3000A Ground Fault Protection breaker tripping

[powersoff]

I had a 3000 amp 480 volt breaker trip on a Ground Fault. The Hmi for the ground fault showed 79Amps. I did find a 36kw heater with all 3phases grounded. It was fed with (3) 50a LPJ fuses. 1 of the fuses was blown.
In the future I would like fuses to open before tripping the 3000 amp breaker. Does anyone have any suggestions as to what sort of protection should be implemented to act faster than a ground fault protection ?

 

[ScottyUK]

In most places using the ground as anything but a protective conductor is a code violation: run a neutral to the star point. It sounds like your breaker reacted in a reasonably foreseeable manner to a badly designed installation.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[powersoff]

Sorry for not being ,clear the connections at the heater were bad/grounded/in water.
I understand that the ground fault breaker did what it was supposed to. But let me assume that in the future something bad happens to another heater( I have many heaters on site all with similar fusing) what I would like to avoid happening is 3000 amp main tripping on a ground fault that feeds dozens of other loads.
The 3 phase 480 volt supply(hot) conductors feeding the heaters are not normally grounded. But today something bad happened.

Thanks,

 

[dpc]

A 3000 A breaker probably cannot trip on 79 amps of ground current. I'm not sure what the lowest setting is, but it will be in the hundreds of amps. Maybe 7900 amps? Or a mis-operation.

The ground fault trip unit will have time delay setting adjustments. For a breaker this large, you will want a fairly long time delay to allow downstream protection time to clear the fault.

What type of trip unit and what pickup and time delay settings?

 

[jraef]

You could install smaller breakers at each load or zone with ground fault trips coordinated to trip before the GF trip in the main.


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[waross]

I assume that the breaker has a shunt trip that is operated by a ground fault detection circuit.
Consider more GFI breakers. Your distribution panels may be retrofitted with ground fault breakers set lower than the main breaker.
As an alternative or addition, the breakers feeding the heaters may be retrofitted with GFI breakers. With proper coordination, the breaker closest to the ground fault will trip first and isolate the fault without interrupting service to other devices.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[rbulsara]

For what it is worth, NEC permits up to 1200A GF pick up and 0.5sec of maximum time delay on 480V main. This for the reasons dpc indicated.

Rafiq Bulsara

http://www.srengineersct.com

 

[rcw retired EE]

Is your service 3-phase, 4-wire? If so, possibly the neutral sensing CT is backwards. Under normal load the meutral current is below the trip point.

When a high harmonic load with lots of neutral currrent comes on, the swapped CT will cause a GF trip.

 

[jghrist]

You cannot generally get fuses to coordinate with a sensitive ground fault device. As Bill suggested, you need GFI breakers on your branch circuits to the heaters to get coordination.

 

[powersoff]

Today I was able to eyeball the Ground Fault Relay it is a Sq D GC200E.
The GF delay was set to 0s and the pickup was at 120A.
I changed the GF delay to .2 sec.

The GC DSP showed the I(g)"last trip" as 79A. After talking to tech support they suggested with no delay, relay actuates faster then display.

 

[powersoff]

My service is 480Y/277 3phase 4 wire.

 

[alehman]

Those settings are still very sensitive for a 3000A service. It probably won't coordinate with downstream fuses unless they are very small (it might coordinate with the 50A heater fuses). You should plot the time-current curves to verify.

Alan
----
"It’s always fun to do the impossible." - Walt Disney

 

[Zogzog]

"The GC DSP showed the I(g)"last trip" as 79A. After talking to tech support they suggested with no delay, relay actuates faster then display. "

Nonsense, the relay has to see Igf above pickup and will record it, something is amiss here, when is the last time this GF system was tested?

 

[jghrist]

What kind of system grounding do you have?

 

[dpc]

If this is a solidly-grounded system, that ground pickup setting on a main breaker is much too low.

In an integral ground trip unit, the lowest available setting would probably be 0.2 or 600 A. We would likely set it much higher than that, with a long time delay.

 

[PHovnanian]

The GF delay was set to 0s and the pickup was at 120A.
I changed the GF delay to .2 sec.

This problem might call for a bit more of a protection coordination study. If that 3000A breaker was originally set as you specified, the intent might have been to protect downstream branch circuits/feeders (arguably a poor design). Now it might be set too high and require some downstream GF protection.

Sadly, the basis for original settings such as this one are rarely documented (other than by the firm that did the design). One often has to reverse engineer things to deduce what they were thinking.

 

[pwrengrds]

I would recommend taking a serious look at every setting on that breaker, if the ground fault is at 120 amps, I would guess that it was never set up in the first place and was just put in a minimum settings. Find the coordination study and set the breaker accordingly, A 3000 amp breaker ground fault should be set so small breakers (<100amps) will clear faults before tripping on ground fault.

 

[powersoff]

I love this site.My favorite response so far " something is amiss". I agree.
If it has been tested the documentation went the way of the dinosaur.
As for the coordination study , they might have taken the dimensions of the gear and figured out they could put it in said location.
Tech support said in the absence of design specs the factory default position for time delay is 0 and pickup 120A.
I believe it has been operating this way for about 8 years.

The end use load with the fault in this case was a 36kw heater fused at 50amps with a 200amp I line breaker feeding the fuses all originating from the 3000amp breaker with the ground fault relay.
Tech support is stating that if ground fault relay set above 200amp pickup all sorts of bad things could/will happen and that I need to implement even more ground fault protection to be safe.
Is this a case of bells and whistles being added on to bells and whistles?
Or does this sound normal to someone.
I have not seen in the field supplementary ground fault protection for motors and heaters.I have seen it for heat tape.
And in my bathroom about once a year the GFI trips .

 

[alehman]

factory default position for time delay is 0 and pickup 120A.

I have never seen a breaker that would allow GF time delay set to 0. What kind of breaker is this?

Tech support is stating that if ground fault relay set above 200amp pickup all sorts of bad things could/will happen

nonsense. I would commonly set it up around 0.5 sec and 1000A. Again, coordination study is needed.

I have not seen in the field supplementary ground fault protection for motors and heaters

A second level of ground fault protection can frequently improve coordination. You can certainly install branch circuit breakers with ground fault trip, for $$. This is not the solution to your present problem.

My favorite quote:

As for the coordination study , they might have taken the dimensions of the gear and figured out they could put it in said location.

Alan
----
"It’s always fun to do the impossible." - Walt Disney

 

[pwrengrds]

When I set up a GF without a coordination study, after looking over the system it generally gets set at 25% of breaker rating and 0.5 seconds, similar to the way Alan commented on setting it up.