Hi Mauri,
We both agree that tetrodes must have the screen properly
bypassed and grounded for stability. I think anyone with a good
understanding for how amplifiers work would agree with that.
> if I remember correctly, the inherent feedthrough capacitance for a
> 4cx1000 in a common grid configuration is less than 0.01 pF, formerly
> 0.00 pf. In my experience the 4cx1000 stability problems are in the
> grounding of screen.
The 30L1 and other amplifiers have a problem in that they "float" or
partially float the control grid from ground. Worse yet, it is a triode
where the single grid is only shield between the highest voltage
part of the output and the input!
If you add any capacitive reactance that has appreciable reactance
at HF between the grid and chassis, it has the effect of reducing
grid shielding on ALL frequencies except the one where the grid
lead inductance and that capacitance are series resonant.
It makes the amplifier more critical to tune and operate, and more
difficult to stabilize.
Not only that, the grid-cathode impedance varies wildly with RF
drive. During some portions of the RF cycle the grid-cathode path
is non-conducting while in other parts of the cycle it is biased on.
Since portions of the RF cycle force the grid-cathode impedance to
a very low value (remember we see a time-averaged impedance
when driving the tube, and so impedance at portions of the RF
cycle is much lower than what we see at the cathode) the
feedback is non-linear and varies with drive power.
Floating the grid through a low-value capacitance is not a good way
to add "negative feedback". As a matter of fact, it makes about as
much sense as the FL-2100 neutralizing system I saw on that
schematic that was posted.
73, Tom W8JI
w8ji@contesting.com
--
FAQ on WWW: http://www.contesting.com/FAQ/amps
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
|