Right you are Karl Arne.
Using any type of coax between the output side of the network and the tube
cathode is an open invitation for trouble. At harmonic frequencies that piece
of coax can look like a quarter wave matching network ! Depending on the
frequency and length of coax involved.
In amplifier design component placement is the name of the game and the input
network is no exception. Mount that network as close to the cathode as possible
and use a piece of copper strap or heavy coax (RG213) outer conducter braid to
attach the two mindful to keep the lead length short.
I went to school with a ham that later became a design engineer for RL Drake.
Among his responsibilities was the design of the L75 Amp. I had a conversation
with him 6 years ago and he spoke of some of the obstacles he had to overcome.
One was the problem of Low Output and poor efficiency on 10 meters. A work
around that he found was to remove the output cap on the input network
bandswitch and attach it directly to the tube cathode. What that accomplished
was to add the stray inductance of the connection lead to the input inductor.
Then it became a simple matter to reduce the inductor size for proper
operation. On the lower bands the output capacitors on the switch were removed
and reduced values installed to reflect a total network capacitance of the two
combined to reflect the proper value. The only drawback to this scheme is the
stray inductance then becomes a second inductor forming a "Double Pi" on the
lower bands. But he found that even on 15 meters the stray inductance
was low enough to have minimal effect. In effect what this approach did was
to transfer the problem to a frequency where it was much more managable to deal
with.
73s
Bob W6AH
-------------- Original message --------------
From: Karl-Arne Markström <sm0aom@telia.com>
> The use of a long coaxial line between the driver and the untuned input of a
> GG amplifier has other drawbacks.
>
> For the fundamental frequency component, the line may appear as a shunt
> capacitance,
> but with increasing harmonic numbers the effects
> of the coaxial line electrical length and its termination
> may result in high equivalent series impedances in series with the RF plate
> current flow.
>
> 73/
>
> Karl-Arne
> SM0AOM
>
>
> ----- Original Message -----
> From: "Gary Schafer"
> To: ; "'Tony King - W4ZT'" ;
> "'AMPS'"
> Cc:
> Sent: Sunday, January 22, 2006 4:59 AM
> Subject: Re: [Amps] Tuned Input
>
>
> > Could you elaborate a little more on this?
> >
> > Why is the efficiency less when a coil input is at the cathode?
> >
> > Also wouldn't "too much coax" provide extra capacitance to ground?
> >
> > I assume that the harmonic that you're talking about is second harmonic
> > energy? Why do we care if it is attenuated or not other than IM
> > performance?
> >
> > Thanks
> > Gary K4FMX
> >
> >
> > > -----Original Message-----
> > > From: amps-bounces@contesting.com [mailto:amps-bounces@contesting.com] On
> > > Behalf Of George badger
> > > Sent: Saturday, January 21, 2006 12:08 AM
> > > To: Tony King - W4ZT; AMPS
> > > Cc: skipp025@yahoo.com
> > > Subject: Re: [Amps] Tuned Input
> > >
> > > I have been using an antenna tuner as an input circuit
> > > for my HB 8877 lineart for years. When I was at EIMAC
> > > I learned that when linear amplifier OEMs complained
> > > of low efficiency it was often caused by too much
> > > inductance in the input cathode drive cicuit. For
> > > example, a T input circuit (LCL} was a disaster on 40
> > > and 80. Similarly, if the lead from the input pi
> > > network to the cathode is too long, a problem is
> > > created on 10M. The reason is that the cathode current
> > > is not sinusoidal so harmonic current must have an
> > > easy path to ground. Unless there is sufficint
> > > capacitance to ground, efficiency suffers.
> > > To avoid this on my linear,I experimented with
> > > a variable capacitor to determine the largest fixed
> > > capacitor I could get away with and still be within
> > > the range of my drake tuner on all bands The
> > > capacitance turned out to be 50 pfd. It is connected
> > > directly from cathode to ground with short leads.
