There is an interesting glitch with the IEEE 1584 equation for arcing short circuit current at or above 1 kV.

The equation for arcing current at higher voltages > 1 kV:

Log Ia = 0.00402 + [0.983 (Log Ibf)]
Ia = 10^(Log Ia)

Where:
Ibf = 3 phase bolted short circuit current in kA
Ia = 3 phase arcing short circuit current
IEEE range of validity 700A to 106 kA

Here’s the glitch. If the bolted current is 1,000 Amps i.e. 1 kA:

Log Ia = 0.00402 + 0.983 X Log (1)

Log of 1 = 0
Therefore:

Log Ia = 0.00402 + 0.983 X 0
Log Ia = 0.00402 + 0
Log Ia = 0.00402

Ia = 10^(Log Ia)
Ia = 10^0.00402 = 1.0093 kA

So….. Ibolted = 1,000 Amps and Iarcing =1,009.3 Amps

Except it is physically impossible to have Iarcing greater than Ibolted.

Oops!

Not very surprising since the IEEE 1584 equations were curve-fitted from the actual data. Some points will be over or under what's physically possible.

It somehow reminds me of 2 AA batteries of mine which, after use, had a negative voltage between the ends (-0.2 V and -0.3 V). I think they gave me too much energy, and now try to get some of it back. Not going to happen...

The glitch is only relevant for bolted faults < 1.734 kA. This probably won't affect many high incident energy locations.

I am sure you are quite correct. I can't imagine that low of a current even sustaining easily.

Just thought the little(?) hiccup was interesting.

Hi Jim, I came up with a similar scenario for voltages under and approaching 1000 volts about three years ago. I said at the time "how can you get an arcing current which is greater than the bolted current. You can easily get arcing current which is 50% higher than the bolted short circuit current value. WDeanT kindly posted the following link which is very interesting. The perils of curve fitting equations!!!


http://us.ferrazshawmut.com/arcflash/getting_help/downloads/ICPS-wilkins011904_web.pdf


Original thread
http://www.arcflashforum.com/showthread.php?t=74

Hi Mike,

Forgot about that post. Good info!

Thanks!