> The very popular Fair-Rite 43 is not immune from
> magnetization, as I have tested it. I suspect
> 73 is at least as bad. The issue comes up for
> me because I use DC in my system to switch relays.
> If the relay DC passed through the transformer due
> to a short or wiring error, I can't just fix the
> problem and ignore the possible core damage.
I just tested three core styles. One was a small bead recommended for choke
baluns, the other two were 43 material and 73 material binocular cores used
in my transformers.
The most interesting thing was I grabbed a bag of 52 cores from a source
OTHER than the factory direct source, and the cores inside were not even the
73 mix labeled on the bag!!! They were 43 material that were mislabeled 73!
They tested identical in every aspect to 43 material cores. This is probably
the most important thing I learned. **Never trust a distributor to keep
things right, no matter how "official" the labeling looks.**
Here's what I found when testing at 3.6MHz:
#43 small beads (like those that fit over RG58 or RG400).
initial .4R 90j, after 12a/turns .6R 110j, after 50a/turns .7R 112j
#43 binocular
initial .9R 203j, 12a/turns 2.8R 246j, 50a/turns 2.3R 280j
#73 binocular
initial R911 355j, after 12a/tu 909R 358j, after 50a/tu 890r 340j
I can't claim to be within a fraction of a percent, but these readings
should be reasonably close because the measurements were on a current
calibration certified Agilent vector network analyzer. There also will be
some minimal connection impedance involved that I did not zero out. I was
mostly interested in a "before/after" test.
The 12a was a steady state dc with sharp rise and fall (a normal SPST
switch). The 50A was a 1/2 second pulsed waveform.
I'm sure every style of core would test differently, even if using the same
basic material. But overall I have to say I just don't see a problem in
normal applications for these cores, at least not in the impedance curves
from 300kHz to 100MHz. I've never noticed one in real life either, not to
say it couldn't happen.
Now the inductance does all go to pieces when I pass dc through the cores,
while the dc is on. I wouldn't run unbalanced high current dc through
windings.
73 Tom
_______________________________________________
Topband mailing list
Topband@contesting.com
http://lists.contesting.com/mailman/listinfo/topband
|