2000A Main Breaker Settings

[steve66]

I suppose if you turn up the settings the HVAC equipment will burn up and then you can replace the HVAC equipment.

How would you get enough ground fault current all the way out at the HVAC that you could trip the GFPE back at the main unless you had some kind of dead short out there.

I've seen a small AC unit trip a 800A breaker several times, and that breaker didn't have any ground fault.

So I completely believe a small AC unit could easily trip a 640A ground fault sensor.

Except for the winding impedance, a dead short is basically the failure mode for a locked up or stalled compressor.

 

[texie]

The OP's situation is a perfect example of why single level GFPE protection is a really poor design choice. Yet I see it regularly do to short sighted thinking to save a buck. As Jim Dungar and others noted, I too have seen many a commercial building blacked out unnecessarily do to this. Sure, the OP has some setting and testing issues on his, but that alone will not be the full cure.

 

[petersonra]

I'm having a hard time understanding how the GFPE could trip on the 2000 amp circuit breaker because of startup current on a motor. Isn't the GFPE measuring the current flowing thru bonding means? So if there's no current flow there how would the GFPE trip?

How is it that startup current could trip the GFPE but the same amount of current flowing to a larger motor under non-startup conditions would not trip the GFPE?

 

[jim dungar]

How is it that startup current could trip the GFPE but the same amount of current flowing to a larger motor under non-startup conditions would not trip the GFPE?

If there is a moderate short to ground in the feed to the motor, wouldn't the main trip on GF whenever the motor pulled more than 640A, but no trip would occur at normal running current?
Cranking up the GF settings might allow the motor to start but would also allow the fault to burn for a longer time, which would make troubleshooting easier because it could be done using human senses like sight and smell.

Long lasting/persistent ground faults is why GF protection is required for 1200A and larger 480Y/277V circuits.

 

[petersonra]

If there is a moderate short to ground in the feed to the motor, wouldn't the main trip on GF whenever the motor pulled more than 640A, but no trip would occur at normal running current?
Cranking up the GF settings might allow the motor to start but would also allow the fault to burn for a longer time, which would make troubleshooting easier because it could be done using human senses like sight and smell.

Long lasting/persistent ground faults is why GF protection is required for 1200A and larger 480Y/277V circuits.

Why would the startup current cause more ground fault current? If there's a ground fault you're going to have a continuous amount of it it is not going to significantly change just because you start the motor.

 

[jim dungar]

Why would the startup current cause more ground fault current? If there's a ground fault you're going to have a continuous amount of it it is not going to significantly change just because you start the motor.

You are correct if there is a bolted fault on a feeder to the motor. But, say you have a fault internal to a motor such that it is not a bolted fault, isn't the amount of current flowing over the fault path a function of the amount of current drawn by the motor? How about if the feeder fault is an arcing fault causing a current divider due to the parallel current path?

 

[zbang]

Throwing ideas at the wall-
Could be that some wire or winding shorts to ground given enough current (e.g. motor starting surge) but reads open when de-energized. A Megger might find that. Or maybe put a peak-reading amp meter on the EGC close to the motor or around all three supply leads and record a startup.

 

[petersonra]

You are correct if there is a bolted fault on a feeder to the motor. But, say you have a fault internal to a motor such that it is not a bolted fault, isn't the amount of current flowing over the fault path a function of the amount of current drawn by the motor? How about if the feeder fault is an arcing fault causing a current divider due to the parallel current path?

In which case you have a real ground fault problem that needs to be addressed. But it has nothing to do with startup current.

In any case with a 40 amp circuit breaker you would have an instantaneous trip that's considerably below 640 amps. With the GFPE delay of a tenth of a second you would think that the 40 amp breaker would trip first.

 

[ATSman]

In which case you have a real ground fault problem that needs to be addressed. But it has nothing to do with startup current.

In any case with a 40 amp circuit breaker you would have an instantaneous trip that's considerably below 640 amps. With the GFPE delay of a tenth of a second you would think that the 40 amp breaker would trip first.

Petersonra, I agree with the first two sentences, but not the rest.
Ok we have been down this road so many times. I tried to find a past post I wrote earlier on this subject but failed. So again, here is my two cents.
First, I can't tell you haw many times a cust calls and said main bkr tripped on GF. Something is wrong with it, can you test it? I ask any work done
in the high rise lately? Any reports of equipment failures? He says just electrician changing light ballasts (277V lighting) who was doing it hot. I go there find the 20A branch breaker tripped in the branch panel in the room he was working in. The main that tripped on GF was 20 or 30 stories down in the basement. The main settings were typically set at 400A @ 0.5 sec on a 4000A bkr. Ok so the shorted hot to ground on the ltg ballast tripped the 20A bkr on inst trip and also went down to the main bkr and tripped it on GF. Nothing wrong with his GF system, just doing it's job.
Senario #2
An old chiller on the roof of another bldg started having problems. One day the 3000A main trips on GF in basement. Find out the start contactor on chiller was arc welded closed. Turned out the compressor motor winding went to ground, tripped the chiller main and took out the main on GF.

Bottom line: a GF in a poorly designed system (due to economics!) will only have 1 level of GF protection: main breaker only! And the reason is that it has to be there by code on a 480V system greater than 1000A, if my memory serves me (I am not a code guy.) And without it and without a field acceptance test on the main bkr GF system by a 3rd party testing co it will not pass inspection by the AHJ and cannot be turned on.
So in this case the OP was on the right track when he suspected the HVAC unit on the roof causing the main to trip. In so many cases it is race and the bkrs in the path of the GF will ALL trip till it reaches the main GF protection. It is a race of OC trips!
Since there is no way to coordinate instantaneous or GF tripping of breakers in series, the only solution I see is to add secondary or a third level of GF protection to limit the outage to a specific area. This is rarely done due to the costs.
One last point. Yes the minimum settings shown on the Main bkr are an indication that they are factory default setting (liability) and a proper startup/commissioning was not performed on this equipment. But experience has shown that even with the proper settings from the engineer, the tripping would still occur.
Ok I will stop rambling and will be interested in hearing any comments on this subject.