a 4 pole motor, the pump/rotor will get to speed in half the time.
Just a thought.

 

[chpumped]

Update Problem Solved!

Thank you everyone for your feedback, mostly for questioning the integrity of the "VFD Specialist" and pointing me back to the vfd after he said it was running fine. I had another local VFD company come out for a more thorough testing on the vfd components like jraef suggested, and within about 5 minutes they had set the torque boost from .05 to zero and it was running very smooth again! Still don't understand why that would degrade over time but as for now it is running quieter and smoother than it ever has...

 

[mikekilroy]

Wow.

initial problem: a week or two later it began shaking very violently at about 50+hz

boost has NOTHING to do with 50hz operation......

Later: So we hooked the original 250 HP 1800 RPM motor back up to the drive and it now will hardly spin with a loud whine at a rough 10-15 RPMs.

Ah, who cares on a v/hz drive? That's below even the motor slip! So a 'dont care.'

Sound like your 2nd vfd "expert" made some other change but could not remember what they were! Or your problem still exists and is simply intermittent.

IMO if a vfd problem comes up again in the future, you should look for a 3rd vfd 'expert.' They say third time is the charm....

http://www.KilroyWasHere<dot>com

 

[jraef]

Yeah, sounds too clean. Most likely the 2nd expert found something else that was really simple, but chose to "retain the aura" of his ability to work a miracle for you so that you will call him first next time.

Can't say I haven't done something similar myself...

"Will work for (the memory of) salami"

 

[chpumped]

I agree with all of you, I don't think we are fully to the bottom of this yet. However in the mean time I did prove that it is not a mechanical or motor issue.