Parallel conductors in EMT (burned Grounds in a Cap bank)

Status
Not open for further replies.

aelectricalman

Senior Member
Location
KY
We specified a capacitor bank on a 480V Delta with a floating ground system. There were supposed to be 4 conduits, each with a set of three 500 kcmil ungrounded feeders (a,b,c) and one 1/0 ground conductor per the design, for the 1600 amp bus that it was feeding. Also note that the bus fed two 800 Amp breakers each set to trip at 700 amps (so wire sizing is ok). The bank was placed immediately on the load side of the gear at the customers main disconnect.

The contractor that did this installation was called back by the owner and was told that during a routine inspection they found all the ground wires from the cap bank had melted insulation (not the conductors). Also, the breaker in the capacitor bank was tripped on one side suggesting an issue. All of the other ungrounded conductors in the bank had not heated up and no fuses were blown. All 50 capacitors and both reactors were resistance tested to ground and frame and were ok. There was no sign the bank was the problem. Upon further inspection, it appears there are two things that were wrong with the installation.

1. The contractor used EMT conduit as a raceway along with a 1/0 conductor for each ground. All knockouts were concentric so the conduits had solid connection to the enclosure. The contractor did not bond the raceways at a bushing, so there was potentially a ground loop (don't know, cannot verify without further tests).

2. The contractor (instead of sticking to the drawings) ran 4 - 4 inch conduits and used only 3 sets of conduits. Inside each set of conduits there was 4 parallel conductors for each phase (yes all four of each conductor is in their each respective conduit) along with 1 - 1/0 ground conductors. Conduit 1 - 4 parallel A phase + 1/0 ground........Conduit 2 - 4 parallel B Phase + 1/0 ground........ Conduit 3 - 4 parallel C phase + 1/0 ground......Conduit 4 is empty. This was not approved like this and I am not sure why it was done this way. I still have to look at the derating but Im not even sure if a 4 inch is acceptable ( but as I said, there is not ungrounded conductor issue).

My question is this....As much as #1 sounds like it is a viable reason for the excess currents, is there any situation where #2 could cause this issue of burned ground conductors? Thanks for your help in advance everyone..
 
Last edited:

Phil Corso

Senior Member
Chris...

1) What is the kVA capacity of the bank in kVA, or Amperes, or Micro-farads?

2) What was the EMT, size, and spacing? Material; magnetic or non-magnetic?

3) What is bonding arrangement, i.e., were the conduits bonded together at both ends?

4) What is conductor insulation?

5) Can you provide some detail about insulation of "all ground-wires from cap" were burned! Meaning, the individual wires from each can?

6) Are the Bleed-Resistors intact?

Regards, Phil Corso
 

Smart $

Esteemed Member
Location
Ohio
I'm inclined to believe the lack of bonding bushings may have had some impact, but only a small, possibly even insignificant amount. The general cause was the iso-phase arrangement. The grounding conductors, paralleled but through separate conduits created a ground loop, with the emf of the iso-phasing induced current through that loop with little resistance. The heat from the grounding conductors and hysteresis between conduit terminations raised the enclosure temperature enough to shift the breaker trip curve to the point of tripping.

Just speculating based on known electrical phenomenon.... :angel:
 

iceworm

Curmudgeon still using printed IEEE Color Books
Location
North of the 65 parallel
Occupation
EE (Field - as little design as possible)
... 1. The contractor used EMT conduit as a raceway along with a 1/0 conductor for each ground. All knockouts were concentric so the conduits had solid connection to the enclosure. The contractor did not bond the raceways at a bushing, so there was potentially a ground loop (don't know, cannot verify without further tests).

2. The contractor (instead of sticking to the drawings) ran 4 - 4 inch conduits and used only 3 sets of conduits. Inside each set of conduits there was 4 parallel conductors for each phase (yes all four of each conductor is in their each respective conduit) along with 1 - 1/0 ground conductors. Conduit 1 - 4 parallel A phase + 1/0 ground........Conduit 2 - 4 parallel B Phase + 1/0 ground........ Conduit 3 - 4 parallel C phase + 1/0 ground......Conduit 4 is empty. This was not approved like this and I am not sure why it was done this way. I still have to look at the derating but Im not even sure if a 4 inch is acceptable ( but as I said, there is not ungrounded conductor issue).

My question is this....As much as #1 sounds like it is a viable reason for the excess currents, is there any situation where #2 could cause this issue of burned ground conductors? ..

This is an absolute classic for #2 causing the problem.

With the phases isolated in a magnetic conduit, there is no way for the magnetic fields to cancel.

Tell the contractor he is pulling the wire out and installing per the drawings. And he is scrapping the existing wire and using new. Which likely he will tell you to pound sand - and you are paying for the job again.

edit fat fingered the send button

edit to add: I think this is what $ is saying.


ice
 
Last edited:

kwired

Electron manager
Location
NE Nebraska
This installation was in violation of 300.3(B). This is an example of what can happen when not following the requirements of that section.


Having only one conductor of a circuit in a metal closure (a raceway in your case) creates a one turn transformer of sorts. You likely had current in the metal raceway as well but apparently had high enough current in the EGC's to overheat the insulation on them.

