3000 amp ground fault question

[powerpete69]

The breaker curve also appears to show that the 600 amp breaker should have tripped as well at 5.13kA........does anyone have an explanation for why it didn't per the breaker curve? The left side of that 600 amp breaker curve appears to be at 2.8kA. Perhaps the actual ground fault current was lower than that? Perhaps the 600A breaker curve does not meet its reality?

 

[hillbilly1]

No.

AF incident energy is based on 3-phase fault currents. The GF current is usually a single-phase line to ground event, and is not considered in an Arc Flash analysis.

Good to know, I was wondering about that.

 

[Hv&Lv]

No.

AF incident energy is based on 3-phase fault currents. The GF current is usually a single-phase line to ground event, and is not considered in an Arc Flash analysis.

So your saying no IE protection isn’t required or taken into account just because it’s single phase...

I realize the1584 guide doesn’t take single phase into account nor show calculations.
that doesn’t mean there isn’t IE on a single phase system, nor does it mean there aren’t calculations available. Hard to do since there aren’t any standards for it.. but the responsiblility is there
To estimate it.
I read a stated level at .2 in a post, so it’s really a moot point..

 

[jim dungar]

So your saying no IE protection isn’t required or taken into account just because it’s single phase...

No, that is not what I said. Proper PPE is always required.

I said that the calculations for incident energy require 3-phase fault currents (typically bolted). Unless you happen to be using a single-phase software package like ArcPro that does consider Line-Ground faults, but they are usually used on open air equipment not enclosed switchboards as shown in the OP.

 

[Hv&Lv]

No, that is not what I said. Proper PPE is always required.

I said that the calculations for incident energy require 3-phase fault currents (typically bolted). Unless you happen to be using a single-phase software package like ArcPro that does consider Line-Ground faults, but they are usually used on open air equipment not enclosed switchboards as shown in the OP.

ArcPro or Arcad what I had in mind, even though the results are consistently higher than the 1584 results

 

[paulengr]

No.

AF incident energy is based on 3-phase fault currents. The GF current is usually a single-phase line to ground event, and is not considered in an Arc Flash analysis.

Wow so utility engineers who often by device design (overhead power lines) can’t get to a three phase fault are doing it wrong?

The reason three phase AF isn’t normally done is convenience. In commercial/industrial equipment the heat from an initial ground fault is enough to rapidly evolve into a three phase arcing fault, within 1-2 cycles. Testing (mostly preliminary) on single phase faults has shown that the incident energy is not much less than the three phase case and there is not enough data for a model so using IEEE 1584 you just assume they are the same as per IEEE 1584. For utilities over 15 kV typically you use ArcPro which models single phase arcs only based on data from Kinetrics labs. There is no three phase 35 kV model.

 

[paulengr]

The breaker curve also appears to show that the 600 amp breaker should have tripped as well at 5.13kA........does anyone have an explanation for why it didn't per the breaker curve? The left side of that 600 amp breaker curve appears to be at 2.8kA. Perhaps the actual ground fault current was lower than that? Perhaps the 600A breaker curve does not meet its reality?

Those SE breakers are getting low in inventory, a sure sign of issues with them. I’ve run into a few where there is maybe one or two left anywhere in North America.

My experience is that the breakers themselves are pretty reliable. The trip units though are not so good. That is the main reason SqD abandoned the series for the Merlin Gerin/Masterpacts. At one customer who just recently started testing I had between 5 and 50% failure rates on trip units on I think 9 different distribution panels in an industrial assembly plant. Switching to a URC trip unit (if available) does the trick or at the relatively low cost for ILines replace the switchboards one at a time using the good spares to maintain the others. The trip units are always the battle.

The fact that short circuit and arcing fault is around 2X long term means adjusting taps is a waste of time. Everything is going to be separated by time. Normally unless you cannot turn it off instantaneous is only used at the lowest (device) level. Distribution breakers either turn it off or set it above the short circuit current so it never trips.

Last thing you might not have done in Easypower. I mostly use SKM and ETap so not sure if it can. Put both ground AND phase current trip settings on the same TCC. With your settings often you will get miscoordination between the two. Phase overcurrent doesn’t care if it’s between phases or phase to ground. That’s when we get into negative sequence current trip functions but these just lead to frustrating troubleshooting so my experience is to avoid using them or plan on taking those 2 AM phone calls as an engineer.

5-10 kA is a wonderful thing from an AIC point of view but leads to long trip times to coordinate and thus high arc flash and damage. Recommend in the future two things. First split the transformers into smaller ones. 3000 A is ridiculous. Other than pulling lots of big wire and bus what do you need that for? 1200 A and lower switches from switchboards to panel boards and the overall cost stays about the same. Second study high resistance grounding closely. Your case sounds like a strong candidate for a retrofit to it. This takes your 5.3 kA ground fault to say 10-15 A. And there are tons of breakers and overload relays on the market (Eaton, ABB) that support this feature so that you can get away from the single breaker solution. In the interim you can set it up for continuous duty and then manually hunt down the ground fault, avoiding what you have now with plant wide black outs because a dinky 2 HP motor burned up and caused a ground fault trip.

 

[powerpete69]

Those SE breakers are getting low in inventory, a sure sign of issues with them. I’ve run into a few where there is maybe one or two left anywhere in North America.

