> Maybe a combination of air wound choke on secondary and
> a floating
> filament transformer would suffice. Use enough choke
> between tube & filament
> transformer secondary to server for maybe 14 MHz to 30 and
> let the floating
> transformer arrangement serve best below the 14Mhz. This
> helps to eliminate
> some cost by eliminating the ferrite rods and the worry
> about saturating
> them.
Hi Larry,
The reactance of the filament choke is in series with the
stray capacitance of the transformer to ground.
Let's say you have 250 ohms inductive reactance from the
choke on 14 MHz, and 250 ohms capacitive reactance from the
transformer's stray capacitance on 7 MHz. Our plan was for
one to "take over" as we change frequency. Let's assume the
reactances change at a linear rate with frequency (they
never do in large components, but we will assume that
perfect world).
The problem is the series reactances cancel each other. On
14 MHz the inductive reactance would be +j250 ohms (the
choke) in series with -j125 ohms (the transformer). Now we
only have +j125 on 14 MHz. We don't have our original 250
ohms until we reach 5 MHz or 19 MHz. We've made a hole in
impedance that reaches zero ohms at 9.9MHz. It can even get
worse than this because of transmission line effects in
physically large components.
It's really much easier just to wind a proper filament choke
for broadband, or if we only operate low frequencies just
use a low capacitance very well insulated transformer with a
fully bypassed primary.
By the way, my first experience with this was trying to do a
filament system in a 3CX10000A7. The peak RF voltage in the
transformer got so high it arced over from primary to
secondary, connecting the power line hot lead to the
filament. After I ruined the transformer, meters, and bias
circuit I thought about the problem and reached the
conclusion it was a bad idea.
73 Tom
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|