Hey guys. Had an HVAC tech contact me about an outside self-contained AC unit for a warehouse, a customer we both work for. They changed the blower motor out because the bearings were growling and got 2 years out of it, and now the bearings are growling again. Their supplier is suggesting we might be getting a current path through the bearings causing arc/pitting and taking out the bearings. I've only heard of this happening, never seen it though. This unit is 480v and VFD driven. I haven't looked at it yet, we are waiting for the new motor to arrive to meet up on it. Any thoughts on what I could be looking for, or to do, to remedy this? About all I can do at the moment is verify connections, and make sure our grounding system is good. Didn't know if anyone here has any experience with this? I suspect maybe something the VFD could be causing this also? I would almost consider it just bad luck but its 1 of 4 identical units, so something must be up. And it's a very clean facility, food grade storage, its kept spotless.
Bearing current
- Thread starter User Name
- Start date
ptonsparky
Tom
- Occupation
- EC - retired
Only one for four identical units is having an issue?
What is the run time of it vs the other three? Fan switch turned to ON?
I would think the Mfg of these units would have suggestions concerning this issue.
What is the run time of it vs the other three? Fan switch turned to ON?
I would think the Mfg of these units would have suggestions concerning this issue.
ptonsparky
Tom
- Occupation
- EC - retired
I couldn't remember where I had seen these.
ptonsparky
Tom
- Occupation
- EC - retired
Which would mean something failed if the units are identical.
- Location
- San Francisco Bay Area, CA, USA
- Occupation
- Electrical Engineer
The ability of a VFD circuit to create bearing currents, and the amount created, is HIGHLY susceptible to numerous field conditions, so it’s entirely possible for “identical” units to have varying degrees of severity, just based on how conductors were run, settings in the drives, slight motor manufacturing variations etc. That said, it’s highly likely that the problem exists in all of them, just to vetting degrees, meaning this is just the first to fail.
If they are buying replacement motors, and you KNOW that this problem exists, tell them to ABSOLUTELY specify that the replacement motors have shaft grounding brushes. It’s as simple as just telling the motor supplier that it’s required.
My preferred method is to incorporate them into the shaft seals of the motor. Impro Seal, one of the best shaft seal systems, offers this option and they are designed to last as long as the seal lasts, so when the motor is serviced to replace the seals, the grounding brushes get replaced too.
If they already bought the replacement motors and they didn’t get shaft grounding brushes, then the next solution (more expensive) is to add a DV/DT filter to the output of the drive.
If they are buying replacement motors, and you KNOW that this problem exists, tell them to ABSOLUTELY specify that the replacement motors have shaft grounding brushes. It’s as simple as just telling the motor supplier that it’s required.
My preferred method is to incorporate them into the shaft seals of the motor. Impro Seal, one of the best shaft seal systems, offers this option and they are designed to last as long as the seal lasts, so when the motor is serviced to replace the seals, the grounding brushes get replaced too.
If they already bought the replacement motors and they didn’t get shaft grounding brushes, then the next solution (more expensive) is to add a DV/DT filter to the output of the drive.
ptonsparky
Tom
- Occupation
- EC - retired
So the mfg difference between supposedly identical pre packaged units would account for the bearing fluting? Or the way the electrician connected that unit vs the other three?
Thanks for all the tips guys really appreciate it! I'll look for an output filter if it's there or not, if not I'll suggest we get one. Or maybe there is a failure with that and no one noticed or looked before. And if we come up empty handed I'll suggest we try the shaft brushes. The previous electrican could have done something sketchy with the install too, I've found corners cut with other stuff in this same warehouse.
Eddie702
Licensed Electrician
- Location
- Western Massachusetts
- Occupation
- Electrician
I did HVAC & Electrical before I retired. This is a very common problem with VFDs. They are severl companies out there with various solutions to this problem.
We had air handlers that were only 1 year old when the bearings started to get noisy
We had air handlers that were only 1 year old when the bearings started to get noisy
- Location
- San Francisco Bay Area, CA, USA
- Occupation
- Electrical Engineer
Not necessarily IF, but more of a how much thing. That’s why I said it is likely this is just the one that failed first.
I was brought in to consult at a Bayer facility where they were losing motors right and left. They were all on the same transfer pumps, same manufacturer, same general setup. Some were lasting just weeks, others were months and a few were years. But every one of them were failing from bearing EDM damage (EDM = Electric Discharge Machining, like a plasma cutter, which is what is happening expect on a micro scale). They added shaft grounding bushings to the motors, problem went away.
- Location
- San Francisco Bay Area, CA, USA
- Occupation
- Electrical Engineer
One possible cut corner issue is grounding. No ground, bad ground, ground not brought back from the motor and terminated to the VFD (not somewhere else first), motors on rubber isolators and not grounded to the building grid, stuff like that. Also if they are exposed to public access, tweakers cutting the ground wires off to sell for scrap.
