Eddy and all, here are some old and well known facts and thoughts on too
much output C.
Eddy already suggested the biggest help; a series L between the plate and
tune C. In addition, the tube does not care what the loaded Q of the
matching network is. It will be just as happy with a Q of 24 as a Q of 12.
Remembering the network efficiency is (Q loaded - Q unloaded) divided by Q
unloaded, the key to efficiency is Q unloaded. The higher the Q unloaded,
the higher the efficiency. Meticulous attention to the 10m resonating
inductance is the key. While 1/8" diameter coil material sounds adequate, even
larger diameters will make for higher Q unloaded and associated losses.
I have never studied the optimum coil dimensions for highest Q in this
application, perhaps someone else here has done that.
I have a homebrew 1.5 KW amplifier for 1296MHz that runs with a loaded Q
of about 300. Total class AB2 efficiency is on the order of 52% and that
takes a Q unloaded of about 1500. The plate resonating inductor is a 10 inch
square by 3/4" high hollow box with the tube mounted in the center. Loaded Q
was determined by varying the drive frequency while peaking the input
circuit and watching for a 3dB reduction in output power.
Cool stuff!
73,
Gerald K5GW
In a message dated 3/28/2011 2:35:37 P.M. Central Daylight Time,
deswynar@xplornet.ca writes:
Hi Roger,
Nice note---thanks!
Yes, the 813 has cost advantages, alright---but you have to jump through
some hoops in order to make them "play" on 15- and 10-meters efficiently,
because of the high interelectrode output capacitance of them (ex. on
10-meters, a paralleled pair display a total of, I believe, 28-pfd., which
is actually more than the plate tuning capacitor value needed on that
band!).
Makes you wonder how that fellow you mentioned got away with FOUR in
parallel! (maybe his was strictly a low-band design...?).
What I did here with mine is employ a 25-uh. 1/4" edge-wound copper ribbon
variable inductor as a pi tank coil...and I incorporated a rugged ceramic
Centralab "bandswitch" that allows me to REMOVE the plate tuning cap out of
the circuit entirely. I then RESONATE the final tank on 15- and 10-meters
by
varying the "L", and using the high interelectorode "C" as a "built-in"
capacitor.
For 10-meters, though, there's another step that needs to be taken: I
switch
in a 0.3-uh. silver-plated 1/8" copper coil in series with the paralleled
plates, to act as an RF transformer of sorts. It's all detailed in my 1985
ARRL HANDBOOK, which references the issue of QST that the idea came out
of...
It's a bit of an exercise, to be sure, but the work-out entails one's GRAY
matter, and NOT one's pocketbook! Hi Hi. I think it's worth it.
~73~ de Eddy VE3CUI - VE3XZ
*******************************************
----- Original Message -----
From: "Roger (sub1)" <sub1@rogerhalstead.com>
To: "Diane & Edward Swynar" <deswynar@xplornet.ca>
Cc: <amps@contesting.com>
Sent: Monday, March 28, 2011 2:04 PM
Subject: Re: [Amps] Homebrew is back..
> On 3/28/2011 3:38 PM, Diane & Edward Swynar wrote:
> > Hi Roger et al,
> >
> > I use a pair of 813s in parallel grounded-grid service here in my
homebrewed
> > KW amplifier, and---I'm blushing as I type this!---but I have NEVER
EVER
> > purchased any of the jugs that are in it, or the spares that I have
here,
> > brand new...
> >
>
> I forgot the old 813. I have 3 of them up on the shelf and keep looking
> for more.
> I used to run the "old legal limit" with a pair of them as you
> describe. I ran into a guy on 40 a while back who claimed he could get
> 2400 PEP out of 4 of them. I don't know if his amp was a monoband or
> covered all of the HF bands.
>
> I see the GS-31 (1KW), GI-46 (1KW), and little GI-7B are at reasonable
> prices and seem popular. With the removable anode coolers on this
> series and the popularity of converting the old Dentron MLA2500B to a
> pair of GI-7Bs that no one has machined a transverse version of the
> cooler to make them even more compatible with the old 8875s I think they
> used.
>
> > They are all either Hamfest finds, or freebies from friends.
> >
> > Life's far too short to pay-off a retiree's ransom for new& exotic
> > tubes---
>
> AMEN!
>
> > and I have yet to hear anyone at the DX end of my QSOs asking me (or
> > expressing any care whatsoever) which tube(s) I might be using! Hi Hi
>
> There does seem to be a lot of interest in what tube(s) are in homebrew
> amps though.
>
> 73
>
> Roger (K8RI)
>
>
> > ~73~ de Eddy VE3CUI - VE3XZ
> >
> >
> > **********************************************
> >
> > ----- Original Message -----
> > From: "Roger (sub1)"<sub1@rogerhalstead.com>
> > To:<amps@contesting.com>
> > Sent: Monday, March 28, 2011 1:11 PM
> > Subject: Re: [Amps] Homebrew is back..
> >
>
>
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|