Ground fault trips upstream breaker instead of downstream even if they are coordinated?

[main197]

Upstream breaker:
Sq D 800A LSIG Micrologic 6.0P
Ground Settings: 160A, 0.4 s
S: 5120A, 0.2S and Inst: 6400A

Downstream breaker:
ABB 400A Ekip Dip LSIG
Ground Settings: 80A, 0.2 s
S: 600A, 0.2 s and Inst: 2200A

Shouldn't the ground fault at the load side of 400A breaker trip it instead of the 800A breaker? The fault current was ~1400A and tripped the 800A instead of the 400A breaker. What could be the reasons? Improper coordination settings? Haven't checked the TCCs using ETAP or SKM.

 

[jim dungar]

We the breakers actyally tested for coordination?

Was this actually a ground fault or the the breaker trip on a different function?

 

[main197]

We the breakers actyally tested for coordination?

Was this actually a ground fault or the the breaker trip on a different function?

No, the breakers were not tested.
Micrologic 6.0P shows it as a ground fault.

 

[jim dungar]

If this were a critical circuit, I would test the breakers individually and in series.

 

[ATSman]

Found this old thread.

Zone Selective Interlocking

I am trying to figure out how does zone selective interlocking (ZSI) work. In traditional coordination (without ZSI), for a downstream fault the downstream breaker will trip instantly and the upstream breaker will not trip. And if we have coordination with ZSI, then again the downstream breaker...

forums.mikeholt.com

 

[jim dungar]

Found this old thread.

Zone Selective Interlocking

I am trying to figure out how does zone selective interlocking (ZSI) work. In traditional coordination (without ZSI), for a downstream fault the downstream breaker will trip instantly and the upstream breaker will not trip. And if we have coordination with ZSI, then again the downstream breaker...

forums.mikeholt.com

It is likely all but impossible to zone interlock breakers from different manufacturers.

 

[David Castor]

Even though it's a ground fault, if the current is high enough, the phase overcurrent elements will see the fault (in one phase) and can also trip. A phase instantaneous trip will be faster than a ground fault with a time delay. Since both of these breakers have instantaneous trip, these can't really be reliably coordinated in the instantaneous region of fault current.

 

[main197]

Even though it's a ground fault, if the current is high enough, the phase overcurrent elements will see the fault (in one phase) and can also trip. A phase instantaneous trip will be faster than a ground fault with a time delay. Since both of these breakers have instantaneous trip, these can't really be reliably coordinated in the instantaneous region of fault current.

That is correct. Micrologic 6.0P showed the fault current of ~1400A. Unless that is a false reading, the current was not high enough to trigger the instantaneous trip. That's why I am scratching my head trying to figure out why it tripped the upstream breaker.

 

[TwoBlocked]

Gotta ask: Where in the system was the fault?

 

[main197]

Gotta ask: Where in the system was the fault?

Somewhere downstream fed by the 400A breaker.

 

[ATSman]

I can't tell you how many times I have seen and electrician changing a 277v (480v system) lighting ballast hot, and shorts to ground. The 20A subpanel breaker trips as well as the 4000A main breaker on the 20th+ flr on ground fault. So many systems are built with GF only on the main and not the secondary feeders because the NEC ONLY requires it on the main. Customer says "something wrong with my GF system". Tests show it is operating properly. The 20A breaker trips on instantaneous at a value high enough to trip the main GF setting of typically set to 10% of the main rating (400A) regardless of the typical GF time delay setting of 0.4 sec. or greater. It turns out to be a crap shoot and usually all or some of the breakers in the string trip.
Solution? Install GF protection on all feeder breakers below the main breaker, localizing the fault and keeping the main on. This is rarely done due to the cost.
After writing this I just realized that there is GF on the 400A bkr. ?!?! Now I'm scratching my head!?!?!
Because the fault current rises so fast, what the OP could try is raising the TD setting on the 800A GF. The GF current setting is not as important as the TD setting (1 sec max.)
Or,
Change out the ABB breaker to Square D and add Zone Interlocking.

 

[David Castor]

That is correct. Micrologic 6.0P showed the fault current of ~1400A. Unless that is a false reading, the current was not high enough to trigger the instantaneous trip. That's why I am scratching my head trying to figure out why it tripped the upstream breaker.

Based on that, the most likely problem is a defective trip unit on the downstream breaker. Is is possible that the ground trip function is somehow turned off on the downstream breaker? I'm not familiar with these trip units. Or is the trip unit supposed to be getting a ground current input from a ground CT and it isn't connected?

 

[hillbilly1]

Is the ground fault CT built into the breaker? (With separate neutral CT) or an external box type CT encircling all of the load conductors?

 

[hillbilly1]

Based on that, the most likely problem is a defective trip unit on the downstream breaker. Is is possible that the ground trip function is somehow turned off on the downstream breaker? I'm not familiar with these trip units. Or is the trip unit supposed to be getting a ground current input from a ground CT and it isn't connected?

I have seen that before, the neutral CT was not connected, so when the neutral current got high enough, it tripped the main because the breaker didn’t “see” that current returning correctly. Also have seen the neutral CT connected backwards cause the same thing.

 

[David Castor]

The ground CT input is often an option for ground fault protection - instead of the unit summing all three phase (and neutral) currents.

 

[mayanees]

Maybe there's a problem with the downstream breaker. Here's what it looks like on a TCC:

 

[ATSman]

The ground CT input is often an option for ground fault protection - instead of the unit summing all three phase (and neutral) currents.

That is correct.....the ground CT (also called a Ground Strap) measures the ground fault current returning on the ground instead of the neutral. The window type is called Zero Sequence design.

 

[ATSman]

I have seen that before, the neutral CT was not connected, so when the neutral current got high enough, it tripped the main because the breaker didn’t “see” that current returning correctly. Also have seen the neutral CT connected backwards cause the same thing.

Yes, and depending on the scheme I have seen the breaker nuisance trip on normal load current if the neu CT was not connected or wired reverse polarity.
We definitely need to know what type of GF scheme is on the 800A breaker and if it is wired properly.

https://electrical-engineering-portal.com/scheme-sensing-ground-faults

https://electrical-engineering-portal.com/ground-fault-protective-schemes

 

[RonUH]

Couldn't something like this also happen as a result of multiple neutral to ground bonds? Seems like a read that somewhere but cannot remember for sure.

 

[ATSman]

Couldn't something like this also happen as a result of multiple neutral to ground bonds? Seems like a read that somewhere but cannot remember for sure.

My experience is the opposite. Multiple neutral grounds would de-sensitize a GFR system. Neu should only be grounded in one spot: upstream of the neu sensor CT (ground strap scheme) at the service transformer: NEC Article 230.95. see diagram in https://electrical-engineering-portal.com/scheme-sensing-ground-faults
On zero sequence schemes where window CT is around all conductors and neu. grounding has to be upstream of window CT so the fault current is not cancelled out and trips breaker.