On Jan 31, 12:40 am, Roy Lewallen wrote:

....

In reality, any transformer will usually behave reasonably well only
when terminated in an impedance near its design impedance.

And, I might add, in the frequency range for which it's designed.
That may be obvious for something like trying to apply an audio
transformer to RF balun service, but it may not be so obvious if you
just pick up a "ham" 4:1 balun and think it's going to work for your
particular frequency range, if it's not designed for that range.

You'll _probably_ have better luck using a balun designed for, say,
50:200 ohms over 1-50MHz with a 3000-j4500 load at 20MHz than if you
try the same load impedance at 2MHz. That's because the transformer
reactance must be high enough at 1MHz to work at 50:200 OK, so there's
at least some hope the reactance will be enough larger at 20MHz to not
screw things up too badly. You will also want to watch out for core
saturation: run at lower power if the load is higher impedance. You
wouldn't want to push the balun too far toward the high end of its
frequency range, either, because there may be stray capacitances
limiting the response at the high end.

As Roy says, it's hard to say what your particular balun will do when
presented with a load so far away from the (presumed) 50:200
impedances it was designed for, unless you measure it with that load.

Cheers,
Tom