Regenerative receiver circuit

[dominicbeesley]

Hi chaps,

I had a spare hour last night so I decided to fiddle around with a germanium transistor (OC44) and the attached circuit which I found here: http://zpostbox.ru/super_regenerator.html

I couldn't get the circuit as drawn to work it just went into saturation at whatever voltage/setting of the regen control. However, with a capacitor between the base of the transistor and the coil and a bias resistor of 470k it works after a fashion.

The trouble is then that the idea of adding a capacitor from emitter to ground to make it a super-regen instead of a plain regen doesn't seem to work. Is there something I'm missing, do I need a different transistor?

I'd be grateful for any ideas on _why_ this circuit doesn't work rather than suggestions of other circuits (I've tried lots of different ones in the past).

Cheers

D

[ms660]

Not sure if I can help but so far as I understand it super regens don't work that well below 30Mhz? Not sure what the upper limit of the OC44 is but someone on the forum will know.

Lawrence.

[Radio Wrangler]

The super regen is an application of a squegging oscillator

It needs a slightly starved power supply and a bit of decoupling capacitance, and the regen receiver is tuned so that it will oscillate.

On power on, the power starving resistor (the pot) charges the decoupler until there is enough volts for the oscillator to burst into oscillation. The current taken jumps up and the voltage on the decoupler drops. At a lower voltage, the oscillator drops out, the current taken drops and the decoupler resumes charging until the cycle repeats.

The presence of a signal helps the oscillator start a little sooner, and run to lower decoupler voltage. So the percentage of time the circuit takes the larger amount of current follows the modulation of the incoming signal.

So you have two oscillatory modes to set up... the receiver transistor oscillating close to the wanted station, and the relaxation oscillation of that oscillator going on and off. The capacitance of the decoupler has to set the squegging at an ultrasonic frequency or it will sound hellish. The old word for sounds pitched too high to be audible was 'supersonic' and we would now say ultrasonic... so this is where the 'super' in superhet and in super-regenerative come from.

Try C between the 1k resistor and the pot. set the pot at low R and adjust the regenerative receiver to be unstable and oscillating at the wanted frequency, then increase the resistance until the oscillation cycles. It's afairly narrow zone, too much and it doesn't start too little R and it never stops.

Superregen's aren't very sociable, but few people listen to MW nowadays.

David

Superregen's aren't very sociable, but few people listen to MW nowadays.

The few would like a clean signal though, stick to ordinary regen, better on MW too. Super regen bandwidth is large.

[dominicbeesley]

Thanks all,

Sorry, I'm not sure I made my question clear. I'm trying to get it to work as a normal regen before I try super-regen. But it won't with the base connected direct to the coil - is that normal, a mistake on the linked website, or something else? Say I wanted to make a simple oscillator in that configuration how would you guys connected the base to the coil?

D

[kalee20]

It does look 'wrong' with the base and collector linked by the coil, especially with a germanium transistor where the base-emitter voltage is only a couple of hundred millivolts.

Can't comment further, apart from saying that while phasing is correct for oscillation, the DC conditions aren't!

[Radio Wrangler]

DC-wise it's going to bias itself with the collector at 1 Vbe above the emitter, which isn't going to do much for gain.

David

[dominicbeesley]

Thanks lads,

I've found a few circuits like this - mostly seemingly of Soviet descent. I wonder if there's something about their germanium transistors?

Cheers

Dom

[ms660]

Yes, some of the designs from the "other side" look a bit simple and weird, the Polyakov mixer is one of them but that definitely works.

Lawrence.

[dominicbeesley]

Thanks Lawrence,

I had a bit of a play in LTSpice and got the circuit with base and collector connected straight to the coil to work with a BC337 transistor (and a noise source to kick it off) though I couldn't get super-regeneration to work as it seemed to draw less current when oscillating. I will persevere with it a little longer when I get time as I find the simplicity intriguing!

I've ordered some high frequency fT > 300MHz soviet germanium transistors and will have a go at building an FM super-regen if I ever get the chance to visit the workshop again in the next few months!

I'm getting quite into Soviet stuff, when I visited Chernobyl last year it was interesting to see some of the seemingly advanced (for 1980's) looking computer equipment that looked like it was being used against very old-fashioned valve equipment and some of the weird stuff (things that looked like old core memories and other odd computer buts) in some of the derelict factories...

D