nuisance GFCI breaker Tripping

Status
Not open for further replies.
C

Cestes

Member
I was recently contacted as a third party testing agency to check a school ground system because of nuisance GFCI breaker tripping (several different breakers in different panels and some even during the evening hours. The design load and tested circuit amperage is less than 6 amperes per 20 amp circuit breaker. Everyone wants to assume that the obvious reasons are not the fault. Items like not having crossed neutrals or grounds on the 3? circuits i.e. 1,3,5, & 2,4,6 have been checked out.

So here are the basics. The school is new... 1200 ampere 480/277v 3? 4W service with the following grounds: water main, ground rod triode, bldg. steel, perimeter 3/0 ground ring, rebar.... the 480 volt gear feeds a 300KVA 50% K-rated 480-208/120volt transformer bonded to the bldg steel with a 3/0 bare copper GEC as well as bonded with bonding jumpers at the ground bushings at the flex fittings. The only grounding of the 1200 ampere 208/120 3? 4W switchboard is a 3/0 CU GEC through each of three secondary flex conduits and mechanically bonding to the transformer chassis (thus mechanically bonding to the building steel). Bldg limitations precluded me from doing a fall of potential test with test points in clean soil. So this left me doing a 2 point earth resistance test of all 5 of the GEC's and a 2 point test between extraneous conductive parts. This test was done on both the 480/277 and 208/120v systems. The test results yielded nothing greater than .01 ohms on each grounded electrode system. The odd thing is a large amount of noise as indicated by a megger model DET 20C. I couldn't get a reading on the ground voltage yet I assume it had to be more than 5 volts to trigger the noise alarm on the meter. Any suggestions would be appreciated.... Thank you
 
So do you have GFCi breakers tripping or the GFI on the 1200A service tripping?

"Bldg limitations precluded me from doing a fall of potential test with test points in clean soil." Building limitations or your test equipment limitations?
 
GFI tripping....

GFI tripping....

it is GFCI branch breakers tripping..... as for limitations on fall of potential. I use a Amprobe DM-III Multitest that will do a Fall of Potential test. The building service is located in a mechanical room located in in the center of a building 100' + leads would be needed for testing. Possible but not feasible... thank you for any suggestions or question you might have.
 
IN either case Building or test equipment limitations I doubt the resistance of the soil is impacting GFP or GFCI's in the building.


Assuming you have checked everything THOROUGHLY as you state.
I would test the GFCI and Over-current portion of the CB's with calibrated testers.
If all test acceptable.
I woulds megger the branch circuits.
Monitor one circuit (high speed disturbance analyzer) and see whats happening when it trips.
 
GFCI tripping

GFCI tripping

O.K. here's what I was told in a pre-test meeting. I brought up the following possibilities and was told that all have been ruled out. also again these breakers are tripping randomly. There are 12 to 16 GFI breakers in each of 4 panels and the same breakers don't trip on a consistent routine. sometimes 1 or 2 in one panel sometimes one or two in each of the 4 panels. Sometimes none trip. sometimes they trip during the day, sometimes they trip during the night. The project engineers and electrical contractor are to the point of grasping at straws. These are what I've been told have been ruled out. Also siemens has replace approximately 10 of the breakers thinking that it might have been a bad batch. Thanks!!

1) The use of after hours cleaning crew using a powered buffer or floor sweeper.
2) An occurrence of a shared neutral between 3? circuits on a GFCI circuit.
3) An occurrence of a shared ground between 3? circuits on a GFCI circuit.
4) A pinched ground or neutral conductor the conduit system causing a ground loop.
 
Last edited:
Here is what I would do--- I would take 2 or 3 of these troublesome GFI breakers and replace them with standard breakers. Then I would put a GfCI Receptacle in line to give you the protection needed for the circuit. Check and see if this trips also.

Don't ask me why because I can't explain it but I had a similar situation years ago with 2 Gfi breakers tripping erratically. I put the receptacles in line and never had a problem. I was at the house just yesterday and the GFI receptacles are still working.
 
Have they ever not experienced tripping problems?

I had one school that suddenly started to experience problems. It turns out that a ballast was replaced in an outdoor wallpak that was not sealed between the fixture and the building. When conditions were wrong, condensation would form in the fixture and trip the GFCI when the photocell turned on.
 
I do not have anything to suggest other than what has already been suggested. I would like to make one statement. I think electrical professionals should stop using the term "Nuisance Tripping" as there is no such thing. GFCI's will trip either because there is a ground fault present beyond what they will tolerate or they are defective. There is no other possibility. There is no nuisance tripping. There is either a wiring problem, defective equipment, or defective GFCI devices. We electricians do not want to admit that our wiring may be at fault, and manufacturers do not want to admit that their equipment or breakers may be at fault so the term nuisance tripping has been invented. It is a false concept. If GFCI's are tripping there is a problem that must be located and corrected. "Nuisance tripping" is an excuse to do nothing.
 
haskindm.....

haskindm.....

