EEVblog #767 - Super Regenerative Receiver Problems

[ElectroIrradiator]

Your description of a Regenerative Receiver sounds more like a Direct Conversion Rx.

That is because a regenerative detector *is* a Direct Conversion Receiver when it is oscillating. It has been roughly 35 years since I built my first regen receiver, so I know how they work.

Somewhere I should also have my little experimental superregen for the 88-108 MHz FM band if anyone would like to see it. (The superregen detector uses slope AM/FM detection like the receiver in the video shown previously). Caution! Contains vacuum tubes. May upset sensitive semiconducting souls.

[vk6zgo]

Your description of a Regenerative Receiver sounds more like a Direct Conversion Rx.

That is because a regenerative detector *is* a Direct Conversion Receiver when it is oscillating. It has been roughly 35 years since I built my first regen receiver, so I know how they work.

But only when it is oscillating---it is not used in the oscillating mode when receiving AM.
A true Direct Conversion receiver has a Local Oscillator which is always oscillating.

I built my first Regen set 56 years ago!

Somewhere I should also have my little experimental superregen for the 88-108 MHz FM band if anyone would like to see it. (The superregen detector uses slope AM/FM detection like the receiver in the video shown previously). Caution! Contains vacuum tubes. May upset sensitive semiconducting souls.

[ElectroIrradiator]

But only when it is oscillating---it is not used in the oscillating mode when receiving AM.
A true Direct Conversion receiver has a Local Oscillator which is always oscillating.

Ah, I see. You are splitting words in order to sound clever. OK, carry on.

[apis]

Ah, so the capacitor isn't involved in the actual mains switching. It really is just the relay, with no MOV or capacitor or anything to protect the relay from arcing while switching off inductive loads?

Shouldn't there be some sort of snubber? Am I correct in thinking this thing really isn't suitable to power inductive loads? Maybe it is only meant to power resistive loads, but I wonder if that is stated clearly on the box it came in?

I don't think there is any, there could be a MOV but couldn't see any. Ideally there would be a snubber inside the inductive load itself.

They need to power the receiver somehow and in cheap low power devices that's typically done with a X2 cap so I'm guessing the cap is for dropping voltage in this case.

[vk3yedotcom]

Ah, I see. You are splitting words in order to sound clever. OK, carry on.

I agree with vk6zgo.

There are significant behaviour differences between a regenerative and direct conversion receiver, even when the former is oscillating.

A regenerative receiver, when brought towards oscillation, narrows its selectivity as the Q of its tuned circuit is effectively heightened with increasing amounts of positive feedback. 

The maximum Q is just before oscillation, but it's still fairly high when it's gently oscillating. 

This effect is helpful when trying to tune a crowded HF band and pick signals apart.

In contrast you don't have that effect with a direct conversion receiver - if you want high selectivity you need to apply audio filtering.

The second difference is that a regenerative detector has a non-linear oscillating detector. 

In contrast a 4 diode balanced mixer, which is common on direct conversion receivers, is basically linear.  This is beneficial when designing a receiver with good strong signal handling characteristics, making a good DC receiver superior, especially if there is reasonable band pass filtering between antenna and detector.

You can blur the lines between the two by having both a regenerative detector (set just before the point of oscillation) and an external beat frequency oscillator on the reception frequency that is near but not physically connected to the detector.  Personally I find this gives the best results when receiving weak signals.  It also overcomes many of the problems of regenerative receivers, eg the local oscillator changing frequency when a different antenna is connected, tuning in the antenna circuit is adjusted or frequency pulling on strong signals.   While off-topic, an external BFO is also beneficial with a crystal set style detector as it provides RF bias, better weak signal reception and even CW/SSB capability.

Superregen detectors are different again - they are way less selective and are only good for AM and wideband FM signals (they'll cope with narrowband FM but you need to crank the volume up and the hiss can be hostile). 

[ElectroIrradiator]

I agree with vk6zgo.

The I am afraid we will have to agree to disagree.

I didn't bring up the DC RX. What I wrote was:

A classic regenerative RX generates a continuous sinewave oscillation at or very near the intended receiving frequency. The RF input signal from the antenna is mixed with the LO signal and an audio or supersonic low frequency signal is the result. This is where the frequently seen audio transformer comes into play in your MW broadcast regenerative RX. So with a regenerative detector you'd expect to see a single frequency being radiated by the detector.

