> From: Rich Measures <measures@vc.net>
To: <amps@contesting.com>
> Date: Tue, 12 Aug 97 04:30:43 +0000
> >range.......for a the ETO Q of 5.3 into 50 ohms..... and pi- tank
> >inductor fixed at 21.57 uH and L inductor fixed at 7.255 uH.....
> >
> >760 ohm load, Cp 600 pF, ...
> ...snip...
> The output R (RL) of the pair of 8877s is approx. 3750V/(2A*2)=938 ohms.
Earlier you gave 1000 ohms, I used that figure.
> Since Q is defined as RL/XC1, and Q is given as 5.3, the reactance of C1
> (which Mr. Rauch refers to as 'Cp') would be 938 ohms/5.3 = 177 ohms. At
> 1.9MHz, C1 is 477pF.
>
> - 600pF does not check out.
It certainly does. This operational example is for a "Q" of approx
5.3 into a 50 ohm load (using your Rp value of 1000 ohms), and gave
the new values of C required to match the PA as the load was changed
without changing the inductors in the tank. (You might take issue
with this... but I think that is how most PA tanks are tuned... by
only adjusting the C's.... with the exception of Henry's.)
The Pi-L has a very wide matching range, even if initial "Q" is set
fairly low. The ONLY time anyone would notice low efficiency is if
the caps simply ran out of adjustment range, or the Q got very high.
On 1.9 MHz with only 500 pF of C at the tube, the pi-L would match
22.8 to 325 ohms. With 600 pF, 8.5 to 760 ohms...all with a design Q
of 5.3 into 50 ohms at 1000 ohms Rp.
My point is this.........
Proper range of components is much more important than this apparent
"fixation" on Q, unless harmonics are a problem. Most PA's with low
efficiency work poorly because they have a cap or inductor that is
out of range (preventing a conjugate match), not because the
operating Q is "off".
ARRL Microsmith is VERY handy for looking at tank performance. Make
the load equal to the tube operating Z, and plug in the values of
tank components. You can vary the values of the adjustable components
and watch the impedance move on the Smith chart. Wherever the
displayed impedance can be moved, the tank will match that load
value.
Tank circuits aren't so mysterious they should be Voo Doo,
especially with all the Smith Chart and transmission line programs
floating around.
73 Tom
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|