My experience is that the breakers themselves are pretty reliable. The trip units though are not so good. That is the main reason SqD abandoned the series for the Merlin Gerin/Masterpacts. At one customer who just recently started testing I had between 5 and 50% failure rates on trip units on I think 9 different distribution panels in an industrial assembly plant. Switching to a URC trip unit (if available) does the trick or at the relatively low cost for ILines replace the switchboards one at a time using the good spares to maintain the others. The trip units are always the battle.

Thanks, will consider doing this if it continues to be a problem

The fact that short circuit and arcing fault is around 2X long term means adjusting taps is a waste of time. Everything is going to be separated by time. Normally unless you cannot turn it off instantaneous is only used at the lowest (device) level. Distribution breakers either turn it off or set it above the short circuit current so it never trips.

Got it.

Last thing you might not have done in Easypower. I mostly use SKM and ETap so not sure if it can. Put both ground AND phase current trip settings on the same TCC. With your settings often you will get miscoordination between the two. Phase overcurrent doesn’t care if it’s between phases or phase to ground. That’s when we get into negative sequence current trip functions but these just lead to frustrating troubleshooting so my experience is to avoid using them or plan on taking those 2 AM phone calls as an engineer.

Didn't I put the ground and phase current together on my TCC on post #3 and #4? From this TCC, It doesn't make sense how the ground fault tripped and the 600 A breaker didn't. Unless the actual line to ground fault was less than 3kA.

5-10 kA is a wonderful thing from an AIC point of view but leads to long trip times to coordinate and thus high arc flash and damage. Recommend in the future two things. First split the transformers into smaller ones. 3000 A is ridiculous. Other than pulling lots of big wire and bus what do you need that for? 1200 A and lower switches from switchboards to panel boards and the overall cost stays about the same. Second study high resistance grounding closely. Your case sounds like a strong candidate for a retrofit to it. This takes your 5.3 kA ground fault to say 10-15 A. And there are tons of breakers and overload relays on the market (Eaton, ABB) that support this feature so that you can get away from the single breaker solution. In the interim you can set it up for continuous duty and then manually hunt down the ground fault, avoiding what you have now with plant wide black outs because a dinky 2 HP motor burned up and caused a ground fault trip.

Unfortunately the plant came with a 1500KVA Transformer, 3000 amp main and 6 sets of 750 MCM cable. I will certainly consider retrofiting the unit.

 

[jim dungar]

Wow so utility engineers who often by device design (overhead power lines) can’t get to a three phase fault are doing it wrong?

The equipment in the OP is an enclosed UL Listed switchboard, it is not an overhead power line. The OP said that Easypower was being used, it is based on IEEE 1584 and does not use L-G fault current in its calculation.

 

[mayanees]

Isn't that breaker equipped with an indicator that tells which function tripped the breaker? That info is critical and should be noted by whoever resets the breaker.

 

[jim dungar]

Those SE breakers are getting low in inventory, a sure sign of issues with them. I’ve run into a few where there is maybe one or two left anywhere in North America.

The SE line of breakers was obsoleted, by Square D almost 20 years ago and are no longer offered by them. As Paul said, even the secondary market of them has dried up. Retrofitting them with a new style breaker is a pretty common.

 

[powerpete69]

Isn't that breaker equipped with an indicator that tells which function tripped the breaker? That info is critical and should be noted by whoever resets the breaker.

Probably, but that breaker was reset 3 months ago. Perhaps it holds a log of the past or something.

 

[jim dungar]

Probably, but that breaker was reset 3 months ago. Perhaps it holds a log of the past or something.

I do not remember the SE breaker as having a trip history. It definitely does not have a display as is found on more recent trip units.

My guess is that you had an arcing ground fault less than 3kA. It does not take a lot of conductor length to limit fault current on branch circuits less than 100A.

 

[powerpete69]

I do not remember the SE breaker as having a trip history. It definitely does not have a display as is found on more recent trip units.

My guess is that you had an arcing ground fault less than 3kA. It does not take a lot of conductor length to limit fault current on branch circuits less than 100A.

Could be, but that 70 amp breaker shows it clearing at .02 seconds.....whereas the ground fault curve shows it won't fully trip till 1.1 seconds....strange.
Unless of course the ground fault on the RTU was less than 1,100 amps....the 70amp breaker doesn't clear till around .7 seconds!!!!
This could very well be the case.
As they say in the business, there is no perfect coordination......just a perfection of imperfections!

 

[jim dungar]

Unless of course the ground fault on the RTU was less than 1,100 amps...

I can believe this was the case. I have 'studied' many branch circuits where the fault current is not high enough to cause protective devices to open instantaneously.

 

[powerpete69]

I can believe this was the case. I have 'studied' many branch circuits where the fault current is not high enough to cause protective devices to open instantaneously.

Yes, per that breaker curve, if the fault current is less than 1,100, it will ride all the way up that teal green leg tripping the ground fault. Then again, the 70 amp breaker did actually trip!~ The ground fault should have kept the 70A breaker from tripping if this was the case!!! Quite the mystery!!

 

[jim dungar]

Yes, per that breaker curve, if the fault current is less than 1,100, it will ride all the way up that teal green leg tripping the ground fault. Then again, the 70 amp breaker did actually trip!~ The ground fault should have kept the 70A breaker from tripping if this was the case!!! Quite the mystery!!

No.
If the current falls anywhere inside of the curve, the breaker may begin to open, but the current may continue for the time it takes to clear the curve on the right hand side.