>> > be a significant high RF voltage across the coil. It's peak-to-peak value
is
>> > typically in the range of VCC. For a 4KVDC amp this results in a peak of
2KV
>> > and and RMS of about 1400V.
>>
>> Say WHAT!!?
>>
>> A typical tube amplifier with a 4KVDC plate supply should exhibit
>> something in the vicinity of 3.5 KV peak RF voltage with an RMS figure of
>> about 2.5KV.
>
>Not quite. The normal properly-tuned peak voltage of a near-class B
>amplifier (class AB approaching B) is about 80-90% of anode voltage in each
>direction,
Mr. Rauch -- 80% of 4000V = 3200V. 90% of 4000V = 3600V. What's the
problem with 3500V?.
>which would translate to 3kV or more across the tank inductance.
>The exact value depends on tank loaded Q, which sets phase shift across the
>inductance. Q's of 10-12 result in about 130-degree shift,
And I thought the phase shift in an inductor was 90?.
>so the full p-p
>voltage from tuning cap to ground and loading cap to ground does NOT appear
>across the inductor.
>
>But is is more than 2kV!!
>
>Worse yet, when the PA is mistuned the peak voltage can be almost anything.
>See the practical demonstration at:
>
>http://www.w8ji.com/demonstation.htm
"In a 3000V PA stage, this would be 10,000V peak! It should make sense
that amplifiers arc from grossly excessive drive power." -- Tom Rauch.
- ... but it doesn't. In a worst case test with a SB-220 (a 3000V
stage), the max peak V across the Tune-C was 3500V - which is c. 100V
from flashing over. The idea of overdrive producing triple V does not
make sense since 3-500Zs are emission limited.
>
>You'd better insulate appropriately for worse-case conditions, which is the
>maximum breakdown voltage of the tank tuning capacitor times at least two or
>three or whatever other component limits the peak voltage under a fault
>condiution.
Kenwood used a 6000V Tune C in the TL-922 and the Tune-C does not protect
the 5000V bandswitch from arc-over. Kenwood apparently sells a heap of
922 bandswitches.
>
>> ...
- R. L. Measures, a.k.a. Rich..., 805.386.3734, AG6K,
www.vcnet.com/measures.
end
|