Gunner wrote:
Any idea of how to calculate the amount of capacitance needed to
reduce idle current in a single phase 220 welder?

Calculating isn't easy. Much easier to add some capacitance and check
current, add some more, keep testing until you find a minimum idle
current.

I believe that for optimal performance you would want SQRT(2*PI/LC) = 60 Hz.

You know everything in that equation except the C. That assumes you know
the effective inductance "looking into" the primary of the transformer.

I think that an LC network looks resistive at the resonant frequency, and
that the resonant frequency is SQRT(2*PI/LC). That's why if you add just
the right amount of capacitance for the inductance of your welder, then
the resonant frequency is the same as that of the power grid and you get
zero imaginary current; i.e. all of your current will be real current.

Real current and imaginary current are electrical engineering terms. Imaginary
current can burn up a wire just like real current, it just doesn't show up on
your power meter.

I used to know all this stuff ..

Grant