[Amps] tuned input network design info

jim.thom jim.thom@telus.net jim.thom at telus.net
Thu Apr 27 12:56:45 EDT 2023


Date: Thu, 27 Apr 2023 05:19:50 +0000
From: John Keating <john.keating at outlook.com>
To: "Amps at contesting.com" <Amps at contesting.com>
Subject: [Amps] tuned input network design info


<If anyone knows of a text with detailed design methodology for tuned input
circuits <for tube amps, please send me a reference.
<Looking for procedure and rationale, not a calculator.

<Thanks,
<John
<AI6LY

There isn't any, and I have looked for years on end.  And orr's very last
handbook shows a chart with various values for different input Z's....and
it's not even close to reality.

I'm talking about a simple PI  net for a tuned input for a high MU  GG tube
amp.

I use the GM3SEK  PI spreadsheet instead.   But a caveat here is in order.
It won't handle 2 x identical Z's, like 50 ohms in..and a 50 ohm tube input
Z.  The 'fix'  is to alter one value.... then it all works.    IE:  50
ohms  to 49.999 ohms.

The GM3SEK pi spreadsheet was for tank output.    Where u stuff in the tube
plate load Z, u instead insert  50 ohms. Insert 9999 uh for the plate
choke. zero everything else out.

On the spreadsheet, you can alter the..'total network Q' .  Down below, it
then spits out the input Q..and also output Q..and these  2 values totalled
always equals the total network Q.   IE: total network q of 4.   And an
input and also output Q of '2' ..but that's for a 50 ohm tube input Z.

You also have to factor in the tubes input C..and also the C from the
cathode end of the fil choke.  Each lead to chassis...( and bypass caps at
cold end of fil choke temp removed).   All that C has to be absorbed by the
calculated  C2 value...IE: less actual C2 required.

U also have to factor in distance from C2 cap to cathode, esp on upper HF
and 6m.

Most books do the coupling cap inputs wrong.  Ideally you wire 2 x caps in
series..and  wire that mess directly across the cathode.  Then the dive is
applied to the junction point of the caps.  Then each cap only handles 1/2
the applied drive current.

On serious qro applications, multiple caps are paralleled 1st.   IE: 2-4 x
.01uf or 4700 pf disc caps are paralleled..and call that assy 'A'.    Then
repeat for...'assy B'.  assy  A+B wired in series..and again, that mess is
wired directly across the cathode....and again, drive is applied to the
junction point.

I design and build a ton of these PI tuned inputs over the years, and still
do on a weekly basis.

On the bigger hb amps,  I gave up on designing 9 x separate, bandswitched
or relay switched tuned inputs.  I use a manually tuned, PI tuned input,
with either a tapped 4 uh coil, or a multronics 0-4.4 uh roller coil.
 Then 2 x broadcast caps for C1 and C2.... and padded on 160m.   Both caps
have 6:1 verniers.  Then it's dead flat swr from 160-10m..and any freq in
between.

Then Q can be optimized better for any band.   You need a lcr meter, and
k6sti's COIL program to construct any of this right.

Jim   VE7RF


More information about the Amps mailing list