GF nuisance tripping-480v system-troubleshooting

[tommom]

I have a 480 volt switchboard with a 1600 amp main circuit breaker with a GE type SS circuit breaker with ground fault trip, and feeder breakers of GE type SGH, SGL, SFH downstream of the main. These breakers feed lighting panels and rooftop units.

Some of the feeder breakers and the main are experiencing nuisance trips (I think I was told that it was at least some of the rooftop feeders). The nuisance trips have subsided since the GE rep came out and cranked the trip settings up, I think.

I figured the best way to troubleshoot is to turn on all the rooftops and lights/loads I can find, and read the current in the main switchboard from neutral to ground. If there is more than "leakage" type current reading there, I could then 1) shut down hvac units 2)shut down lights/loads 3)trip breakers if necessary to identify which circuits have an effect on the amount of neutral-ground current. Once the offending circuits are found, they can be traced down to figure out where the problem is, fix it and reset the GF sensors to some level that might kind of coordinate (I didn't include a zone-selective scheme on the GF - mea culpa).


I am assuming the contractor somehow made a downstream connection between neutral and ground - otherwise we would have had an equipment burndown by now.

Any thoughts?

 

[ozmosis]

hmmm, increasing the trip level is not a good idea. Sounds a bit like "...this darn fuse keeps blowing. Hey, let's put a bigger one in!"
I think the exercise you mentioned to trace where the fault is coming from is the best policy. Once you have tracked it down and you know the cause, then you can remedy it.

 

[Skogsgurra]

Any frequency inverters?

Gunnar Englund

http://www.gke.org

 

[rbulsara]

Well, first and foremost is to verify if the existing settings are good or not and check the coordinaiton. They may be set too low. This will determine they are in fact nuisance trips.

Are all trips on ground fault or do they inlcude overcurrent trips?

Does the feeders have GFP as well?

If over current trips, have someone verify that all breakers are sized correctly. It is not uncommon to find motor feeder breakers undersized.

 

[Senselessticker]

I agree with rbulsara. I'd would check coordination first.

 

[swn1]

Do your main breakers have neutral CTs? Are they wired up correctly? When feeding line-to-neutral loads you have to either set the GF trip pickup above the worst-case neutral current level or measure the neutral current.

Possible problems include incorrect ratio, incorrect polarity, incorrect placement, and accidental shorting of the neutral CT wiring. All will cause nuisance tripping on GF only when unbalanced line-to-neutral loads are present.

Because the neutral conductor does not pass through the CB frame, the CT is added to the switchgear and wired into the breaker. Lots of room for error.

 

[davidbeach]

The neutral sensor also needs to be orientated correctly, if reversed it only makes things worse.

 

[tommom]

Thanks for all the info. I did suspect incorrect wiring of the trip unit, but SUPPOSEDLY the GE rep went out and looked at the unit. I'll have to get cut sheets of the units involved.

However, since there are apparently more than one unit which is tripping, that would seem to indicate the wiring problem (assuming there is one, and I think it reasonable to do so) is out on the system and not at the unit.

I ALWAYS suspect shorted, malfunctioning or backwards-wired CT's on this sort of problem but it would appear in this case the problem is elsewhere if the information is being transferred to me correctly - that is, that more than one breaker is tripping. However, it's possible they're ALL wired wrong.

I will look at the coordination, but GF on low-voltage breakers is notoriously difficult to coordinate without zone-selection. Further, even with bad coordination, that should only cause a problem if the breaker trips are set WAY too low (a possibility, ie that the GF trip unit is reading CT error current), or if there is some other reason there is current between ground and neutral. But lack of coordination doesn't cause trips, although very low settings might.

THis is a commercial building - there are VFD's but they are only like 15 HP, maybe 3-4 of them.

 

[Skogsgurra]

The power level of the VFDs is not important. The switching frequency is. And cable length. But most important is the fact that you have them. Very few installations where VFDs are involved can have RCDs (or GF) especially if there are any filters used on the VFDs. European RCDs have a 30 mA trip setting and they can not be used. There are 300 mA devices and it is quite often difficult to have reliable operation even at that high setting.

What are your GF trip settiings?

Gunnar Englund

http://www.gke.org

 

[tommom]

I'll have to check the GF trip settings, but these are not in the mA range. The main breaker's GF setting goes up to 1200 amps on the 480 volt side - I don't know what it's set at.

The system we're talking about is power ground fault - not personnel protection, if I'm understanding Gunnar's question correctly.

 

[davidbeach]

Circuit breaker GF units can also be susceptible to false residual due saturation of the "current sensors" used in the breakers. They might be something like a C10, or maybe a C5 CT.

 

[Skogsgurra]

Yes tommom, I was thinking low level ground fault. We have a problem with terminology. About time to put together an FAQ on US vs European terminology. I think that we should start a thread on that. We will need all forces to join. But it will be worth the effort.

I need to think about it.

Gunnar Englund

http://www.gke.org

 

[ozmosis]

ah,yes. That was why I was a little concerned about simply increasing the trip level, being a European. Seemed a little dangerous to me to do that in the first instant.

 

[mc5w]

There are also some other potential problems. One of them could be that the neutral is regrounded somewhere, possibly by and electrician who only got a 75% on the Code exam. Regrounding of the neutral happens a lot here is the U.S. because a lot of guys do no the difference between a service panel and a feeder panel.

Further confusing the problem was that during World War 2 the War Materials Board changed the grounding of electric ranges and dryers to TNC using the neutral as an equipment grounding conductor. Took us more than 50 years to get everybody to stop doing that with new installations but now we have a Connector Conspiracy Problem.

What is a real laugher is that Duquesne Light Company in Pittsburgh PA uses form 2s meters on 120 volt 2 wire services with 1 current coil in the neutral. Since some downstream fuseboxes have the neutral permanently bonded to the box the water pipes and equipment ground wires tend to bypass 1/2 of the meter!

Could be worse. In one building hillbilly electricians used 3-wire range outlets on 240 volts corner grounded only it was a C phase grounded system. When a tenant rented some 240 volt portable heaters he got a surprise when 1 of them contacted grounded metal.