I have more work done and have a well working set now.
The design is settled, an AGC controlled oscillator/ mixer and a neutralised reflex IF/AF amplifier/output.
that is a lot for a twin pentode!
I am not sure if I have an "Autodyne" or a "Neutrodyne" but it sure works well.
I have never used the neutralising technique before and it was like a magic trick as I could introduce the neutralisation on the working hot chassis and see the effect on the go.

Normally you go to extreme lengths to keep input and output wiring on a high gain stage apart.
In this case without neutralising, the reflex stage starts off barking mad unstable with about 170Kc self oscillation RF drive to the speaker.
With that happening the stage is just choked up with no reflex Audio output.
Simply attaching a short length of wire to the plate pin and bringing that adjacent to the grid wiring made the audio suddenly appear out of the fog of RF.
With 5 turns loosely coupling the plate wire to the grid wiring all trace of RF disappears and the valve delivers audio out of the plate and the stage is stable.
I could re-set the IF transformers for maximum passband and gain, still stable.
Simply amazing.
I am sure there is a bunch of old radio engineers rolling their eyes saying, "so what! we all knew that 100 years ago!"
To a person who has only made a few scratch built superhets it is amazing.

I will send a part 2 to Brad to attach to post #1 which will bring the story up to the point just before adding neutralisation.

A part 3 will follow up once I transcribe my workshop notes and photos to file and that will have the gain and AGC tests on the first pentode section and the magic neutralising on the reflex section.

Fred.

You might find that you could replace the 5 turns with a 1.5pf cap or something. I've seen that idea used in a radio.
However, if it's working it might be safer to leave it - up to you of course.

Hi Rob, yep that would be the correct thing in production.
Twisting the cables is a good variable cheap and high voltage rating alternative!
Ian recalled the go with HMV was to wire in a piece of "figure 8" flex and trim it to suit each production set.

The Radio (Radiotron) Designers Handbook has a host of references to re-generation and neutralising.
I read a couple of pages and fell asleep.
But at least it gave me a glimpse of the theory behind the technique.
The practical advice in neutralising a reflex IF stage was to do what you suggest put in a fixed value of a low size.
Any amount of neutralising is a benefit, so a "fits all" size works nicely and works with production sets.
I bet the HMV wiring guys just cut off a 1.5231" length and stuck that in each set and so long as the IF gain and stability met the target no change needed.

Fred.

Twisting the cables is a good variable cheap and high voltage rating alternative!

Yes, getting a high voltage 1.5pf cap might be a problem these days, you'd probably have to put a 0.01 high voltage in series with a low voltage 1.5pf, but the size of it all would undoubtedly cause instability problems of its own. So having a short twisted wire would be the best solution.

Well done Fred in getting this unlikely scenario to actually work.

Part 2 uploaded.

Great article Fred. Your knowledge of the theoretical aspects is astounding.

I noticed a few typos though, the final schematic is dated 19/10/2022, which is sometime in the future, while "End of part 2. 20/10/22" is also yet to happen.

Your reference to a transmitter makes me wonder if it's possible to construct an AM transmitter with just this valve. An output of 1 watt should be achievable.

Hi Rob, yeah Damm Tardis, must get the hyper timey wimy modulator re calibrated keeps jumping forward 1 month!
(Nothing to do with being old and stupid!)

Those 10 pin valves are very interesting, the performance means they cut the valve count in a TV dramatically, 6X9, 6Y9 and 6V9.
I was not even aware of them until Ian R pointed them out.

I do not know much about transmitters but the 6Y9 could be quite handy, it runs at HF being intended for video use and the big half has very high gain and a chunky plate structure. The only transmitter I have made are the AM repeaters and this valve having 2 pentodes would make an easy tank oscillator and modulator.

I'll stick to making funny superhets!

Fred.

Those 10 pin valves are very interesting, the performance means they cut the valve count in a TV dramatically, 6X9, 6Y9 and 6V9.
I was not even aware of them until Ian R pointed them out.

I knew they existed because I have a bunch of them in my collection. They were the valves used before the last-gasp Compactron (12-pin) series, and as transistors took over.

There's another one in that group, the 6U9 - just as common as the other 3 you mentioned.

Something some of you radio guys may be able to tell me.
Any idea what radio models the IF trannies I used came off?
I took some close ups of the cans in part 2 with numbers showing.
I have a box of coils and cans bought as junk so have no idea of the breed????

Reason I want to know is to present as much info for anybody that may like to have a go at this design.
Ian has proposed a power supply using stock Jaycar transformers in a doubler circuit, so a "Fred special" tranny is not needed. The OPT is stock Jaycar and a valve-type sensitive 4-ohm speaker would probably be better than the brick like Jaycar speaker I used.

It may be possible to use a transistor coil set also a Jaycar item.
But.
I simply used valve IF's and tuning coil sizes that were era compatible with valves and using litz wire that can handle milliamps.
There would also be still floating around Aegis or Kingsley coil sets as I have used in other superhets.
A tuning coil and IF set from a 1970's 6BE6/6BA6 HMV or STC like thing would be perfect being small in size and if the donor had a matching mini-gang as well that would be perfect.

It would be nice if somebody could take up the challenge and make a 6Y9 set.

Fred.

The 1st IF looks like Astor (the cutouts in the side of the can that engaged the mounting clips are the giveaway) and the 2nd is definitely HMV.

Some scrap battery valve chassis would be a good source of parts. I have a garage full of them (but they all work!).

Kriesler 11/99s would also be a good chassis to base this on, but did you ever find an 11/99 that didn't still work?

I raised it on the UK forum in response to a "minimalist superhet challenge" post and a few guys reckoned they'd build one.
Using a PFL200 (17 volt heater) of course.

Ian, that Kriesler 11-99 design could be a good unit to get all the critical parts needed.
It appears that thousands of them were made so hopefully there are a good number still left.
Had a look at one on Ebay, I bid on it and that will probably give me a non working example to play with.

The circuit is brilliant, what a minimalistic design.
With 110 volt HT every thing is loafing along, no wonder they last forever!

The coil set would be suitable and even the power transformer might be ok.
That shows 115 volt AC on the D11-99 circuit, that would give 140 volt DC with a full wave bridge.

A "kit set" could be based around a dead 11-99 or even previous mantle models.
What could go wrong?

Fred.

I have finished work on the prototype and will send a "part 3" to Brad to add to the others at the top of the thread.
Having done that prototype I have a bit of an idea of the compromises I have built into the layout and the valve.
I have a Kriesler 11-99 on the way. It will be typical of a dead set you may buy as junk, but hopefully the coil set and tranny, speaker, will be ok.
That can form a "kitset" that anyone can copy.
Work should start next week and I will update on this project as we go.

Cheers, Fred.

Part 3 uploaded.

The performance you are getting is, as I recall, similar to that of 1930s 5 valve autodyne radios. The 1st stage 57 never added much in terms of gain.

They needed a long wire antenna and had no AGC. Volume was controlled by throttling back the IF stage and shorting the antenna.

I'm impressed that you were able to make a practical, usable radio out of it.

Fred, Ian and all other contributors

I just want to say how much I have enjoyed following this thread and I am in total awe of your combined knowledge.

Many thanks

Harold