While browsing the web, I found this simple superhet project:
https://hackaday.com/2021/04/19/a-superheterodyne-receiver-with-a-74xx-twist/
https://acidbourbon.wordpress.com/2021/04/11/a-74xx-defined-radio/
It uses an IF of 100 kHz and an opamp for the IF filtering. Because of the low IF, and no RF filtering, there will be the problem of receiving images 200 kHz away from the desired signal, which might be solved by double conversion.
The simple VFO is kind of interesting to me. It just uses a 74HC4046 PLL chip (available for less than a dollar from one of my local suppliers) with a couple of discrete components. I do wonder about how much noise is introduced by such a simple VCO though.
@Sean O'Connor wrote:
This is an interesting idea. I'm kind of drifting off-topic now, but this could also be an interesting converter circuit for a regenerative superhet, because the cross-coupled oscillator tends to generate a stable voltage output (due to the limiting behavior), and this helps with providing a hopefully constant load to the following regenerative detector stage.
I keep asking myself what is the simplest regenerative superhet possible that (1) doesn't require adjusting the regeneration control as the LO is tuned, i.e., constant loading on the regenerative stage and (2) is sensitive enough to hear band noise with a small ferrite antenna, or even better, sensitive enough to hear the mixer noise.
I don't think this goal can be accomplished (easily) with just 2 transistors (one RF->IF self-oscillating mixer, then a regenerative IF stage), though it's barely possible. The problem with the self-oscillating mixer is the varying LO output level as the LO is tuned, causing varying loading on the regenerative stage, requiring adjustment of the regeneration level as the set is tuned. However, vladn's previous work on hybrid feedback oscillators might allow a one-transistor self-oscillating-mixer with constant output level to be designed. But it would require high-Q components with low ESR and no unpredictable losses, which is hard to achieve in practice.
So probably in practice the cross-coupled oscillator (2 transistors), wired as a self-oscillating mixer as you describe, plus one regenerative detector transistor might work as the simplest-possible. You'd still need an audio amp or audio transformer though.
A long time ago on the old board I posted a superhet circuit based on the cross coupled oscillator that you know better than anyone at this stage. I don't have a circuit diagram at the moment. Somewhere on the internet is an improved copy someone constructed. I'll look more on the internet for it today.
Basically it was the cross coupled oscillator, with the one free base connected to a ferrite rod antenna for preselection. And the free collector connected to an IF transformer.
I would design it slightly different today with a Seiler oscillator as the (oscillating) current source to the differential pair in place of the single resistor and then just a ferrite rod antenna connected to the base of one transistor and the IF transformer to the collector of the other transistor.
That results in a very stable, low phase noise oscillator as one input to an active multiplier (mixer) circuit and the signal from the ferrite rod antenna as the other multipler input.
https://www.seas.ucla.edu/brweb/papers/Journals/BR_Magzine1.pdf