Ned Simmons writes:

On Tue, 04 Sep 2007 21:34:16 -0600, Joe Pfeiffer
wrote:

Ned Simmons writes:

Seems like a perfectly good analogy to me. The cord is sized
considering the nature of the load, not the capacity of the circuit
that's feeding it. The welder in question draws about 30A @ 30% duty
cycle.

You're assuming the conditions of setting the duty cycle of the welder
also apply to the duty cycle of the cord. I don't see *any* reason to
believe that's going to be the case. When I see a duty cycle rating
for an extension cord, I'll believe that it applies to it. In the
mean time, I'll use a cord rated for the peak load that it will be
called on to handle.

See the NEC article on welders. It recognizes the intermittent nature
of welding and allows conductors with an ampacity lower than the
breaker on a welder circuit.

You're right (and I'm surprised) -- the supply to a welder can be
lower than called for by the peak current draw by roughly the square
root of the duty cycle. Given that the duty cycle is defined over a
ten minute period, this means you can run double the rated current
through a cord for roughly two minutes! Yikes...

Though that's also assuming the welder doesn't have a specified I1eff
rating -- if it does, that trumps the calculation.