[Moon-net] 70cm 4cx250b amp question.

Leif Asbrink leif at sm5bsz.com
Fri Nov 7 19:44:39 CET 2008

Hi Jordan,

> It looks like I'll need a big resistor in the grid cavity... I didn't think 
> a wire-wound power resistor would be a friendly thing in there, but at 10w 
> dissipation, do I have an option...? I could parallel some 2w carbons, but 
> that would also end up being a sizeable lot of 5 resistors....
The purpose of the resistor is to prevent the screen grid voltage
from going above 360V in case the current from the screen grid
has the wrong sign. 

When the plate RF swing is low while the plate current is high, the number 
of electrons that hit the screen grid will create an even higher number 
of secondary electrons that will travel to the always much more positive 
plate. The inflow of these extra electrons create the reversed sign.
Under full load, the RF swing will bring the plate voltage below
the screen grid periodically which makes G2 the most positive electrode.
The current pulses thus created are averaged by the G2 capacitor and 
constitute the normal screen grid current in these tubes.

There should be a VERY low RF voltage on the screen grid. Capacitors
should be used (possibly one with a series resonance at the frequency
of operation). The resistor should not see any RF voltage and since
the bandwidth of your modulation envelope should be very small, a couple
of kHz only, the variations in the screen grid current would not
contain very high frequencies. The screen grid current could have
frequency components up to 100kHz or so, not more, so there is no 
reason at all to mount it inside the grid cavity. You may even place 
the screen grid resistor in your power supply unit. Use wirewound
if that is what you have at hand. Just make sure the resistor will 
always be connected to the screen grid whenever there is a plate 

> I'd like to have 10mA on idle, so with 360v it looks like 36K @ 10w will 
> do....
That will give plenty of safety margin:-)

> I'll be sharing a power supply between amps, with separate filament 
> transformers for each amp.... The screen and bias V's can be handled with a 
> rotary switch, but what is a practical method of switching the 2200V B+...? 
I would leave the 2200V connected all the time to both amplifiers.
Possibly I would make the screen grid negative in rx mode.
(Listen to the antenna output from the PA to see if it produces any 
noise in Rx mode. If it does, make G1 or G2 or both more negative. 
Never below -250V on G1.)


Leif / SM5BSZ

Moon-Net posting and subscription instructions are at http://www.nlsa.com/nets/moon-net-help.html

More information about the Moon-net mailing list