Lighting Breaker Tripping

[chrisrappl]

I could use some help troubleshooting an intermittently tripping lighting feeder breaker.
The service is 277/480 3 phase/ 4 wire ? 2500 amp main breaker with ground fault protection (GFP).
Downstream are two switchboard sections containing GE TJL breakers. The lighting breaker (Lighting panels on the 1st and 2nd floor) and the breakers for the 1st and 2nd Fl power panels are equipped with GFP; the breakers for the UPS and EHVAC are not equipped with GFP.
Since the main has GFP, I do not think that the code requires GFP on the lighting and power feeder breakers; however the original engineer (20+ years ago) must have had his reasons.
The lighting feeder breaker (TJL4V2603) with micro versa trip (T4VTG) is a 300a breaker that normally is loaded to about 10% (A-33.9a, B-27.1a, C-19.4a). It had tripped previously (maybe once every few years). In the past month or so it has tripped almost every week. (No changes or additions that we are aware of ? the lighting panels also contain some single phase 277v circuits for VAVs - a few years ago an underground backflow preventer was replaced with an above ground unit that may have a heater for freeze protection. I need to investigate since the pole lighting is the only power in the vicinity.) I don?t think that these GE breakers are known to nuisance trip when lightly loaded and unbalanced, but it?s not a good situation.
12/2/10 7:40pm
12/10/10 7:39pm
12/16/10 6:30pm
12/21/10 7:43pm
12/24/10 6:04pm
Since all the tripping has occurred after dark I looked into the five exterior lighting circuits (photo cell controlled through contactors). We made sure that all the pole lights, bollards and exterior building lights were in good operable condition and then operated with the two pole light circuits off (breaker tripped), with the two bollard circuits off (breaker tripped) and with the building exterior lighting circuit off ( breaker tripped). After the 3rd or 4th trip, the building technician noted that each time, as he was heading over to the building to reset the breaker, he was passing the night cleaning lady as she was leaving the building. We have spoken to the cleaning lady and are in the process of reviewing the security cameras for clues. (Maybe those are her nights for cleaning the breakers?
On the 21st, the lighting breaker and the breaker directly beneath it (power ? 2nd fl.) were tripped. The 2nd floor power feeder is a 3 wire feeder with no neutral however; there is a pair of control wires coming to the breaker from an unused neutral CT. Interestingly the micro versa trip on the power breaker is a T4VTNG. I would assume that the ?N? designates neutral. As far as the 2nd floor power breaker being tripped, I suspect human error. The tech who responded did not know how to reset the lighting breaker and he may have accidentally pushed the trip button on the breaker below while trying to reset the lighting breaker, but? It may be a clue.
The electrical room where the breakers are located also contains a large UPS. The original UPS had battery racks of large wet cells that overcharged a couple of years ago and more than half of the electrolyte boiled out into the atmosphere of the room. Acid swab tests on the walls and equipment were unremarkable. Mostly the water evaporated leaving a higher acid concentration in the cells, but? It may also be a clue. Solid state boards in the breakers could have been affected.
I?m sure that I need to continue looking for a downstream ground fault; however, I think that I should schedule a shutdown and trace the CT control wires to each breaker, making sure that the lighting neutral CT is connected to the lighting breaker. I also need to know if the lighting breaker should have the T4VTNG micro versa trip unit. The T4VTG units do have two control terminals and each has a pair of wires coming up form their respective, unused neutral CTs, so they must be designed for neutral current, but why the difference of the ?N? in the designation? And why isn?t the one feeder with a neutral connected to the trip unit with an ?N??
I wish I knew why the original engineer required GFP on the lighting and power panel feeders. I am almost tempted to replace the T4VTG trip unit with a T4VT unit with no GFP, or just replace the TJL breaker with a basic TJK frame breaker with no GFP. However, I would not want to take a chance that whatever is causing the lighting breaker to trip could cause the main breaker to trip since it also has ground fault protection.

 

[Dennis Alwon]

I will preface this by saying I know nothing of GFP protection. It does sound like the cleaning lady may have something to do with it. Is it possible???? that a vacuum or some appliance she uses can cause this? Just a thought. I know it is not the same as GFCI but not sure-- Is the GFP adjustable?

 

[zog]

Before you go changing the design of the protective system and/or replacing breakers you may want to do some actual testing. Have you tested the breakers? Have you done any insulation resistance testing? Those are the 1st 2 things that need to be done to find out what is happening here.