I agree with your conclusion on the use of "nuisance". What makes this a nuisance is that everyone involved in this project swears it's not their fault that 8 to 10 GFCI breakers trip on a daily basis. some when not in use. So based on this observation I choose nuisance because why would a GFCI breaker react to a "assumed" fault today and not tomorrow, this morning and not tonight..... Can ground noise cause GFCI tripping? Can a 480 volt VFD for the school chiller inject enough noise on the ground system to cause an impact on GFCI breakers in branch panelboards? (suggested by electrical engineer).... If we make the assumption that everything normal is correct...... collectively what else in a building can cause a random tripping of GFCI breakers... Again, Thanks!!
 
Cestes said:
I agree with your conclusion on the use of "nuisance". What makes this a nuisance is that everyone involved in this project swears it's not their fault that 8 to 10 GFCI breakers trip on a daily basis. some when not in use. So based on this observation I choose nuisance because why would a GFCI breaker react to a "assumed" fault today and not tomorrow, this morning and not tonight..... Can ground noise cause GFCI tripping? Can a 480 volt VFD for the school chiller inject enough noise on the ground system to cause an impact on GFCI breakers in branch panelboards? (suggested by electrical engineer).... If we make the assumption that everything normal is correct...... collectively what else in a building can cause a random tripping of GFCI breakers... Again, Thanks!!
Click to expand...

A GFCI does not care anything about ground at all. It only cares if the current flowing out of the device is equal to the current flowing into it. So, unless your grounds and neutrals are bonded downstream of the GFCI, currents on the ground can not flow "backwards" into the GFCI.
 
Try using a high speed scope:
H-N
N-G
H-G

Make sure all the school electrical equipment is running.

We have had problems with VFD's interference etc. and breaker trips.

Good Luck
 
I have used the term "nuisance tripping" many times. I also believe that it is possible for GFIs, GFCIs or RCCDs (Residual Current Circuit Breaker) to trip when you would not normally expect them to.
Thus the term nuisance. I do not consider the use of that term to be unprofessional. Troubleshooting such situations can be a nuisance.

In some instances "leakage current" due to EMI filters (etc) may be putting a GF(C)Is on the edge of tripping and when noisy loads are introduced they then trip.
Different GFIs may have different reaction times and be more or less susceptible to transient or noise currents.
We supplied equipment to a Semiconductor manufacturer that required an RCCD on the main power input to the equipment (208 V @ 200A -3Ph). They often nuisance tripped.
This equipment incorporated several VFDs and other electrically noisy devices.

Because this nuisance tripping caused expensive downtime the company spent the time to have several different brands of RCCDs EMI tested and then began specifying that only certain brands that passed the EMI testing could be used.
In one instance we simply substituted the EMI tested brand for another untested unit and the nuisance tripping stopped.
 
" I think electrical professionals should stop using the term "Nuisance Tripping" as there is no such thing. GFCI's will trip either because there is a ground fault present beyond what they will tolerate or they are defective. There is no other possibility. There is no nuisance tripping. "

Amen to that brother!

" The building service is located in a mechanical room located in in the center of a building 100' + leads would be needed for testing."

Cestes, that is pretty normal, most fall of potential tests require the Z rod to be >100 Ft from the rod under test to be outside the sphere of influence of the other 2 rods. Our sets have 400ft leads, sometimes even those are not long enough.
 
Cestes,

In post #4 one of the items ruled out is,' an occurance of a shared ground

on a three phase circuit' . I don't see where this would make any difference!

My point being, if the troubleshooters are wasting their time on the grounds

of a three phase circuit, maybe it's time to get some new troubleshooters.
 
How long are the branch circuits?

I second Frank's skepticism of the troubleshooter's abilities if the equipment grounding conductors (EGCs) were on the list. These conductors should naturally be joined several times along the way, in a commercial setting.

If the wiring method is EMT, and they were suspended with metallic hangers, then there are multiple electrical paths for a ground fault to return to the SDS or service, in addition to the intended ones.

Given this skepticism, I would tear into the circuits myself to verify that the neutrals were indeed not mixed together among multiwire circuits (MWBCs), and that the neutrals were not touching the EGCs.

I also agree with Jim Dungar's comments about the irrelevance of the grounding electrode system to the problem.

I haven't seen multiwire circuits used on GFCI breakers, myself, I would think that would be cost-prohibitive. You might simply verify that these are not single-pole GFCI breakers supplying MWBCs, that would not work. The breakers would have to be three-pole, 120/208 rated GFCI breakers.
 
Cestes, that is pretty normal, most fall of potential tests require the Z rod to be >100 Ft from the rod under test to be outside the sphere of influence of the other 2 rods. Our sets have 400ft leads, sometimes even those are not long enough.
Click to expand...

Depending on the size of the electrode grid will determine the distance you need to "go out" to place test electrodes. We have had specs requiring 1000'+. The issue here becomes neighbors, not test leads.
 
Thank you all.....

Thank you all.....

Gentlemen,

Thank you one and all for your consice information. I do not claim Juris (electrical) Prudence on all matters. I'll be the first to admit this one has me stumped because I have to prove something wrong based on someone elses assumptions and declarations..... i.e. "we've checked that and ruled it out" So thank each of you..... Keep sending any ideas and thoughts you might have....

Chuck Estes
 
Status
Not open for further replies.
Top