Yeah notably last year I found one of the 120/240v subpanels in the building had no equipment ground run to it. Just the two phases and neutral. Pulled in pvc conduit, so no conduit path for it either. So I already know grounding was omitted at one point in the project.
cadpoint
Senior Member
- Location
- Durham, NC
I wondering that myself! Are the line side circuits all run in solo conduits out of the VFD's?
junkhound
Senior Member
- Location
- Renton, WA
- Occupation
- EE, power electronics specialty
I noticed a lot of generalization, but no one has described for User Name the mechanism for the ESD/EDM currents.
VFD operate at 20 kHz (units with IGBTs) up to few hundred kHz for more advanced units with FET switching
These high frequency signals have a circuit path from the collector or drain of the switching elements to the heat sinks, the current return for this path is thru the motor winding to stator capacitance, then thru the stator to rotor capacitance, thence thru the bearings to grounded motor frame. The bearing balls roll on a thin film of oil, above 300 volts high frequency (the EDM damage mechanism) that film and associated airgaps arc. Note that this can, in some cases, mean a POORLY grounded motor frame (as NEC grounds defined) can actually have less bearing damage!
Grounding the shaft provides on means of a direct path of the high frequency currents back to the heat sink without going thru the bearings. Grounding here means connection to the VFD heat sink, not earth or equipment ground. Again,this can result in a poorly grounded (as NEC grounds defined) VFD showing less bearing damage! Somewhat counter intuitive to folks with NEC grounding understandings only.
One can visualize all sorts of various high frequency current paths when other grounds are missing or very long high inductance paths.
IEEExplore has literally hundreds of papers written on the subject with fixes and causes discussed.
VFD operate at 20 kHz (units with IGBTs) up to few hundred kHz for more advanced units with FET switching
These high frequency signals have a circuit path from the collector or drain of the switching elements to the heat sinks, the current return for this path is thru the motor winding to stator capacitance, then thru the stator to rotor capacitance, thence thru the bearings to grounded motor frame. The bearing balls roll on a thin film of oil, above 300 volts high frequency (the EDM damage mechanism) that film and associated airgaps arc. Note that this can, in some cases, mean a POORLY grounded motor frame (as NEC grounds defined) can actually have less bearing damage!
Grounding the shaft provides on means of a direct path of the high frequency currents back to the heat sink without going thru the bearings. Grounding here means connection to the VFD heat sink, not earth or equipment ground. Again,this can result in a poorly grounded (as NEC grounds defined) VFD showing less bearing damage! Somewhat counter intuitive to folks with NEC grounding understandings only.
One can visualize all sorts of various high frequency current paths when other grounds are missing or very long high inductance paths.
IEEExplore has literally hundreds of papers written on the subject with fixes and causes discussed.
junkhound
Senior Member
- Location
- Renton, WA
- Occupation
- EE, power electronics specialty
See previous post, in some cases, motors on rubber isolators can be a solution to bearing currents! But obviously gives rise to a whole new set of safety problems.
- Location
- Placerville, CA, USA
- Occupation
- Retired PV System Designer
But, since the mechanism of bearing damage is specifically the voltage difference between the motor shaft and the motor frame (hence applied to the bearing oil film), what is really necessary is not grounding, but bonding. A low impedance path between the motor shaft and the motor frame will prevent the damage, totally independently of any of the finer details of the motor frame grounding. "Proper" grounding of the motor shaft and the motor frame separately is one way of achieving bonding, but is indirect and potentially complex.
Having a potential on the motor frame because of improper or inadequate grounding is a totally separate issue from EDM damage.
Just as in parts of the NEC, the difference between grounding and bonding, and the reasons for both, need to be understood.
junkhound
Senior Member
- Location
- Renton, WA
- Occupation
- EE, power electronics specialty
Thanks for the further clarifications.
I attempted to do so by using the phrases (as NEC grounds described) and 'NEC grounding understanding only.'
Yours is a more complete explanation of what I attempted with those phrases.
I attempted to do so by using the phrases (as NEC grounds described) and 'NEC grounding understanding only.'
Yours is a more complete explanation of what I attempted with those phrases.
kwired
Electron manager
- Location
- NE Nebraska
What you just described sort of explains why appliances with VFD driven motors don't play well with a GFCI in the supply circuit as well.
I have encountered clothes washers that isolate the motor frame from the EGC through a "door switch" so to speak. There is a EGC to the motor frame but this switch opens continuity of it as long as the rear access panel of the appliance is in place. This lets the machine run on a GFCI as long as the cover is on, or in my case when I first discovered this as long as the switch is intact. The switch mounting provisions broke on the one I discovered this concept on and that is why it kept tripping GFCI. Appliance repair man told owner the GFCI was bad
They called me and I right away figured GFCI was likely fine and there was a fault in the appliance.