Current On The Equipment Grounding System

[Phil Devine]

We are experiencing problems with some of the electrical systems in one of our new schools. We wanted to make sure that the neutral conductors are only bonded in one place at the main service disconnect. We have 7 ? 15 amperes flowing in a main equipment grounding conductor bonding the switchgear to the building steel frame. There is a debate here whether that is normal or not and whether further testing of the ground and neutral connections and grounded conductor insulation system is necessary.

Is it normal to have current flow on the grounding conductor(s) of a system (like a school)?

Is there an acceptable current value on an equipment grounding conductor?

 

[lpelectric]

Sure does not sound normal to me. Stray current flow suggests a problem that needs some in-depth investigation. I'd look first for obvious neutral-to-ground faults. If that doesn't work, how about the primary current returning over the earth and getting into the ufer system and into piping bonded to the GEC.

 

[haskindm]

There should be no current flow on the grounding conductors. Where are you reading this current?
There may be an incorrect neutral/ground connection somewhere.
The other possibility is that it is from neutral current traveling through the earth that is finding a return path through your grounding system. The current may be from outside of your facility. You may need to get the power company involved to locate the source if that is the case. One way to test is to shut down all power to the school. If the current is still present on the grounding conductor, then the source is from outside of your facility. If the current goes away, then you have a good indication of improper ground/neutral connections in your facility.

 

[don_resqcapt19]

We need more details. There is always some current on the grounding electrode conductors as they provide a parallel path for grounded conductor current. In the case of buildings with a common metal underground water service, it is not uncommon to fine 20% of more of the grounded conductor current on the water line. So please provide more information about the grounding electrode(s) used. Also is there more than one service for this building?
Don

 

[Phil Devine]

There are several points in the equipment grounding system. As far as I know, there is only one main service disconnect.

There are several transformers and we verified (and changed a few) of the secondary connections to ensure the grounded conductor was only bonded at the transformer case or the service panel it feedd and not both.

The main equipment grounding electrodes are located outside of the building, and there is a gas line and several fire sprinkler risers also bonded to the main equipment grounding electrode. The conductor between the switchgear and the building frame is the one that has current flow.

To my knowledge, there are no other buildings connected or fed from this system.

The insulation resistance was tested on the main neutral conductor when separated from ground and it failed at 1Kv. It was checked for resistance and was over 1,000 ohms. We are told that the neutral insulation resistance is probably reading low because there are several devices connected to the neutral providing parallel paths to ground.

Our initial problems were light switches failing arcing to ground) and a possible shock of an employee. The inrush current of the electronic ballast is suspect of causing the switches to fail.

The current on the equipment grounding conductor is still present and we are wondering if that is a normal occurance.

 

[brian john]

Are there multiple utility transformers?

Is the service 480/277 VAC? If so do you measure ground current at the SDS's?

Is the generator neutral grounded at the generator, 3 pole or 4 pole ATS?

Metal water pipe as noted with adjacent buildings supplied from same transformer?

Have you meggered the panel feeders lifted/isolated from the main service?

Have you taken zero sequence measurments at all the distribution panels?

I have more but these are good for starters

 

[brian john]

Forgot to add a very common source of neutral ground current. The utility will ground the neutral, you electrical contractor does the same in the main service, if there is a metallic path between the utility transformer pad and the building you set up a parallel path with the neutral resulting in current. Metallic pipe, rebar (pad and building same foundation).

If there are any downstream neutral ground connections (NEC violation) you make this situation worse.

 

[Pierre C Belarge]

Our initial problems were light switches failing arcing to ground) and a possible shock of an employee. The inrush current of the electronic ballast is suspect of causing the switches to fail.

The current on the equipment grounding conductor is still present and we are wondering if that is a normal occurance.

As Don mentioned, it is not easy to troubleshoot a system via posts on a website, and the more information, the better the answers may be. There are some very experienced people who have already responded...you will get some good advice from them.

My response is in regards to the portion of your post that I highlight (and enlarged for our readers who may be colorblind).

This does not seem to me to be a grounding issue, but possibly a "bonding" issue somewhere within the building past the first point of disconnect - service.
It may also be fault current with an ineffective ground fault current path that is following steel to where you are getting your measurement.

If it is possible, I would try to shut down panels one at a time to see if the measurement goes away. If that does work, then you can narrow your search.
Of course this is just my thoughts.

Good luck and please keep us in the loop.

 

[dereckbc]

I am guessing here and just putting in my 2-cents worth.

