[HBR] Mixer questions and other tech questions..

[Peter Bertini]

On Tue, Aug 19, 2008 at 8:21 PM, Walt Hutchens <waltah at earthlink.net> wrote:


> That means that the 1st mixer almost has to be singled ended, since a
> balanced oscillator output over this wide range is a very difficult
> job.  On 10, for example, stray capacitances will greatly affect the
> amplitude and phase of the mixer drive.  Trying to make this work with
> conventional plug in coils would be a real challenge.
snip
> The advantages of a beam tube (6JH8, 7360, etc.) 1st mixer are very
> high gain if the signal is placed on the control grid and there is
> ample LO injection. However, 'ample' is at least several volts p-p and
> 10 or more would be better. The LO signal isn't rejected at the
> (push-pull) mixer output. That's not too big a deal since the HBR LO
> would be well above the IF and well above or below the signal
> frequency, but it will mean a couple or possibly more spurs to think
> about -- the basic HBR design is almost entirely free of this issue.

> One tweak that will help in the front end is a 6EH7 as an RF amplifier.
> This tube gives far better dynamic range than any of the tubes used in
> the original design, plus higher gain if you want -- the HBR's have a
> separate variable cathode resistor RF gain control. Combine the 6EH7
> with a 12AT7 or (better) a 6922 or 6DJ8 1st mixer in a push-push
> circuit. (Cathodes are in parallel with about a 470 ohm UNBYPASSED
> cathode resistor, one grid driven by signal, other by the LO, plates
> in parallel. Grids return to ground through say 100k each.) This mixer
> is very resistant to crossmodulation AND is semi-balanced against both
> oscillator and signal appearing in the plate circuit. While the gain
> is somewhat lower than other circuits, the RF amp will more than make
> up, especially with the 6EH7.
>

>
> Walt Hutchens
> KJ4KV
> ************************************

Hello Walt, and group..

Thanks for the feedback. What you've described above using the 12AT7
sounds like a Pullen mixer; except the Pullen runs a  reduced plate
voltage on the mixer plate. I've decided to go with 6ES8 dual triodes
in the first
and second mixer stages; the 6ES8 Pullen mixer has been used in 75A4 mixer
upgrades with excellent results. I've decided using balanced mixers would
require too much reworking of W6TC's original design; and reading through the
literature I see even he discovered some issues with the 2nd harmonic mixing
scheme when he went to the combined triode/pentode mixer in later HBR
reiterations.

I do intend to to try the 6JH8 mixer for the PD, I'm thinking a pentode (6U8 or
similar) for the BFO,  leaving  the triode available for  a phase splitter for
balanced LO (BFO) injection. Probably overkill, but I have 6JH8s on hand.

Regarding the 6EH7, I'm getting strapped for filament current on the Hammond
transformer that's on hand for this project :)   Secondly, with that
much gain, I'm
wondering what the ENR would be?  I vaguely remember reading about some
concerns regarding loading of the RF tuned circuits with higher gain devices?
Is that an issue?

The notch filter:  I'm still deliberating this one, since I will be
including a Q-Multiplier,
and adding the second triode for the Notch filter function isn't much work!
The peaking feature would be best for CW selectivity, while the notch
feature seems
to be mostly needed for SSB or AM...  I am not sure if there is sufficient
advantage to having BOTH available at the same time?? It certainly
complicates the IF
design doing so... I'd love some feedback from folks who've been there and tried
both. Or, whether just the Q-Multiplier Peak or Null approach is adequate.

I've been thinking it would be nice to find a few junker HR-10
Heathkits for the
purpose of salvaging two sets of  half-lattice xtal filters to make
good first IF 1600kc
four pole filter for the HBR!!!! That would be the cat's meow. I
suspect the three section
tuning cap might be a good sub for the Millen unit used in the HBR,
but I haven't
checked the parts values to be sure.

Another thought in the back of my mind is that it MIGHT make a lot of sense
for a NEW builder to investigate some of the new AM/FM three section caps that
are appearing surplus.  Using six pin coil sockets (instead of five)  could
give the option of using either the 23pF FM tuning sections for
conventionally wound
HBR coils, while allowing use of the 365pF sections for custom coils that could
cover significant portions of the SW spectrum.

That also raises the possibility of being able to somehow switch to a single
conversion scheme (85kc IF)  just by switching coils, this would allow
BCB and 160 meter
coverage. I think that would be a neat thing to achievet.   Again, a
lot of this
is mental exercise on my part, but I do have 90 percent of the parts on hand.

One more thought: I recently picked up several sections of 1-1/2 and 1-1/4" OD
fiberglass tubing to make my coil forms  (using salvaged tubes for the bases.)
I was wondering if anyone else had tried using this material.  One
final question,
where the heck can you find N750 dogbones in this day and age???? Surplus
Sales has them, but I refuse to pay a buck a piece for them, especially when
they have to be handpicked for each osc. coil!!!

Pete  k1zjh