Re: Recent s.s.t.c work

["Tesla list" <tesla@xxxxxxxxxx>]

Original poster: Steve Conner <steve@xxxxxxxxxxxx>

> a) 10 V peak across that 1 ohm resistor
> b) 25 V pk across the homemade 4 m-ohm
> c) 25 V peak across the 100 mV/50 A shunt
> All of the above cannot be true at the same time.

I wouldn't ever trust the signal from a resistive
shunt at high frequencies, because of common-mode
effects and L*di/dt. When there's a lot of I, even a
tiny amount of L (or M for that matter) causes a di/dt
signal big enough to swamp the wanted I*R signal.
Also, the common-mode performance of diff amps is
never that good at high frequencies.

The overall result is that measurements with shunts,
especially floating ones, tend to exaggerate pulses
and spikes and make them look a lot bigger. You can
use shunts successfully but you really have to know
what you're doing. You need to use low-inductance
shunts such as tubular or hairpin designs, and arrange
the signal wiring very carefully to avoid mutual
induction between the power and signal circuits. There
is a whole chapter in Craggs and Meek on shunts for
measuring high current impulses.

It seems easier to get a good result with
ferrite-cored current transformers or Rogowski coils,
so that is what we all use on our pulsed SSTCs.

Steve Conner
http://www.scopeboy.com/