Gentlemen:

I monitor (vibration analysis) 3 1000 HP Westinghouse Motors (Vertical Pump Applications) controlled by VFD's. Both upper bearing (Kingsbury Thrust and lower guide bearing SKF 6036) are insulated. One motor in particular passes electrical currents through the lower bearing causing electrical fluting damage to the lower guide bearing. In an attempt to establish root cause, we have noticed that all three motors have an "insulated ground" fitting on the lower bearing (see photos). The electrical personnel say this probe is in contact with the outer race. My question to the forum is what is the purpose of such a ground probe and secondly, should it be grounded during normal operation? All comments are greatly appreciated. Thanks - Mike

Insulated_Ground.doc (106 Kb, 92 downloads) Insulated Ground

My guess would be it is a thermocouple probe for the bottom bearing. Apparently the bearing is insulated from the motor bottom bearing bracket to prevent circulating currents, so they bring the probe leads out of the bracket through an insulated bushing. If the temperature indication is not used I would think those probe leads should not be grounded but should instead be safely taped off to prevent any contact. (my guess).

Do you know who installed this jumper (OEM or plant or shop?). What is it attached to on the lower right side... just bolted to the bracket or is that another insulated bushing to run a lead through... and to where from there?

This message has been edited. Last edited by: electricpete,

Electricpete - Thanks for replying to this request. The thermocouple probe for this bearing is 180 degrees opposite this "ground probe". This "jumper" has been installed on all 3 motors since original installation approx. 14 years ago by the OEM. The "jumper" is bolted to the motor frame without insulated bushings etc. As I stated in the previous post, only 1 of the motors is exhibiting electrical fluting on the lower guide bearing with bearing replacement intervals at ~6 months.
We have left the ground "jumper" off on the last bearing replacement - OK so far but nobody can explain the purpose of the "insulated ground".

I don't know but I think it is some kind of instrumentation. Maybe just a well where you could insert a second temperature probe or some kind of vibration probe. Or less likely a point to measure voltage on the bearing.

Whatever it is, it looks like it is intentionally insulated from the motor bearing bracket at the bushing where it comes through the bearing bracket, do you agree? (Also during disassembly, does the bottom bearing appear to be insulated from it's housing?). If so, that strongly suggests that it was intended to be insulated from the bottom bracket, and therefore shouldn't be jumpered to the same bottom bracket.

I will also add since this is a large vfd motor, one would expect that both bearings will be insulated since the capacitive coupling of the high frequency components fo the vfd provides a path to complete the current loop if even one bearing is uninsulated (not so for non-vfd motor where we generally only need to insulated just one bearing). So if you have a large vfd which needs both bearings insulated, why would you ground one of those bearings to the bottom bracket? Doesn't sound right. The only time I would see this could be acceptable is if the bottom bracket is itself insulated from the stator frame to interrupt the current flow. (is it?)

This message has been edited. Last edited by: electricpete,

Electricpete - Appreciate the input and observations. To answer your questions: Yes, the "probe" is intentionally insulated from the motor bearing bracket - Yes, both bearings are insulated from the their respective housings. Your observations are right inline with ours and this probe should not be grounded to the motor frame. We have discussed this issue with a local reputable motor shop and they too are clueless. Seems everyone is in agreement on the issue, that it should not be grounded to the motor frame. Only time will tell - takes approx. 2 months for the bearing defects to appear in the vibration data. - Mike

If the bearing is not insulated properly, current will still pass thru it. causing the flutting.
You can detect this electric arc before the bearing gets damage, by looking at your vibration data in the 2000-2500 Hz range, you will see inner race frequency been generated that is not harmonic to the fundamental. As the flutting starts the inner race frequencies will beging to show, down into the 1000's Hz range and when it becomes harmonic to the fundamental deterioration has started. You can put brushes that contact the shaft and then are grounded giving the current a way out bypassing the bearing.

Correct me if I am wrong, but if you ground the outer race of the bearing, does that not make the path complete for shaft currents to pass through it. Seems to me that defeats the purpose of insulating the outer race. If there are concerns about errant shaft currents, why not add a shaft grounding brush. May be redundant, but how much does a rebuild cost?

Just a mechanical guys 2 cents,
Gary B

To elchaco:
Resistance readings before the motor is ever started up is the way to prevent damage. If you see vibration in the 2k to 2.5k region, too late! EDM takes hours to days to start a bearing on his downward path to ruin (wow, poetic!).
To Data Dog (Man, I hate that term. Take pride in your job):
Please consider that the temperature probes must also be insulated. They will reconnect the ground path if they are not. And, I agree with all of the other responders that told you take disconnect any 'ground to bearing race' connection.
By the way, the insulated bearings remove the current path for low frequency EDM, but the higher frequency currents (5 MHz) capacitively couple across the airgap and charge the rotor. If both bearings are insulated, the coupling, if conductive, will carry these currents into the driven equipment.
The best, slam dunk, fix for VFD applications is the proper grounding in the power cables. This ground path needs to provide the lowest impedance path for the high frequency current back to the drive ground bus. Twisted copper wire (good for 60 Hertz) doesn't work! A 360 degree braided shield surrounding all of the power leads works well. Also a 360 degree copper film shield. Remember, if you do not take these high frequency currents back through a path you provide, they will find their own path back. At best, this will be from the motor frame through your plant ground grid connection. Still not good, since you are contaminating the entire plant ground grid with high frequency noise. Could be an issue for other sensitive equipment.

I thought I would answer but Ron Brook has done such a great job!

I would add only that the motor frame should be properly grounded to both power cable as well as grounding grid using high frequency grounding straps. In addition you can use ferrite filters to reduce the common mode rejection current. If you have problem with EDM you should make these measurements. When you detect a problem in vibration spectra it is too late. The damage has already occurred.

The common mode rejection current can be measured using rogowski coils and an oscilloscope. This is should be done right after you put motor back to service.

We have in operation a total of 13 large Hp / low rpm VFD controlled vertical pump motors, all supplied by the same Italian manufacturer. Only 4 of them have both bearings insulated with the lower drive end bearing having an insulated grounding stud in contact with the outer race of the bearing. Per manufacturer's instructions this stud is to be bonded to the motor frame ( = ground ) during normal operation. The purpose of the insulated ( from the frame ) stud is to allow for testing, i.e. meggering of the bearings insulation systems resistance to ground. For the duration of the test the bonding to ground jumper is disconnected.
Why both bearings are isolated from the frame for these 4 machines ? Nobody could tell us. The other 9 motors have only the top NDE bearing housing insulated.

quote:

Why both bearings are isolated from the frame for these 4 machines ?

On motors for use with VFD's, we need both bearings insulated. Non-vfd motor bearings need only one bearing insualted, usually the outboard end.

The reason is that the current needs to flow in a loop. For standard motor, this involves two bearings. For vfd motor there is high frequency content which creates capacitive linkage to ground (Xc = 1/<2*Pi*f*C> gets smaller as f gets larger) so that a path to ground thru one bearing is enough to cause a problem.

I apologize if I'm telling you what you already know.

I don't know why the motors would be different. If all for vfd application, I would think that all need both bearings insulated.