High Amperage on ground possibly from UPS

[Bstir]

Hello,
Looking for some insight to a problem we are experiencing with high amperage reading on a ground wire. Lot going on here so bear with me, problem started yesterday with higher/low voltages coming into building 261V 287V 310V. While troubleshooting using a thermal imagining gun we found one branch circuit that had a ground heating up the hottest, 210 degrees, was at a UPS that feeds out to a CT scan. When we put an amp meter on the ground we had 160 amps, phases where around 250 amps, and only 4 amps on neutral. We had utility company come who confirmed high voltage was a problem from their substation and they corrected it. The amperage and the heat went down significantly but didn't go away currently values voltage 278 278 277, amperage 54 63 50, amperage on ground 41, amperage on neutral 4 amps. So when we started looking into the UPS possible being the issue we came across this see pics below. I have also attached the manual for the UPS.
Do you think this is where we are getting our imbalance from?
Answer to some questions you may have
Building transformer is Wye to Wye
CAT scan is straight 3 phase 480V that requires a Isolated ground same size as feeders and a redundant isolated ground #4
AT PANEL Isolated ground is bounded with bug to panel and redundant to ground bar
AT UPS Input isolated ground is not on grounding electrode but to the frame
AT UPS Input redundant ground is on output neutral bar but bonded to frame with that copper plate you see in pic
AT UPS Output isolated ground and redundant are also on Output neutral bar with Input redundant ground
Yes shutting down CAT scan would confirm that is where the imbalance is from but trying to find a time we can put it out of service is not easy

 

[ATSman]

Well for one thing, in the top left pic, if that is an isolated ground conductor it should not be landed on the equipment frame. Should be landed on the grounding electrode terminal, which in our opinion, should be insulated from the frame. The purpose of an isolated ground is to be separate from any equip frame or building steel. If not, the steel can act like an antenna and affect the critical control circuits of the UPS and its operation.
Any current flow on the ground wire is a no no. The ground wire should only carry current during a fault in order to trip the GFR protection device.
Can you send us the consulting engineer's (stamped) drawing for this installation, which is the first place I would start to resolve this problem..

 

[infinity]

While troubleshooting using a thermal imagining gun we found one branch circuit that had a ground heating up the hottest, 210 degrees, was at a UPS that feeds out to a CT scan. When we put an amp meter on the ground we had 160 amps, phases where around 250 amps, and only 4 amps on neutral.

Can you elaborate on what the "ground" conductor actually is? Is it an EGC run with the branch circuit?

 

[ATSman]

Based on the current readings, it appears the load current is returning on the grounding electrode instead of the neutral conductor.
Almost like the neutral is not grounded or not grounded in the right place.
The plan 3-line drawing would be a big help.

 

[synchro]

Do you have a current probe like a flexible Rogowski coil that you can place around groups of conductors? For example, around all the phases together, around all the phases with the neutral, and then with either or both ground conductors also included. In other words, try to find how the currents are flowing between the conductors by making more measurements, and then finding an explanation that fits the measured data.

Also, measure currents on the conductors going both into and out of the UPS.
A few other questions:
Does the CT Scan machine use a neutral?
Is there any other L-N load on the branch circuit before the UPS? I'm just asking because a possible scenario is that there is an open neutral in the panel for the branch circuit, and the 4A on the input neutral conductor is being fed "backwards" into this upstream L-N load from the N-G bond in the UPS. That could make the current on the neutral conductor relatively small and independent of the current drawn from the line conductors by the UPS and/or the the CT scan machine. But this is just speculation without more info.
Have you measured the currrents for the branch circuit in the panel, assuming your measurements have been made at the UPS itself?

 

[Bstir]

 

[Bstir]

Do you have a current probe like a flexible Rogowski coil that you can place around groups of conductors? For example, around all the phases together, around all the phases with the neutral, and then with either or both ground conductors also included. In other words, try to find how the currents are flowing between the conductors by making more measurements, and then finding an explanation that fits the measured data.

Also, measure currents on the conductors going both into and out of the UPS.
A few other questions:
Does the CT Scan machine use a neutral?
Is there any other L-N load on the branch circuit before the UPS? I'm just asking because a possible scenario is that there is an open neutral in the panel for the branch circuit, and the 4A on the input neutral conductor is being fed "backwards" into this upstream L-N load from the N-G bond in the UPS. That could make the current on the neutral conductor relatively small and independent of the current drawn from the line conductors by the UPS and/or the the CT scan machine. But this is just speculation without more info.
Have you measured the currrents for the branch circuit in the panel, assuming your measurements have been made at the UPS itself?

We do have a tester with flexible probes so I will take those measurements

CAT scan does not require a Neutral

UPS/CAT scan is fed from a panel that has Neutral feeders but no other Neutrals out. Panel feeds cooling equipment and out to a transformer that feeds a 120/208 panel this transformer is Delta Wye so they did not run a neutral to the transformer from 480 panel.

Thanks

 

[Bstir]

Well for one thing, in the top left pic, if that is an isolated ground conductor it should not be landed on the equipment frame. Should be landed on the grounding electrode terminal, which in our opinion, should be insulated from the frame. The purpose of an isolated ground is to be separate from any equip frame or building steel. If not, the steel can act like an antenna and affect the critical control circuits of the UPS and its operation.
Any current flow on the ground wire is a no no. The ground wire should only carry current during a fault in order to trip the GFR protection device.
Can you send us the consulting engineer's (stamped) drawing for this installation, which is the first place I would start to resolve this problem..

Yes I agree the input Isolated ground is not wired correctly. The input redundant isolated ground is going to the output neutral and bonded to the frame as well which also seems wrong. I believe this to be the problem that all the grounds serving the CAT scan should be isolated and neutral ground bond removed.