Putting all conductors of the circuit in the raceway, or same number conductors from every phase in one raceway where there are paralleled sets means the magnetic fields around those conductors cancel one another out and doesn't cause magnetic effects on the closure (raceway in your case).

Now bond three different raceways together on the ends that are not done correctly and you have three (out of phase from one another) sources and circulating currents between them all. Kind of like creating a low voltage secondary three phase transformer and shorting the secondary conductors together in a way.
 

jim dungar

Moderator
Staff member
Location
Wisconsin
Occupation
PE (Retired) - Power Systems
#2 is definitely a problem, and needs to be the first item fixed.

EMT and lack of bonding bushings is a basically a non-issue.

You said this is a 480V ungrounded system, are your cap banks connected in a grounded wye?
 

Sierrasparky

Senior Member
Location
USA
Occupation
Electrician ,contractor
Looks to me like the only solution is to remove and replace the conductors with new as per plan and standard practice. Was this ec a newbie or what. :?
 

aelectricalman

Senior Member
Location
KY
Chris...

1) What is the kVA capacity of the bank in kVA, or Amperes, or Micro-farads? 800 kVAR, 1600 Amp. I dont know what the KVA is off hand until I get to work tomorrow but I do know the bank was sized appropriately at 95%

2) What was the EMT, size, and spacing? Material; magnetic or non-magnetic? 4in, 4in. cooper, non ferrous.

3) What is bonding arrangement, i.e., were the conduits bonded together at both ends? Stated above. No bonding.

4) What is conductor insulation? THHN/THWN-2

5) Can you provide some detail about insulation of "all ground-wires from cap" were burned! Meaning, the individual wires from each can? The conductors from the capacitors to the gear. I stated the capacitors and bank was unharmed...Only the conductors installed by electrician were damaged.

6) Are the Bleed-Resistors intact? yes. No damage to the unit. Caps all tested appropriate uF

Regards, Phil Corso

See above, in the quotes.

Regards, Chris Hill
 

aelectricalman

Senior Member
Location
KY
Why did you reply to the one post that had the least to do with diagnosing or fixing the problem?

Because he is the only one that asked questions (and yes, I agree it has nothing to do with the problem). I have not had a second to review the responses but I will in a few minutes.

Secondly, my browser had not refreshed to show the other responses and since he was first, I saw it without seeing the others.

Sorry for the inconvenience.
 

aelectricalman

Senior Member
Location
KY
I'm inclined to believe the lack of bonding bushings may have had some impact, but only a small, possibly even insignificant amount. The general cause was the iso-phase arrangement. The grounding conductors, paralleled but through separate conduits created a ground loop, with the emf of the iso-phasing induced current through that loop with little resistance. The heat from the grounding conductors and hysteresis between conduit terminations raised the enclosure temperature enough to shift the breaker trip curve to the point of tripping.

Just speculating based on known electrical phenomenon.... :angel:

Thanks for the insight. That is what I suspected. I just didn't know if the iso phasing was a smoking gun. I have little experience with the effects of this. Although I understand the engineering behind what is occuring, I didnt know the potential full range of consequences of the situation. Thank you again.
 

aelectricalman

Senior Member
Location
KY
This installation was in violation of 300.3(B). This is an example of what can happen when not following the requirements of that section.


Having only one conductor of a circuit in a metal closure (a raceway in your case) creates a one turn transformer of sorts. You likely had current in the metal raceway as well but apparently had high enough current in the EGC's to overheat the insulation on them.

Putting all conductors of the circuit in the raceway, or same number conductors from every phase in one raceway where there are paralleled sets means the magnetic fields around those conductors cancel one another out and doesn't cause magnetic effects on the closure (raceway in your case).

Now bond three different raceways together on the ends that are not done correctly and you have three (out of phase from one another) sources and circulating currents between them all. Kind of like creating a low voltage secondary three phase transformer and shorting the secondary conductors together in a way.

Thanks for the NEC citation. That is very helpful..
 

aelectricalman

Senior Member
Location
KY
Looks to me like the only solution is to remove and replace the conductors with new as per plan and standard practice. Was this ec a newbie or what. :?

The sad thing is this was an EC that I know and trust. He has about 20 years experience with commercial/Industrial. It was a lapse in judgement I suppose. He probably walked away and let his journeyman do this....Then when he came back the day was shot and he went with it....(speculation). Either way, I would have thought he knew better.

We work for them on this job so it won't be coming out of my pocket. Ha Ha.
 

aelectricalman

Senior Member
Location
KY
There are a lot of good responses here. Thank you all for the sanity check. I will update everyone when we have resolution.
 

Sierrasparky

Senior Member
Location
USA
Occupation
Electrician ,contractor
The sad thing is this was an EC that I know and trust. He has about 20 years experience with commercial/Industrial. It was a lapse in judgement I suppose. He probably walked away and let his journeyman do this....Then when he came back the day was shot and he went with it....(speculation). Either way, I would have thought he knew better.

We work for them on this job so it won't be coming out of my pocket. Ha Ha.


20 year veteran of the electrical field!, well I hope it was one of his workers that did it and not himself. ( really does not change much other than some piece of mind. )

I suppose neither will do that again. Boy the lessons we learn sometimes. I never stop learning myself.
 
Status
Not open for further replies.
Top