> From: Martin Ellis <jmellis@ihug.co.nz>
> Good morning Carl,
> The best looking homebrew 6m linear I have seen was made by
> Jim WA7JCO in Phoenix AZ. He showed me >1.5kw keydown on 6m
> with an 8877. There was no parallel plate capacitor - instead
> he used a vacuum variable as a tunable series capacitor.
> This replaced the blocking capacitor, and by placing a
> capacitor in series with the <large> anode capaciance
> it enabled an efficient tank Q to be achieved.
Bad theory there Martin.
Adding a series C does NOT reduce thank Q, contrary to what Jim
might claim. It actually increases tank Q, because of the extra
stray L and C of the vacuum cap.
What happens is this...
The tube out C STILL remains the tank input C. All the
additional series C of the vacuum cap does is cancel some of the
inductive reactance of the tank inductor, requiring the designer to
add more inductance for the same resonant frequency.
What you really wind up with is a series resonant L/C circuit in
series with used to be the original tank inductance
There are actually two ways to analyze that circuit, and both come
out the same.
Another way is to consider the tube output resistance pulls at
the center of a capacitive divider, with the tube C to ground and the
vacuum cap in series with the load. If you analyze this circuit, you
see the impedance of the anode is stepped UP by the series
capacitance.
This means the input of the remaining L section, the inductance and
the loading capacitance, sees a higher impedance and Q is increased.
When the accumulated Q of all the components is added, it is actually
much worse than Q of a conventional pi-network or double L network
using only the tube's output capacitance as the input C of the tank.
He would actually have had LESS tank Q if he used a conventional
circuit with a flapper plate near the tune anode to trim the tank
in.
> I believe he told me that no suppressor circuit was required
> in this case.
8877's are nearly always unconditionally stable without ANY
suppressor as long as you ground the grid properly. Most coaxial
grid connection tubes are.
73, Tom W8JI
--
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
|