> > > W6TC
> > > --- Tony King - W4ZT wrote:
> > >
> > > > Ian White, GM3SEK wrote:
> > > > > Roger D Johnson wrote:
> > > > >> Tony King - W4ZT wrote:
> > > > >>> I don't know if any of you have looked into this
> > > > particular product but
> > > > >>> at first glance it appears that this is one
> > > > solution for the tuned input
> > > > >>> problem many of us face and at the right price
> > > > too:
> > > > >>>
> > > > >>>
> > > >
> > >
> > > > >>>
> > > > >>> 73, Tony W4ZT
> > > > >> The purpose of the tuned input is to provide a
> > > > "flywheel" effect to
> > > > >> smooth out the variations of amplifier input
> > > > impedance over the
> > > > >> operating cycle. The Q of the circuit determines
> > > > how much "flywheel"
> > > > >> effect we have. The lowest recommended value I've
> > > > seen is 2. Solid
> > > > >> state transceivers want even more...perhaps 3 to
> > > > 5.
> > > > >>
> > > > >> The LDG tuner is an L network in which the Q is
> > > > dependent on the ratio
> > > > >> of impedances to be matched. For normal input
> > > > impedances, the resulting
> > > > >> Q will be very low. It's hard to write the
> > > > formula in text format but
> > > > >> in words it's as follows: divide the higher
> > > > impedance by the lower,
> > > > >> subtract 1 and then take the square root. For
> > > > example to match an
> > > > >> impedance of 100 ohms to 50 ohms, the resulting Q
> > > > is only 1!
> > > > >>
> > > > >> With a 3 terminal network you can choose the Q
> > > > independently of the
> > > > >> matching and would seem to be the way to go.
> > > > >>
> > > > >
> > > > > The recommendation for an input tuned circuit Q of
> > > > 2-4 came from a 1961
> > > > > article by Eimac authors in QST. That
> > > > recommendation was specifically
> > > > > based on a pi tank, and it was also pointed out
> > > > that the output
> > > > > capacitor provides a direct shunt path from
> > > > cathode to ground for
> > > > > circulating harmonic currents. But when an
> > > > L-network is being used in
> > > > > impedance step-down mode, it doesn't even have an
> > > > output C.
> > > > >
> > > > > Now it may be that the designers know all about
> > > > these things, have
> > > > > checked their effects on amplifier IMD
> > > > performance, and have found some
> > > > > reason to ignore the Eimac recommendations. If
> > > > that is the case, it
> > > > > would be good to hear why.
> > > > >
> > > > >
> > > > >
> > > > > 73 from
> > > > > Ian GM3SEK
> > > >
> > > > These are very good points that raise questions that
> > > > must be answered
> > > > before I would consider using such a tuner as the
> > > > input circuit. I'm
> > > > with you Ian, I'd like to see the answers to the
> > > > questions.
> > > >
> > > > It's been said many times that the best input
> > > > circuit is a pi-network
> > > > which will provide the fly-wheel effect to help
> > > > smooth the dynamic
> > > > changes in input impedance. As Skipp025 said, on
> > > > another list, "The
> > > > automated antenna tuners might tend to hunt around
> > > > looking for the best
> > > > match." If that happens, we've just moved the
> > > > problem from the
> > > > transceiver to the amp.
> > > >
> > > > Bill, WA4LAV, has an interesting suggestion about
> > > > shunting the cathode
> > > > with a capacitor and I wonder what everyone thinks
> > > > of that?
> > > >
> > > > 73, Tony W4ZT
> > > >
> > > >
> > > > _______________________________________________
> > > > 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
> >
> >
> > _______________________________________________
> > Amps mailing list
> > Amps@contesting.com
> > http://lists.contesting.com/mailman/listinfo/amps
> >
> >
> > --
> > No virus found in this incoming message.
> > Checked by AVG Free Edition.
> > Version: 7.1.375 / Virus Database: 267.14.21/236 - Release Date: 2006-01-20
> >
>
>
> --
> No virus found in this outgoing message.
> Checked by AVG Free Edition.
> Version: 7.1.375 / Virus Database: 267.14.21/236 - Release Date: 2006-01-20
>
> _______________________________________________
> 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
|