Why are we having this debate, based on what I wrote?

[SeanB]

Those relays are very prone to sticking, or to arcing themselves to pieces. They are the standard relay used in those sealers you see al laround for sealing plastic bags, where they drive the 56VAC transformer used to heat the sealing element strip.

Failure mode is either the contacts arc away ( lovely fireworks show and a burnt relay case with lots of smoke) or they simply weld closed. I buy a local version that is slightly better rated, with a 5A rating, and with slightly thicker current carrying pins, which means I have to drill the 2 holes in the SRBP board oversize to fit them.

Standard addition is a VDR across the contacts, to reduce the arcing on the coils. They lead a hard life in those sealers, and I keep a few spares along with all the electronics for the control, which is a 5W resistor, some C106D SCR's, some 6V8 zener diodes and the single NPN transistor used there, along with the 10uF 400V capacitor and the 100uF 25V timing cap, and always add a 220k resistor across the control pot to reduce on time, as you never need more than a 2s pulse to seal, and they go all the way to 8s which cooks the element and the PVC bags.

[VK3DRB]

433.920 MHz is the "CB band" or "Kiddies band" in the remote control world. I had a non-technical friend whose garage door opener became very insensitive to the remote, making it next to useless. He told me he paid a ripoff merchant about $1,000 to get the main board replaced, but the problem was still there. I investigated it and found one of his neighbours had installed some crappy Chinese weather station atop his house that transmitted on 433.920 MHz, wiping out the frequency for a radius for several houses around. 433.920 MHz is a risky frequency to use in Australia because you have no idea what RF crap is around to interfere on that frequency. Most modern car remote door openers have moved off that frequency.

Was this Arlec piece of crap really designed in Australia? http://www.arlec.com.au/innovation-design/ I doubt it!



 

[vk6zgo]

I agree with vk6zgo.

The I am afraid we will have to agree to disagree.

I didn't bring up the DC RX. What I wrote was:

A classic regenerative RX generates a continuous sinewave oscillation at or very near the intended receiving frequency. The RF input signal from the antenna is mixed with the LO signal and an audio or supersonic low frequency signal is the result. This is where the frequently seen audio transformer comes into play in your MW broadcast regenerative RX. So with a regenerative detector you'd expect to see a single frequency being radiated by the detector.

Why are we having this debate, based on what I wrote?

We are having this debate precisely because of what you wrote.

A Regenerative Receiver only generates a continuous sinewave oscillation at or near the received frequency in the special cases where it is receiving CW or SSB signals.

When receiving AM signals,the level of feedback is kept below that which will cause oscillation.

Positive feedback at just below the threshold of oscillation has the following beneficial effect--cancellation of losses in the Receiver input tuned circuit,leading to higher circuit Q,which in turn increases both selectivity & sensitivity of the receiver.
It is this effect which is the primary benefit of Regeneration,rather than oscillation.

You can receive AM in the oscillating mode,but the "L.O." thus formed is unstable,so the demodulated audio drifts off into distortion due to audible beat notes mixed with the desired audio
I know this happens--I've tried it!.

It is hard enough to receive AM in this manner using a Superhet with a stable BFO!

[G0HZU]

For a bit of fun I tried capturing an on and an off burst from the STATUS (UK model) transmitter using an Agilent E4406A and the 89601 VSA SW from Agilent. This allows the burst to be captured and saved as I and Q data to various file formats. It looks like simple OOK.

For a bit more fun I then transferred the captured IQ burst directly into an Agilent vector sig gen as I and Q bin data (contained in I.bin and Q.bin files) and played the file data back at several mW via an antenna.

This basic 'grabber' and replay system worked perfectly (no surprise) as I guessed there would be no security in the form of rolling codes etc.
So the vector sig gen could turn the unit on and off the same as the real remote.

[99tito99]

The scariest thing is you have stuffed your “crawl space” with OneHungLow random/spontaneous deflagration units.

Are crawl spaces flammable? I mean, obviously spontaneous deflagration units are to be avoided, but perhaps a crawlspace is an ideal location for them?

Although your thought that this is the best spot if not to hot maybe right, this void devoid of sight causes some fright. (wow, that's a stinker!)

Cheers,
Mark
*****************