Arc Flash in a tie breaker compartment? 1

[JFLovvorn]

I did some searching and reading through these forums before posting, and was unable to find an answer. If this topic has already been covered, I apologize and if a link could be provided I'll happily delete this post (can I do that? I'm new here) and go read the original thread.

That said, here's my question.

I'm working on a power study and arc flash analysis using SKM Power*Tools software for an approximately 200-bus system at a local plant. For the most part it's pretty straightforward - IEEE 1584 and NFPA 70E have been helpful, and the IEEE Red Book as well. However, there are several ties in the plant, including three at the 13.8kV level in the main electrical substations. The utility company feeds 161kV into the yard, where the plant has six transformers with OCB protection. Each transformer steps down to 13.8kV then feeds through a 13.8kV main breaker into switchgear. A tie breaker/bus is located between the switchgear in pairs (so 1 can be tied to 2, 3 to 4, and 5 to 6).

Now, I can calculate the arc flash at the main 13.8kV bus with no problem during normal operation (all ties open). I can also calculate with the tie breaker closed. However, it seems to me that there would always be a high flash value present in the tie breaker compartment because it has two energized buses at all times, fed by different transformers. If a fault occurred in the tie compartment, couldn't it have contribution from each bus?

My question is, would the tie compartment need to be labeled with the tie-closed flash label, or would it be a summation of the two tie-open flash values? It seems to me that the worst-case scenario would be a fault starting at one side of the open tie breaker, burning through almost its entire lifetime, then spreading enough to trigger a fault on the other side of the tie breaker from the other bus. In essence, two faults from two separate busses in one compartment. However, I'm not sure how likely that is. It seems more likely that the tie-closed flash value would be the value needed at a tie. I haven't been able to find any literature that discusses this scenario, though, and I'd love to find an IEEE or other standard to back me up on this.

If anyone can point me toward some literature, it would be very helpful. Or some discussion on the subject from more experienced engineers would be wonderful. I'm a year and a half out of school and this is the biggest project I've had so far.

Thanks in advance for any help, and sorry for the long-winded post!

 

[davidbeach]

If the gear can run tie closed, along with both mains closed, the whole gear should be labeled for that condition. Even in the tie compartment, you are likely to only get contribution from one side or the other, unless you are having a fault that goes around the breaker.

 

[arcflash99]

Well, I'm not an engineer, but, this can be a difficult topic and takes some thought into publishing an ATPV that will be used to protect people's lives.

We operate with bus ties closed in most of our plant. Therefore most of our calculations are derived with bus ties closed. We use a bus tie closed operation to provide an ATPV for the bus and for the tie breaker.

Good luck on finding published information.

Even if the ties run open I would label each bus with the highest value, unless tie closed exceeds 100 Cal/cm2.

 

[dpc]

I agree with David Beach. If you can run with tie closed and both mains closed, this condition needs to be evaluated for all equipment not just the tie breaker. For the tie breaker, you need to determine which upstream devices must open to clear the fault, since you obviously can't count on the tie breaker itself.

You need to consider multiple operating conditions and label each bus or piece of equipment for the worst-case arc energy. This may or may not correspond to the maximum fault current. The tie breaker is not really any different in this regard.

 

[waross]

Do I undestand your question correctly?
Case 1> Tie breaker closed. Contribution from both supplies. Energy is total current (both supplies) times maximum clearing time. No problem
Case 2> Tie breaker open. Contribution from one side or the other.
Energy is current from one source times clearing time. No problem.
Case 3> Tie breaker is open. A fault develops on one side and involves the other side.
If the involvement is almost instantaneous, the energy will be similar to case 1.
If the second side is involved just as the first side clears, the arc will persist until the second side clears.
Question;
In the event of case 3, how to calculate arc flash.
Is it reasonable to assume that the arc will involve both sides and calculate the energy as the current and clearing time of the first side, plus the current and clearing time of the second side.
Have I correctly restated your original question?
Comments?
respectfully

 

[JFLovvorn]

waross, you have stated the question correctly. Evaluating Case 1 and Case 2 is no problem, and I have done so. Case 3 is the problem. Is it even a feasible scenario? It seems to me that the highest possible rating in Case 3 would come about by adding the flash value of Bus 1 to the flash value of Bus 2, I just don't know if Case 3 is even possible.

 

[dpc]

That is actually a very interesting question, now that I understand it. For a fault within the tie breaker cubicle itself, it might be possible for there to be contribution from both sides due to the conductive plasma that is created by the arc.

The same situation would occur at an open main breaker when operating with the tie closed.

In most cases, I suspect the fault would be limited to one side or the other, but I'm not sure you can rule out the possibility that the fault could extend itself to involve both sides. I'm hoping someone else can convince me this can't happen. But based on some of the fireballs I've seen on video, I suspect it's possible.

AFAIK, the IEEE-1584 or NFPA-70E standards do not specifically address this issue.

 

[JFLovvorn]

I agree, it's an interesting question. I think for now I'm going to suggest that the tie breakers be labeled with the worst-case scenario as follows:

With the tie breaker and both mains closed, the entire lineup will have one value.

With the tie breaker open and both mains closed, each lineup will have its own value. The tie breaker compartment will be labeled with a separate value consisting of the sum of the two hazard values from each bus. This should represent a worst-case scenario.

I'll of course bring to the attention of the customer that this is not a scenario covered by any standard I've read. I'll run it by my managing PE first (along with everything else in the study).

It seems to me this would be a pretty low-percentage chance for the most part. But a few of the busses at this location have pretty high fault levels, and I suspect they could spread pretty easily.

I appreciate the discussion, everyone.

 

[redtailET]

Case 3 is very possible. The fault from one side can ionize the air and cause a separate fault on the other side. You would have two separate faults taking place on the open tie.