This sounds like a large facility with a lot of wiring, electronic equipment, and lighting.

On a large system you have a fair amount of cable capacitance coupling between conductors, mainly between Phase conductors and EGC being in such close proximity in raceways. This will cause current to leak on grounding conductors. It is normal and nothing you can do about it.

In addition electronic equipment will have devices like filter capacitors for FCC purposes and MOV’s installed between L-G and N-G internal to the equipment. This device will leak current into EGC’s and add up. It is also possible one of the devices has failed shorted or partially shorted amplifying the problem. If memory serves me correctly HID lighting ballast have components (filters) installed between L-G

TVSS devices also leak, and fail shorted to ground.

So what I am saying is it might be normal, and the only way to trace it down is by divide and concur, or trial and error to find the source or sources.

 

[brian john]

dereckbc:

Truth be known it is most likely a combination of some of what I posted and your post. As I have stated before of the buildings we have tested 90% or more have neutral ground issues, of the 10% left they have service grounding that can not be avoided due to utility practices and the NEC, this results in ground current, then there is the leakage issue.

But IMO if proper testing was performed at installation the downstream grounded neutral issues could be avoided.

 

[MrHandy2005]

Only one main service disconnect??? I find it hard to bel;ieve that there is only one service to a school. There must be secondary boxes along the way> I would shut one off at a time to see if there is something wrong in that particular service box. Remember, neutral and earth ground are not bonded in secondary services.

 

[brian john]

Only one main service disconnect??? believe that there is only one service to a school.

Yes, quite common in all but the biggest schools in our area anyway.

I find it hard to believe There must be secondary boxes along the way.

Yep commonly called distribution panels.

I would shut one off at a time to see if there is something wrong in that particular service box.

Kinda of hard in a school to just SHUT OFF, better to perform a through investigation, then you have ammunition to take to management to arrange an outage, generally over a holiday weekend.

Remember, neutral and earth ground are not bonded in secondary services.

They durn site better be if you wish to pass inspection.

 

[Phil Devine]

The contractor replaced the lighting ballasts with those of a different manufaturer. The in rush current is about 1/2 of the inrush of the original ballasts. All efforts have been concentrated on the switch failures and the ballast/switch compatability. No answer or detrmination has been made regarding the gound currents. The neutral of both the primary and secondary winings of the step down transformers was found to be bonded. The primary bond was removed. The grounding conductor currents reduced from 15 amps to about 9 amps. The ground currents come form the secondary distribution panels. The feeder breakers were turned on one by one and the ground current increased with each feeder breaker turned on. We are still concerned that our ground current problem is not normal/may be hazardous. I drafted and forwarded the following questiosn from the testing companies report:



Is 9 amperes a safe and acceptable value of current on the equipment grounding conductor bonding the ground system to the building steel?


What is the minimum acceptable insulation resistance between the grounded neutral conductor and the equipment grounding conductor? Is the value of 980 ohms acceptable?


What could be benefited from perfoming the following tests:

ground loop impedance test
clamp-on ground test of distribution and panel grounds and other accessible points?

I know this is hard to do by forum however, we think the source of currents on the grounding conductors warrants investigation. We are not convinced that the problem is solved because we installed new ballasts to the light fixtures.

 

[brian john]

. The neutral of both the primary and secondary windings of the step down transformers was found to be bonded.

Primary Neutral? You are utilizing wye wye at 480/277 to 208/120. This seems to be to be screaming RED FLAG to me I would check this issue.

T he ground currents come form the secondary distribution panels. The feeder breakers were turned on one by one and the ground current increased with each feeder breaker turned on. We are still concerned that our ground current problem is not normal/may be hazardous.

This could be from branch circuits from one SDS tied to another SDS or accidental/intentionally ground neutrals downstream from the SDS. SDS (separately derived system ie. Transformers)

Is 9 amperes a safe and acceptable value of current on the equipment grounding conductor bonding the ground system to the building steel?

Could be depends on the source of this current. It is very hard to measure all the ground current until you have resolved all NEC violations (if they exist)

What is the minimum acceptable insulation resistance between the grounded neutral conductor and the equipment grounding conductor? Is the value of 980 ohms acceptable?

Depends on what was connected to the neutral at the time, I would isolate all neutral feeders to megger them, then you would expect readings in the megohms, same with the branch circuits.

What could be benefited from performing the following tests:

ground loop impedance test

clamp-on ground test of distribution and panel grounds and other accessible points?

