> 50 microhenries of plate choke with the shunt plate tuning capacitor, and
> the capacity of the loading capacitor in parallel (as it is connected
> through the long lead that is the tank coil), bearing in mind that if the
> load is not a resistor, but an antenna, at low frequencies, it will tend
> to look like a high impedance.
>
> We are not talking about 'choke resonance' where the choke with its
> inherent distributed parasitic componets of reactance resonates, but of
> choke resonance where the choke is tuned by the other discrete components
> in the circuit.
The units I measured had a parallel anode resonance (with the
output port terminated) of over 10,000 ohms somewhat below the
AM BC band.
The circuit involves the blocking capacitor, the plate choke, the
shunting capacitance below the plate choke, and the additional
choke in series with the power supply lead. The power supply
looks like a near-perfect short at LF because of the electrolytics
and other components. The reactance of the blocking cap parallel
tuned the complex network of the two chokes in series, with a
capacitor shunting the mid-point between the two chokes.
By adding series inductance to the output port, the amps I tested
would take right off at VLF.
Loading the grid chokes with a 1000 ohm resistor stopped the
problems, but the best bet is to remove the parallel resonant circuit
on the grid.
Placing unloaded chokes in grid leads is very poor engineering.
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
|