[Amps] Figuring out tube output impedance

Xmitters@aol.com Xmitters at aol.com
Mon Feb 27 14:14:45 EST 2006


In a message dated 2/27/06 12:42:21 PM Central Standard Time, 
amps-request at contesting.com writes:

<< I see a lot of  formulae(is that plural???) on determining the output 
impedance of a tube.
 some say just plain plate volts divided by plate current, some have a 1.8 x 
formula  and so on. I Want to determine
 this gu-84b impedance so I can start work on the tank setup. I run 3.9kv and 
I'm GUESSING 2 amp.
  I stuck a 2k in there and played some with the values so far and things 
look ok, no suprises. this is
  my motive for the question. lets hear some opinions! thanks as always.
  >>

Chuck,

There is a real good way to calculate tube parameters that will give the best 
accuracy available on paper. That is, to use the Eimac Tube performance 
Computer (TPC) clear plastic overlay, a set of constant current curves, and a 
properly thought out operating line for the power output/performance desired. The 
Eimac TPC actually samples the instantaneous currents at 15 degree intervals. 
What this does that the magic constants don't do, is it takes into account the 
actual tube characteristics of the tube you are using. The result is a set of 
more accurate parameters.

The "magic" constants are based on a piece-wise sinusoidal plate current flow 
and an integration to calculate the probable DC and RF components of the 
plate current. A really stinko mathematical approach _IF_ you are looking for 
accuracy. The RCA Transmitting Tubes TT-5 has a nice write-up about how to use 
these approximations (their K values) and they are, in fact, wonderful for doing 
quick approximations.

But if best-accuracy on paper is what you're after, then the Eimac TPC and 
associated formulas is the only way to go, short of real world measurements on a 
working prototype. The Eimac TPC and instruction note is free for the asking 
at Eimac. No serious amplifier designer should be without one. BTW the Eimac 
TPC works for everyone's constant current curves. So you can use non-Eimac 
tubes and data sheets so long as that data sheet includes constant current curves. 

You _cannot_ use the Eimac TPC with plate family characteristic curves. 
Whether you use Eimac or "constants" depends entirely on the required mathematical 
model accuracy for the system you are analyzing. My choices are as follows:

For approximations, RCA K values and other "constants" is the way to go.
For circuit design and target operating parameters, it's the Eimac TPC, hands 
down.

Both algorithms have their place and work wonders when properly applied.



Jeff Glass, BSEE CSRE
Chief Engineer 
WNIU WNIJ
Northern Illinois University


More information about the Amps mailing list