In searching for the source of ground currents in my experience little help I have outlined above the procedures I have utilized to locate and isolate ground current. And while I am not the sharpest knife in the draw, I do have years of expierence doing this particular type of investigation.

I know this is hard to do by forum however, we think the source of currents on the grounding conductors warrants investigation. We are not convinced that the problem is solved because we installed new ballasts to the light fixtures.

Neither am I if you can still measure ground current, whether it is an issue or not, can be determined with a few test.

 

[Pierre C Belarge]

"Quote:
I know this is hard to do by forum however, we think the source of currents on the grounding conductors warrants investigation. We are not convinced that the problem is solved because we installed new ballasts to the light fixtures. "


I do not believe changing the ballasts was the only cause of your issues. This is a school, and I cannot imagine that 9 amps of current on the grounding conductors is acceptable.
You and your team should spend as much time as is necessary to locate this problem.


If that is not acceptable, I have seen where Brian John has in the past offered his services to people out of his normal work territory. Maybe he can help you.

Brian - hope you do not mind, but I think this issue needs an expert such as yourself to locate a potentially very dangerous situation in a school.

 

[Phil Devine]

The transformers are delta-wye that the bonding jumpers were removed from. Sorry for the typo. We are awaiting the design engineer's resonse to the test report to determine where to go from here. I appreciate everyone's input and advice. I will keep you posted as this unfolds.

 

[jim dungar]

The transformers are delta-wye that the bonding jumpers were removed from. Sorry for the typo. We are awaiting the design engineer's resonse to the test report to determine where to go from here. I appreciate everyone's input and advice. I will keep you posted as this unfolds.

There needs to be a bonding jumper from the core assembly to the enclosure and from the enclosure the building ground reference just like any piece of equipment.

There needs to be a bonding jumper, but only one, between X0 and the GEC. This jumper can be located in the transformer or in the first disconnect after the transformer.

 

[Phil Devine]

Switch Failure From Ballast Inrush Current

Switch Failure From Ballast Inrush Current

We are concerned about the ground currents and awaiting the design engineer's analysis.

The contractor has focused on the ballast and switch compatibility with regards to the in-rush current. Has anyone had any exderience or switch failures/noncompatibility from flourescent light inrush current? We have six flourescent fixtures controled by each switch. The switches have failed and arced to the metal switch plate. The contacts of the switches when diesected showed much pitting even afeter short use. We repalced 15 ampere switches with 20 ampere switches from the same manufacturer and got the same results of contact pitting and audible popping and snapping inside the switch when operated.

One of our electrical suppliers had recommended use of thermistors as inrush current limiters.

 

[Phil Devine]

We Had The Design Engineer And The Testing Firm At The School Yesterday. We Found 9 Amperes Still On The Equipment Bonding Jumper Connecting The Building Steel To The Main Equipment Ground Buss. There Was Minimal Currents On The Conductors Connecting The Multiple Water Lines, Fire Sprinkler Risers And The Equipment Grounding Electrodes To The Main Grounding Buss. We First Connected The Main Equipment Ground Buss To The Utility Transformer Case With A Jumper And Had No Change In Any Readings. Further Trouble Shooting Indicated That The Currents Were Coming From The Low Voltage Transformers. We Found That As We Opened Feeder Breakers To The Sub Distribution Panels, The Current On The Conductor Bonding The Building Steel To The Main Equipment Ground Buss Went Down Proportionately. We Removed The Bond To The Building Steel And There Was No Change Or Increase In Any Of The Other Bonding Conductor Currents That Connected To The Piping Systems. There Is Very Little Building Steel And It Is Isolated From The Rest Of The Building/steel. The Engineer Had The Building Steel Directly Bonded To The Piping Systems Instead Of The Main Equipment Ground Buss. It Seems We Had A Return Path For Neutral Current Through The Bonding Conductor To The Steel. The Steel Had Multiple Coinnections To The Water System And The Neutral Through Pipe Hangers Etc. Pretty Interesting And Sharp Analysis From The Engineer. Thought I Would Share That With You All.

We Also Found 4 Circuits In The Sub Distribution Panels That Have Bonds Between Gound And The Neutrals With The Conductors Isolated That Read Good With A Multimeter But A Short When Tested With A 500 Volt Megger. We Have Some More Troubleshooting To Do. Sorry For The Caps Lock, I Am Too Lazy To Retype.

 

[zog]

"Read Good With A Multimeter But A Short When Tested With A 500 Volt Megger"

thats why you use the right tool for the right job