I have a 400A I-Line panel which calculates out at 0.3 cal/cm2 base on its upstream protection. A 100A Type FA breaker in this panel feeds a 51 kVA drive isolation transformer (460:460) which drops the bolted fault current on the secondary side to 1050 A (1 kA). The software equations tell me this will result in 6.7 cal/cm2 at the little drive since it will arc for over 2 seconds. Maybe I've watched Mythbusters one to many times but I do not think the physics would match the equations at this low end.

The cheapest solution is a transformer bypass mode for maintenance but you can never be sure people will remember to turn these things back. OK having them suit up in HRC 2 gear might be cheaper but I think this would sound the BS alarm in most practicing electricians (a similar 75 kVA drive has HRC 3 by the math)

Any thoughts out there?

On what basis do you think the physics would not match the equations? The IEEE-1584 equations are empirical - based on arc flash test results.

Just a hunch - but my guess s that they used curve-fitting techniques which are most accurate mid-range and fall apart at the boundary conditions. What these boundaries are and how far the fall apart, who knows. I doubt there is a lot of measured data at 1kA.

I have also seen the math suggest a 60A fused disconnect fed via 6 AWG from 1000 MCM overhead feeders would be HRC 4. I do not believe there is enough copper to expand to that level and that the wire would essentially act as a really slow fuse. I am not endorsing this design but it is quite common in older steel and rail plants.

Yes, IEEE 1584 is a curve fit exercise. There are problems at the end points but possibly not what you described. See here:
http://ep-us.mersen.com/resources/media/articles/WP-Improved-Method-ArcFlash-Analysis.pdf

The "bogus fit" problem shows up on the high end and sometimes on the low end but you can easily sanity check when this happens (Iarc > Ibf).

As to your example, this happens in the real world quite often. As you lower the available fault current, the inverse time curve hands you a great deal of pain. If the upstream protection is a fuse (you did not specify), it likely trips in 1/4 cycle. Let's just say for argument's sake that it's 10 kA available fault current and assume that incident energy is linearly related to current (it's not by just saying). Now you are reducing the bolted fault current to 1050 A, and the arcing fault current is 85% of that or around 850 A. This might be beneath the instantaneous trip of the 100 A circuit breaker in which case 2 seconds is not unreasonable, or 120 cycles. So we've reduced the available fault current by roughly 10 fold but also increased the trip time by nearly 500 times. So some more napkin math suggests that the incident energy is about 50 times. 0.3 x 50 = 15 cal/cm^2 so the 6 cal/cm^2 value doesn't seem so unreasonable any more.

This is one of the little lessons to learn about arc flash...if at all possible, try to set your protective devices so that the arcing fault current is within the device's instantaneous or at least short time trip curve. Otherwise, be prepared to deal with some very large values.

As to the example of using a wire as a fuse, again, not an uncommon scenario. The jumper to medium/high voltage surge protectors is quite often sized intentionally to be used as a "copper fuse". This does not show up in the model either. Sometimes you have to use a little thought when applying IEEE 1584.

The Mersen paper is an interesting read. There is certainly a lot to try to model during a brief span of time. The zero-crossing re-ignition was a concept I had not heard of before.

My point though is that I am still not a believer that a 850A arcing fault will sustain for 120 cycles. The PG&E data is a small sample but seems to support that notion. Is there another sample set which shows low fault currents can sustain that long?

[ATTACH=full]213[/ATTACH]

There is nothing preventing you from using some other value.
It is just there is currently no industry accepted methodology which you can use to support your decision.

If you feel you can defend your position 'in a court of law', go ahead.
NFPA70E purposely leaves the risk and probability portions of Electrical Safe work Practices up to each individual company.

There is definitely evidence for a lack of self sustaining conditions at the lower voltages. Right now 1584 has a cutoff of 240 V fed from 125 kVA or less transformer. Some test work has shown that this cutoff is not low enough. I have seen test data from the DC folks suggesting that for DC arcs alone, they are not self-sustaining below around 50-78 volts (depending on whose data you look at). From wherever this absolute threshold lies as you go up in voltage, it takes progressively less current to heat up the air enough to sustain reignition of the arc in an AC arc (the arc doesn't go through extinction in DC). In fact that is one of the basic principles of circuit breakers...cooling the arc so that it cannot reignite. So someone needs to do a lot of testing to verify where this point happens and what influences it to find a realistic cutoff where there just isn't enough arc energy to get the incident energy up to a reasonable concern. If you have a few thousand or more to spare though, the joint IEEE/NFPA/EPRI Committee would happily put it to good use.

I may see what I can do with a few university and industry contacts regarding testing at this lower end (might be a good PhD dissertation). As industries become more aware of arc flash and its implications we have no trouble convincing them to use racking robots and remote operators on high energy devices. The pushback now is at the low end of the arcing current where customers can't believe current-limiting fuses won't fix everything. We ask them to spend a lot of money remediating designs to cover our bases legally but may in fact be money not needed to be spent in the first place.

Thanks for everyone's perspectives!