You can put DC on those filaments and read the current drain, after a suitable
warmup time. With that you can use Ohm's law. You need the resistance once
heated.

Of course a resistor will increase warmup time, so keep that in mind if you
want to avoid cathode stripping.

A resistor is not the best method because when plate current rises so does the
cathode temperature. The resistor abates the decrease in current which is
figured in by the designer. Delta temperature is bad for cathodes. Whether this
is significant in your case or not, I don't know.

Actually at this point I'm wondering just how you know the filament voltage is
high. Y'know you could get the supply running into a resistor, well a series of
resistors and just use a coil around one of the feed wires. Picking it up
inductvely should avoid most RF drain.

At any rate if you can get insulated wires through the core of the transformer
you can wind a "bucking" winding. You'll have to see which phase it is, and
then it goes in series with the original winding, but out of phase. This will
give you a solid voltage at a lower level.

Good luck with it.

JURB