[ARC5] HF Converter - the mixer.

[Tom Lee]

Hi Ken,

Your mention of the Pullen circuit takes me back in time -- thanks! It 
is indeed an excellent mixer, and particularly attractive for upgrading 
classic VT receivers. It certainly exhibits better noise performance 
than do non-triode circuits. It still suffers from spurious responses to 
image frequencies, but the laziness of a tube's characteristics 
(relative to a transistor's) means that the conversion gain to the image 
terms will be quite a bit lower than the primary conversion gain. As 
long as the interference environment is benign, or the receiver is 
preceded by a good preselector, the image problem can be managed well. 
LO feedthrough is still a problem for this circuit (as for any 
single-balanced design), but brute-force filtering handles that.

If one were carrying out a blank-sheet design for maximum dynamic range, 
I'd still elect a more modern double-balanced architecture. But if I 
wished to stay with the hollow state, I'd probably choose the Pullen for 
all but the simplest, low-tube count receivers. And if I wanted to 
upgrade an existing receiver, retrofitting it with a Pullen mixer would 
be a great idea.

Thanks very much for posting those links!

-- Cheers,
Tom

-- 
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
http://www-smirc.stanford.edu

On 3/16/2022 22:54, Kenneth G. Gordon wrote:
> I find it somewhat odd that so few people consider the so-called "Pullen Mixer" for such an
> application.
>
>  From my study of that mixer, it seems to me to be a nearly perfect mixer for this purpose.
>
> I wonder if Dr. Tom would consider commenting on this?
>
> Here are some links to it which I have published on my website for a number of years.
>
> http://www.w7ekb.com/glowbugs/rx/receivers.htm
>
> ...at 9) about half-way down the page...
>
> and
>
> http://www.w7ekb.com/glowbugs/rx/Pullen/PullenMixer.pdf
>
> http://www.w7ekb.com/glowbugs/rx/Pullen/75A4Pullen.pdf
>
> One thing that is not clear from any of the above is that to properly implement this mixer will
> require 4 triodes, not 3, 1 of which would have to be a cathode-follower between the
> triode-HFO and the mixer-proper.
>
> I had planned to use two 7963 submini dual triodes in my implementation of this circuit.
>
> These could be mounted on an octal tube base, with other components, and simply plugged
> into, say, an "ARC-5" receiver in place of the 12K8 for instance. Other parameters would
> have to be "tweaked", of course.
>
> It also appears that the higher the transconductance of the triodes involved, the better this
> mixer circuit works.
>
> Ken W7EKB