Light fixture fault..Trips Switch Gear...For the ENTIRE 3 Story School.

Status
Not open for further replies.
A

Amon

Member
Location
Denver, CO
Balancing not right??


An 18ga neutral and hot inside of a light fixture were tied together. Light was turned on... Entire building loses power. Same thing happens with a ground fault.

This is almost a 30,000 sq ft building. Completely shut down. Not a single breaker or sub panel tripped. Only the main throw out at service.
 
Sounds like the main has never had the trip settings calibrated.
Seeing a low current GF and opening.
 
As others have said the system is not properly coordinated.

The main certainly has ground fault protection. It is probably set to it's minimum value rather than a properly engineered value.

All the other breakers probably don't have ground fault protection except perhaps some GFCI breakers. If you compare the trip curves of the breakers, the main is currently more sensitive to this fault then the down stream breakers.
 
No coordination- and it does not surprise me in that breaker dials come set to the lowest setting and are rarely adjusted. Have a coordination study done by an engineer to figure out where to set the dials.
 
mbrooke said:
No coordination- and it does not surprise me in that breaker dials come set to the lowest setting and are rarely adjusted. Have a coordination study done by an engineer to figure out where to set the dials.
Click to expand...

or just turn the dial until your fault doesn't trip the main before the branch circuit OCPD. You ought to be able to get that done in 10 or 15 tries.:happysad:
 
winnie said:
As others have said the system is not properly coordinated.

The main certainly has ground fault protection. It is probably set to it's minimum value rather than a properly engineered value.

All the other breakers probably don't have ground fault protection except perhaps some GFCI breakers. If you compare the trip curves of the breakers, the main is currently more sensitive to this fault then the down stream breakers.
Click to expand...


I can feel you on the trip value settings however....When you read the fault message on the switch gear after a trip it is showing 345amp 430amp faults......I get the minimum value deal but thats a pretty dadgum big number in my mind I don't know much though. Sure there are not any GFCI breakers...but still.. I don't understand how it does not trip the 20amp lighting circuit, does NOT trip the 150amp Sub panel for that circuit, does NOT trip the CT cab (400amp)....Trips out at the street...How could the minimum setting on the switch gear keep a 20amp breaker from tripping first? It makes it to the street before the 20amp can open up??
 
Amon said:
How could the minimum setting on the switch gear keep a 20amp breaker from tripping first? It makes it to the street before the 20amp can open up??
Click to expand...

Very very easily.

I cannot remember seeing a 1200A, or larger, ground fault relay coordinate with a 20A breaker unless the settings have been turned higher than minimum. Our standard procedure is to use a 30A breaker as the benchmark device and then develop GF settings above that. It is not unreasonable to see a 20A hold 200A for several seconds.

The biggest issue I see is the time element. GF devices usually have settings of 0.0, 0.1, 0.2, 0.3, and 0.4 secs, often with minimum pickup points of 200A.
Small breakers on the otherhand are usually provided with instantaneous clearing points a little over 10X. These small breakers also are often able to handle inrush and motor starting currents of roughly 6 times for 20 sec. It is not unreasonable to see a 20A breaker carry 200A for several seconds before clearing it.

But look at all the money that was saved by not paying for a coordination study.
 
Amon said:
....Trips out at the street...How could the minimum setting on the switch gear keep a 20amp breaker from tripping first? It makes it to the street before the 20amp can open up??
Click to expand...

The whole point of a breaker is to detect some sort of anomalous situation and shut down the circuit, while at the same time permitting normal operation. Breakers also have a certain amount of manufacturing tolerance.

Breakers need to permit normal loading, and also the 'inrush' of various loads (high current drawn by a load for a short duration). If you look at the 'time current curve' for a standard 'QO' breaker, at 1 second it will trip at somewhere between 6 and 12x its handle rating. At 0.3 second it will trip somewhere between 9x and 22x its handle rating. This means that a 20A breaker might reasonably 'hold' 300A for 0.3 seconds, and be completely within spec.

Normal breakers look for excess current flow to detect the anomalous situation. They are not smart, just letting a thermal element heat up or a magnet to pull in order to trip the breaker.

The ground fault detection capability of the main adds some serious complexity to this issue. Aside from a very small amount of leakage and capacitive charging current, no current should flow to ground. So any significant ground fault current indicates an anomalous situation. A 1000A breaker with 2000A of current (for a short period of time) might be perfectly normal, but a 1000A breaker with even 50A of ground fault current indicates a real fault.

So if you have a main breaker (with ground fault detection) picking up ground fault current, and it hasn't been properly set up to let the smaller breaker trip first (on excess current), it is very understandable that the main breaker will trip first.

-Jon
 
Amon said:
I can feel you on the trip value settings however....When you read the fault message on the switch gear after a trip it is showing 345amp 430amp faults......I get the minimum value deal but thats a pretty dadgum big number in my mind I don't know much though. Sure there are not any GFCI breakers...but still.. I don't understand how it does not trip the 20amp lighting circuit, does NOT trip the 150amp Sub panel for that circuit, does NOT trip the CT cab (400amp)....Trips out at the street...How could the minimum setting on the switch gear keep a 20amp breaker from tripping first? It makes it to the street before the 20amp can open up??
Click to expand...



Speed. If 5 amps triggers the GFP to unlatch opening the main breaker all taking place in 6 cycles total before the 20amp's bimetal has a chance to heat up to the point that it can pull back and unlatch (that taking say 20 cycles) then its a given.


If 400amps is above the 20amp's magnetic pickup then the possibility of either one or even both simultaneously tripping goes up, but its still about who can win the race.
 
It happens a lot.
When I called the ee he told me to turn it up a bit, we were still roughing in so all new gear etc and a cord ~300' from the gear was pinched into the metal stud track; cord was plugged into 100a sub with gfi receipts.., tripped the main. Told the ee I couldn't figure out how, GE breaker had all the computer features you could buy (high priced school tax), and by the way the instructions were shipped in for you to pickup. I reset breaker with no change he picked up instructions I called about a week later he told me to just reset again if it trips. I doubt it ever was changed.
 
sameguy said:
It happens a lot.
When I called the ee he told me to turn it up a bit, we were still roughing in so all new gear etc and a cord ~300' from the gear was pinched into the metal stud track; cord was plugged into 100a sub with gfi receipts.., tripped the main. Told the ee I couldn't figure out how, GE breaker had all the computer features you could buy (high priced school tax), and by the way the instructions were shipped in for you to pickup. I reset breaker with no change he picked up instructions I called about a week later he told me to just reset again if it trips. I doubt it ever was changed.
Click to expand...


Interesting. Same story.. Nobody really cares on their end... One guy said. "At least its working"....But its not lol
 
Amon said:
Interesting. Same story.. Nobody really cares on their end... One guy said. "At least its working"....But its not lol
Click to expand...

Look at the bright side, if an arc flash ever takes place you know the OCPD will respond faster then a QO.
 
Status
Not open for further replies.
Top