Hi Pete,
> OK, I'm confused. I'm building an electronic bias switch and vacuum relay
> QSK into my SB-220, having finally been jarred out of procrastination by
> the failure of the changeover relay this past weekend. In the process, I
> have become aware of (apparently) wide variation in the bias voltages that
> various modifiers recommend.
>
> To recap, the standard bias source is a 5.1 volt zener. Harbach uses 8 x
> 1N4005s, back connected, to produce 6.8 volts. One mod I've seen uses an
> 8.2V zener for SSB and a 6.8V for CW, but the designer says that the 6.8V
> makes the 3-500s too hard to drive, so he just uses the 8.2V position all
> the time and runs in SSB position.
>
> If the 5.1 volt original bias produces Class B operation at the higher SSB
> plate voltage, what does 8.2V produce? Any difference in IMD, one way or
> the another? What about for CW?
'
In a grounded grid PA, the output and input are in series. That is
what reduces gain, because it adds heavy amounts of negative
feedback.
With so much negative feedback, there is very little difference in
IMD performance with changes in bias as long as the bias source
is stable (at the envelope change rate, not the RF cycle by cycle
rate).
This is also the reason why cathode driven systems often greatly
out perform grid driven systems in IMD performance.
The ability to adjust bias has more to do with in-shack heat control
than IMD performance in a grounded grid PA, so long as
conduction angle remains substantially more than 180 degrees.
Typical solid state exciters are in the range of -25 to -30 dB third
order IMD (standard test, not ARRL below PEP test), while a
typical cathode driven tube PA with high-mu triodes is in the -35 to -
50 dB third order IMD. Obviously exciters are the biggest problem
when a GG amp follows the exciter, unless you use a G2DAF bias
system or something equally silly.
Changing bias on a 3CX1200 from 7.5 volts to 15 volts at 3600 volts
on the anode only changed IMD about 1 dB in tests here, lowering
some products while slightly raising others. I expect the 220 would
be about the same.
By the way, the SB-220's IMD performance will become better as
tuning is adjusted and also from band to band if you get rid of the
capacitors and chokes on the grids by grounding the grids directly
to the chassis. It will also become more stable.
When the grids are floated with that weird system, tuning and
operating frequency changes the phase of the feedback making the
amp behave like an un- or poorly- neutralized PA.
73, Tom W8JI
w8ji@contesting.com
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|