 

[chrisrappl]

Thanks for the replies. I don't have a way of testing the breakers. I would consider insulation testing, if I had a clue where to start. Since this is so intermittant, I think that it probably is an equipment fault rather than cable insulation failure. I did asked to review the security cameras to see if a truck or something was happening in the parking lot, but since the breaker continued to trip with those circuits de-energized, I'm not convinced that it is the exterior lighting. I was thinking that if I temporarily eliminated the GFP, I might force the affected branch circuit overcurrent device to trip and that would show me where to start troubleshooting.

As far as the cleaning lady connection - I don't see how plugging in a vacuum cleaner in a 120v receptacle could affect a breaker feeding two 277/480v lighting panels. It might have more to do with what she is doing with lighting or even temperature since there are some 277v VAVs fed from these lighting panels.

 

[LYLE2231]

just thinking out loud here...Have you checked the tightness at the breaker lugs?
Then I'd move down stream to the sw bd that feeds the poles.
Hope this helps. Lyle

 

[Dennis Alwon]

As far as the cleaning lady connection - I don't see how plugging in a vacuum cleaner in a 120v receptacle could affect a breaker feeding two 277/480v lighting panels. It might have more to do with what she is doing with lighting or even temperature since there are some 277v VAVs fed from these lighting panels.

Yeah, I forgot it was 277V... dopey me.

 

[LarryFine]

I don't have a way of testing the breakers.

Substitution, especially when you have more than one of the same type of device. Swap two and see if the issue stays with the wiring or moves with the device.

 

[zog]

Substitution, especially when you have more than one of the same type of device. Swap two and see if the issue stays with the wiring or moves with the device.

So if he replaces a tripping breaker and the replacement does not trip is the breaker not tripping bad or was the one that was tripping bad? How do you know without testing? A breaker is supposed to trip when a fault occurs and most breaker failures will cause them not to trip.

 

[defears]

Rent a recording meter. See which phase is causing the problem and then record each breaker on that phase until you find the circuit.

 

[zog]

Rent a recording meter. See which phase is causing the problem and then record each breaker on that phase until you find the circuit.

Good advice

 

[chrisrappl]

Good Idea. I have a Dranitz Disturbance Analyzer. I forgot that it was up in the shop. I'll hook it up and see if it can capture anything. If it does show an anomaly on one phase, at least I've narrowed my troubleshooting to 1/3 of the circuits involved.

Thanks

 

[chrisrappl]

Update

Update

The breaker tripped again on 12/31/10 @ 6:00 pm and on 1/4/11 @ 11:47 pm. The only constant so far is that it always trips when the building is unoccupied. Remember that this breaker is only loaded to about 10% when the building is fully occupied.
At this point I am thinking that the reduced load when the building is unoccupied might be a factor.
Maybe the GFP of the breaker is looking at imbalance as a percentage and nuisance tripping when the load is the lightest and possibly most unbalanced (percentage wise)?
Early next week I will hook up the recording meter and see if I can capture anything.
I have scheduled a shutdown for 1/15/11 and hope to have a replacement breaker by then.
During the shutdown I will do an insulation test of the feeder conductors, trace the control wires to the neutral CTs, and make sure that the neutral conductor is connected to the proper CT.
Can anyone tell me the difference between GE Micro-Versa-Trip units T4VTG and T4VTNG?
I need to determine which would be the proper unit for a 3 phase / 4 wire feeder serving single phase 277v lighting and VAV branch circuits.

 

[zog]

The breaker tripped again on 12/31/10 @ 6:00 pm and on 1/4/11 @ 11:47 pm. The only constant so far is that it always trips when the building is unoccupied. Remember that this breaker is only loaded to about 10% when the building is fully occupied.
At this point I am thinking that the reduced load when the building is unoccupied might be a factor.
Maybe the GFP of the breaker is looking at imbalance as a percentage and nuisance tripping when the load is the lightest and possibly most unbalanced (percentage wise)?

Possible, the accuracy of those CT greatly dimishes below 20% load.

Early next week I will hook up the recording meter and see if I can capture anything.
I have scheduled a shutdown for 1/15/11 and hope to have a replacement breaker by then.

I should have those in stock new and reconditioned if you need them.

Can anyone tell me the difference between GE Micro-Versa-Trip units T4VTG and T4VTNG?.

The N means it is nuetral sensing and uses an (external) 4th CT

 

[chrisrappl]

Zog,

How can I contact you?

Chris

 

[zog]

OK, at <20% loading the CT's will not even power up that trip unit, at which time anything can happen, a small spike on a single phase from turning on oooh lets say a vacuum cleaner could cause the trip unit to think there is a ground fault and trip.

Another possible issue, the TJL breakers have a rubber bumper in the contact assembly that often melts and gums up the mechanism (A Reason they redesigned this line) and can cause the breaker to trip, not trip, not close, all of the above. Like pouring rubber cement in the mech.