Welders & Voltage Spikes?

[lmcgill]

Hello, I am a junior electrical engineer who has been tasked with getting to the bottom of an odd situation one of our client's are having. In short, our client (local high school) has a waste lift station with a duplex pump system installed. They keep burning up the pumps. The pump vendor tore down and reported that the pumps have most likely been "damaged by voltage spikes". The pumps are fed from a panel inside the main building that also feeds a series of welders. We're wondering if the welders are somehow causing irregular and harmful spikes on the feed out to the pumps. The voltage at the pumps is in an acceptable range so voltage drop is not an issue.

Questions are: Has anyone had experience with welder causing damage to other loads fed from the same panel? Also, can anyone point me to a piece of equipment that would be capable of measuring/logging the irregularities?

A locla electrician says he has logging equipment that he can put on the feed to the motor for three days and provide us a print out. I wonder, however, if the sampling resolution is good enough to pick up any of these spikes. Any insight into this would be greatly helpful. Thanks,

 

[just the cowboy]

Is the pump powered by a drive or across the line? I have heard of drives causing damage to bearings in motors and if is a pump that has a custom front flange as part of the motor it may affect it.

 

[Gregg Harris]

Probable and more likely with multiple welders being used simultaneously as in a class room setting.

 

[jim dungar]

Don't forget to look into the bonding of the pumps and the welders.

 

[zbang]

Don't forget to look into the bonding of the pumps and the welders.

Yep, make sure the pumps & pipes aren't accidentally part of the welder's return lead.

 

[Jraef]

You didn't say what the damage was, bearing damage or winding insulation damage. Let's assume for a minute that it's bearing damage, because that HAS BEEN associated with welding operations.

So to elaborate on what the others are saying, it is not likely "spikes" as in large voltage transient events, but rather a series of small "noise" spikes caused by the welders that is creating havoc for you. If you do some research on "common mode noise" cross referenced to welders, you may see some info on what is referred to as EDM damage to motor bearings. EDM stands for "Electrical Discharge Machining", essentially what a welder is actually, because the damage that happens appears to the untrained eye to be something that looks as though a welder was at work. But how it relates to the motor bearings is that welders, like VFDs, can create high frequency common mode noise, which interacts with the stator and rotor in a motor to create a capacitive effect across the air gap, essentially a voltage potential that builds up between them and has to go somewhere. Since the motor is riding on bearings, that voltage dissipates across the bearing races through the bearings, and "pits" the races at a microscopic level as it passes by. Then because once there is a pit, the next time it builds up, that pit is an easier spot for the dissipation, it happens there again, and again, and again etc. etc. etc. The result is at first a "frosting" look to the bearing races, followed by a "washboard" appearance where there are deep ruts across the race surface. That is followed very shortly by failure of the bearings. VFDs are now a well known cause of this, hence the question, but long before VFDs were doing this, welders were doing it first. In fact when it first started showing up in motors run by VFDs, many people mistakenly assumed someone had been welding near by.

Generally, this can be fixed by a much more conscientious grounding regimen. The NEC rules on grounding are for protection of equipment and personnel from fire and accidents, NOT from common mode noise. If all someone did was the bare minimum per the NEC, it's really easy for common mode noise to sneak into a system from welders. That can also be the result of a good system gone bad over time, i.e. corroded terminations, cut or missing EGCs, etc. Start there.

If however the motor damage was insulation failure, that's a whole different kettle of fish. Might still be related to common mode noise, but it could be a lot of other things too. If you get a chance to see the damaged windings, compare them to what you see on this page and report back. A good motor shop will have access to this too, but not all motor shops are good motor shops...
http://www.rescuemotors.com/motor windings.html

 

[lmcgill]

Wow, thanks for all of the help. Let me provide some more details. First, the owner contacted their supplier of these motors and the supplier did have the pumps diagnosed but I cannot comment on the quality of this diagnosis. The motor shop said both of the motors "suffered a blowout in the windings" and that they were "tough to see and diagnose". They said "the windings showed no real signs of heat build up, the windings were bright, ho heat discoloration, and the ties strings were not brittle". It is the opinion of the motor shop that "the units had seen voltage spikes".

I would like to do two things, perform a more detailed inspection of the welding area to see what kind of grounding is there and put a monitor on the feed to the motors to see if we can catch any voltage irregularities.

Thanks for the info it was very helpful and insightful. I'll report back because I know you are all losing sleep over this one!!

 

[Gregg Harris]

Thanks for the info it was very helpful and insightful. I'll report back because I know you are all losing sleep over this one!!


Electricians do not sleep we stay wired.