Huge Arc Flash Boundaries

[BigJohn1]

The utility I work for recently had an arc flash analysis done for our plants. It was a given that we had equipment that was "Dangerous" and would have large arc flash boundaries.

What I wasn't expecting was to find that there are a bunch of boundaries in excess of 100 feet, and sometimes much more.

Much of this is high-power generating equipment, many MVA, but I'm still having a very difficult time believing that it has the ability to induce second-degree burns across the distance of half a football field. I strongly suspect this is a case of garbage-in-garbage-out, and some of the numbers for the equations were wrong.

Does anyone have any real world experience calculating flash boundaries where they ran into numbers this high? Is it possible for these hazards to be legitimate? Thanks.

-John

 

[alehman]

What's your clearing time?

Alan
“The engineer's first problem in any design situation is to discover what the problem really is.” Unk.

 

[undervoltage]

Primarily it depends upon the bolted fault current, so the higher the bolted fault current, the larger will be the arc flash boundry.

I have worked upto 13.8kV Arc Flash System design. What is your switchgear ratings?

 

[Zogzog]

Those values are common in generation plants. Which is why most utilities in the US have gone to remote racking and switching. Hard for us to give you a detailed answer without seeing the study.

 

[dpc]

It's not unusual to see very long distances like this for high energy situations when using IEEE 1584 equations. I would not assume there is an error in the calculation.

David Castor

http://www.cvoes.com

 

[ScottyUK]

Normally big generating plant is of a phase-segregated design outside of the machine, sometimes including the GSU transformer itself. It is fairly hard to get an inter-phase flashover on a phase segregated design, and normally the system is high resistance earthed to limit damage to things like the stator core so the fault level to earth is relatively low. Where are you seeing these long distances, and are they actually applicable? If you can't get an inter-phase fault then there's not much point in calculating how far away the boundary is for one.


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

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

 

[BigJohn1]

Thanks for the replies. I know y'all can't give definite answers without all the technical data, but I'd just never seen distances anywhere near that great, nor was I expecting anything remotely like that, so I wanted some more knowledgeable opinions.

I'm still amazed, but I'm much more convinced of the authenticity of the results.

-John

 

[BigJohn1]

And to answer your question, Scotty, unfortunately there is definitely the possibility of phase-to-phase faults in a lot of this stuff: Much of it is open air-insulated equipment ranging from 13.8 to 24.4kV.

As far as clearing times, those are all over the map. We have some utility ties that apparently won't clear for almost 2 seconds, which is pretty phenomenal, but that protection isn't under our control.

-John

 

[rmw]

And to think of all the years we walked all around that stuff without even a care and then they discovered it was dangerous.

rmw

 

[rasevskii]

One can think of many open-air substations in the middle of residential areas, that have been there for decades, and now we discover arc-flash boundaries of "hundreds of feet" in some cases...With the substation fence a few tens of feet away from the houses.

I could point out a location where 110KV overhead lines (four circuits) run directly over houses and yards, but the lines were built in 1908 and the houses in the 1920s ...

Lawyer's Paradise.

rasevskii

 

[racobb]

What software was used for the calculations. If it is open air then I would most likely use ArcPro.

Alan

 

[exciting]

Note that the empirically derived equations of IEEE Std 1584 are only valid up to 15 kV, above that the so-called Lee equations are used. These are theoretical values and can easily lead to incredibe incident energy levels and boundary distances (especially with the 2 second clearing times you mentioned).

When comparing both methods around 15 kV it seems a bit silly to use these overly conservative results, but to my knowledge it's the only thing available for industrial applications. NESC tables may apply in your situation giving you a HRC and boundary straight away, but I'm not very familiar with those.

 

[BigJohn1]

Exciting,

I didn't realize that above 15kV results were based on more theoretical calculations. That's interesting to know and probably would explain some of this, however some of these thousand-inch boundaries are for 13.8kV equipment.

Rasevskii,

I did think of the many substations in residential areas, and I believe those are reasonably safe, which is part of why I'm having so much trouble with this. I've seen a lot of high-power equipment go "boom" and I've never seen an electrical incident so intense that it would've caused 2nd degree burns from 50 meters away. You've gotta admit, that's a pretty phenomenal statement, though I do understand these survey results err very much on the side of safety.

-John

 

[jghrist]

The calculations of arc flash boundaries for arcs in enclosures are extrapolations of test results. In reality, when you are far away from the arc, the energy would dissipate more like an arc in open air because the focusing effect of the enclosure would be negligible at large distances. If you fix the "dangerous" arc flash locations, you will eliminate the very large boundaries.

 

[wbd]

A couple of thoughts:

1. IEEE 1584 allows you to use a cutoff time of 2 secs with some other considerations. Usually the default time in software programs is around 1000 secs or longer. This will make the IE large.

2. Is there specific areas where the large AF Boundary is? There may be ways of reducing this with changing trip settings on protective gear or installing a arc flash protective relay.

3. On the substation comments, if you look at the NESC for the arc flash hazard it is a single phase fault. These usually have less energy than a 3 phase fault.

For a lot more information on arc flash, go to

http://www.arcflashforum.com