[AMRadio] Biasing a Class C modulated final

Bernie Doran qedconsultants at embarqmail.com
Wed Feb 18 12:03:38 EST 2015

Hi Paul: I assume  you are using the 813s in parallel, if so they will not
like going above 20 meters.   I strongly suggest Eimac's " care and feeding
of power grid tubes"   they recommend and I agree that one side of the
filament should be grounded directly to a common point, all bypass caps go
to the same point.  do not use .01 caps for bypass  of the filaments it can
cause a resonate circuit with the filaments of tubes with long leads or
filaments, instead use nothing larger that .002 or so.   I also like placing
a 100 OHM non inductive resister in each screen and control grid lead,
bypassing the end away from the tubes screen with another .002 back to the
common ground. the grid leak resistor is attached to the two 100 Ohm
resistors and bypassed with the .002 cap at the downstream end. do not use a
choke at the grid circuit.  the grid leak also serves as a base line load
and makes the amp less sensitive as do the resistors in the screen circuit.
813s are very easy to drive and like to take off on their own.    it is a
lot easier to do all this first and not have to go back in and root around.
The same with provisions for neutralizing, I can hear people now saying you
don't neutralize tetrodes! well show me a pair of 813s that show a slight
peak in grid current at the exact point of the plate current dip!!

Use  shielded wire for fil, screen, and bias leads.   now this is a little
extreme but next time I am going to run the fil, screen, and bias in
separate small dia copper tube grounded only at the common point. and in
addition to the .002 I would use a 100 to 250 MMF Mica in parallel   Also
strongly suggest a vacuum variable for the plate tuning cap.    It took me a
long time to realize that no matter what you do the circuits have multiple
resonations  the .002 cap has an internal resonance depending on lead length
etc.  The screen grid, control grid and filament basically form a triode.
The plate leads of parallel tubes along with the grid leads  form a push
pull circuit at some frequency,   and on and on.   The National R175  make a
good plate choke for shunt fed circuits.  does not go to 160.    Don't push
the  poor 813s, 2500 is a little high for them but should work, just don't
run at more that 800 watts input.    a bit over 300 MA.  The plate
resistance would be about 4000 Ohms and with about 30 MMF of output
capacitance you are just about the limit for a reasonable tank Q on 20
meters.   Use air on them and heat radiating plate caps and they probably
would last your life time.  Too much air is indicated when you blow them out
of their sockets  or can not hear the receiver!!!

I love 813s they are my favorite tube along with 304TLs, obviously I am an
old Goat. if I insisted on 20 through 10 meters I would go with 4-250As in
push pull.  I only have a pair of 833s  in operation at this time but a mess
of 813s and 304TLs in boxes plus the 4-250s  and for the QRP amp 4-1000s.!
None for sale untill I croak.  I hope you did not fall  asleap reading this!
Bernie   W8RPW
----- Original Message ----- 
From: "Paul Baldock" <paul at paulbaldock.com>
To: "Bernie Doran" <qedconsultants at embarqmail.com>
Sent: Tuesday, February 17, 2015 9:02 PM
Subject: Re: [AMRadio] Biasing a Class C modulated final

> At 01:35 PM 2/17/2015, Bernie  Doran wrote:
>>  if you do have a seperate screen supply use a choke in series and they
>> will self modulate the screens.
> That's the plan.
>>  in the screens, you would need about ten H at 100 ma.    what plate
>> voltage are you going to use?
> Around 2500V
> - Paul
> ---
> This email has been checked for viruses by Avast antivirus software.
> http://www.avast.com

More information about the AMRadio mailing list

This page last updated 21 Feb 2018.