Date: Thu, 21 Feb 2013 11:15:41 -0500
From: Eddy Swynar <deswynar@xplornet.ca>
To: "Mike Lord" <k4elv@bellsouth.net>
Cc: amps@contesting.com
Subject: Re: [Amps] 813 efficiency
On 2013-02-20, at 3:18 PM, Mike Lord wrote:
> Some time ago (maybe 2 years) there were comments filed on this site
> regarding increasing the efficiency of the 813 tube on 21 - 28 mhz. I've
> looked through my history of this group but failed to find a reference. I
> also checked some GE Ham News articles that I've saved. I realize the tube
> wasn't designed for this spectrum. However, I do seem to recall some
> references being made on this site based upon personal experiences. I really
> don't intend to waste the groups time, but I sure would appreciate any
> circuit suggestions that would lead to increased efficiency of this tube.
>
Hi Mike,
There was an article in an issue of QST (when it was still of the small-format,
I believe), as well reference made in the large-format HANDBOOKS, regarding
inserting a small coil in series with the output of tubes like the 813 that, in
effect, acted like an "L" network, thus raising the efficiency of the tube
tremendously on the higher bands...
I went through the mathematical gyrations (and they are considerable!) when I
built my 2x813 linear here some 22 years ago: I ended-up inserting a
switch-selectable 0.3 uh. coil in the plate leads right at the 813 plate
caps...a mercury switch (glued atop a popsicle stick!) allows to coil to be
"in", or "out". When it's "in", the amplifier seems to perform every bit as
efficiently on 10-meters, as it does on 160...
Mind you, I incorporated a switch in the plate tuning capacitor, too, allowing
me to switch it out entirely, & use the high interelectrode "C" of the tubes to
resonate on the higher bands, ALL AIDED & ABETTED BY A VARIABLE PI-TANK COIL
that I use here (26 uh. of 1.4" edge-wound copper ribbon).
There's considerable work involved, but it CAN be done.
Barring this, take a look at the scheme used for 10-meters only in the 2x813
amplifier that's featured in the 1968 ARRL HANDBOOK...
~73~ de Eddy VE3CUI – VE3XZ
## eddy, toss the books. Spend 5 mins with the GM3SEK PI software instead.
The L PI is the ticket, but it can take many variations. With higher amounts
of L in the plate lead ( anywhere b4 the tune cap)
you require less L in the main 10m coil..and vice versa. You can see that
effect right away in the software. There is NO reason to have to switch that
coil out of the circuit, why bother ?
## and why would you want to switch the coil in the plate lead at the plate
caps ? You shoulda just put the coil AFTER the plate block caps, and b4 the
tune cap. Then it wouldn’t have any B+ on your switching contacts.
The way you have done it, you have both B+ and RF voltage on it. When driven
full bore, the anode..and your mercury relay contacts...will have almost twice
the key down B+ voltage on it. The RF in the tank circuit
is AC. And AC will back feed through the plate block caps... and onto the
anode..and eveything in its path.
## remove your popsicle stick, mercury relay, etc... and toss em. Insert
your L coil between plate block caps and tune cap. If no room insert in
plate lead. Toss any parasitic suppressor, the amp doesn’t
need it on any band.
later... Jim VE7RF
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|