EMRFD Message Archive 6444
Message Date From Subject 6444 2011-07-17 08:35:02 qrp.gaijin Single-signal regenerative receiver
I'm very interested in and would like to ask a question about regenerative receivers. The issue at hand: I've recently discovered a simple way to turn any regenerative receiver into a single-signal receiver that can suppress one sideband; typically regenerative receivers are thought of as being unable to suppress the opposite sideband. My question is if any of you have seen or heard of a similar scheme being used before, perhaps in the olden days of regenerative or superhet receivers.
Assume we are interested in single-signal CW reception with a 1 kHz beat note. The proposed idea uses three combined techniques: (1) a superheterodyne-style single-conversion architecture with (2) an audible heterodyne beat note (instead of a supersonic heterodyne) of 10 kHz used as an intermediate frequency with sharp 10 kHz audio filtering to remove one sideband before final down-conversion to 1 kHz, and (3) a regenerative front-end RF mixer run just below oscillation threshold at the reception frequency in order to remove the close-in RF image (caused by the very-very low IF choice) by using Q-multiplication instead of resorting to double conversion.
Has this scheme been tried before? It's almost like a superhet, except the IF is an audio frequency heterodyne beat note, so strictly speaking it isn't a superheterodyne receiver. Since an audio frequency heterodyne and a regenerative mixer are required, I'll call this scheme an "audiogenerodyne".
I implemented this scheme and found during testing that it seems to work well, allowing a marked attenuation (dependent only on the degree of audio filtering) of signals on the undesired side of zero-beat. The interesting practical implication of this architecture is that any existing regenerative receiver can be used (repurposed) for single-single CW reception without making any circuit changes to the existing regenerative receiver. The reasons this is possible are (1) the "regenerative mixer" that this scheme requires is expected to output an audio frequency IF, and by definition any regenerative receiver already outputs audio frequencies, so any unmodified regenerative receiver can serve as the mixer; and, (2) the local oscillator requires no direct connection to the regenerative receiver, and stray coupling can be used for LO injection, just as old-time regen-plus-beat-frequency-oscillator designs sometimes did.
My implementation used an existing MOSFET regenerative receiver with a nearby BFO tuned to the regenerative receiver's frequency plus 10 kHz for the mixing. Audio IF filtering and final audio down-mixing were done with audio DSP software on a PC using the PC sound card for audio input. Architecturally it looks like a superhet, but as mentioned above the IF is at audio. Operationally, instead of a superhet, it feels more like a operating a regen with a BFO, plus some black-box audio filtering. It feels like this particular combination of architectural features is in a grey zone between regens and superhets.
In practice this isn't completely trivial to tune, which is probably why there don't seem to be any designs (that I could find) pursuing this approach. But for someone who likes operating regenerative receivers, it's kind of fun to try this scheme, especially because no alteration to the original regenerative receiver circuitry is needed to get single-single reception. I think this particular technique is interesting to and maybe useful for only those people who are trying to get more out of their regenerative receivers - not interesting for absolute high-performance reception, but interesting perhaps as an extension to the fascinating classic technology of regeneration.
What do you think? Old trick, or new trick? I haven't seen the historical survey literature on regenerative receivers mention trying something like this, but maybe some early superhet developements were done in a similar direction.6445 2011-07-17 09:02:35 Alex P Re: Single-signal regenerative receiver
There's alot to digest in your post. Maybe you could draw a simple block diagram and post it so that the architecture can be more clear?
Alex
6446 2011-07-17 10:10:19 dave Re: Single-signal regenerative receiver
image frequency, which would be 10 KHz above the BFO, or 20 KHz above the
desired signal.
Dave - WB6DHW
<http://wb6dhw.com>
6447 2011-07-17 10:22:59 Ashhar Farhan Re: Single-signal regenerative receiver
achieve a single signal response. you could even pull the oscillator to
tune the band, and then of course, if you increased the emitter current, it
would radiate back a cw signal. how's that for an omnidirectional cw rig (as
opposed to a bidirectional tcvr) ?
- farhan
p.s. i have had too much caffiene today, i am maybe ranting.
6448 2011-07-17 10:39:50 Roelof Bakker Re: Single-signal regenerative receiver
A German pre-war receiver used a similar technique.
I have even read a Dutch article what described this phenomena called:
"You don't know what you are missing".
As this is generally true, in this case it refers to the fact that the
peak of the selectivity coincides with zero beat in a regenerative
receiver. I will try to find it back.
73,
Roelof Bakker, pa0rdt6449 2011-07-17 15:15:49 qrp.gaijin Re: Single-signal regenerative receiver 6450 2011-07-17 15:55:04 qrp.gaijin Re: Single-signal regenerative receiver 6451 2011-07-17 17:54:29 kb1gmx Re: Single-signal regenerative receiver 6452 2011-07-17 18:02:00 Chris Trask Re: Single-signal regenerative receiver
> I'm very interested in and would like to ask a question about
> regenerative receivers. The issue at hand: I've recently
> discovered a simple way to turn any regenerative receiver
> into a single-signal receiver that can suppress one sideband;
>
>
><
> Has this scheme been tried before? It's almost like a superhet,
> except the IF is an audio frequency heterodyne beat note, so
> strictly speaking it isn't a superheterodyne receiver. Since
> an audio frequency heterodyne and a regenerative mixer are
> required, I'll call this scheme an "audiogenerodyne".
>
>
>This is a very intersting idea. I'd like to see the nuts-and-bolts
> What do you think? Old trick, or new trick? I haven't seen the
> historical survey literature on regenerative receivers mention
> trying something like this, but maybe some early superhet
> developements were done in a similar direction.
>
side (schematic) of this scheme.
I've used the high audio frequency IF for an earlier version of my
adjacent-signal interference canceller as it was more manageable for
experimentation.
Chris Trask
N7ZWY / WDX3HLB
Senior Member IEEE6453 2011-07-17 19:14:44 qrp.gaijin Re: Single-signal regenerative receiver
> > except the IF is an audio frequency heterodyne beat note
> This is a very intersting idea. I'd like to see the nuts-and-bolts
> side (schematic) of this scheme.
I just used the equipment I had available to test the idea. Specifically:
1) This regenerative receiver as a RF-to-audio-frequency-IF mixer: http://www.electronics-tutorials.com/receivers/regen-radio-receiver.htm
2) This grid-dip oscillator as a BFO: http://www.b-kainka.de/bastel53.htm (second circuit from the bottom). The BFO was held about 10 cm from the regen's LC tank with no physical connection.
3) The JACK-Rack audio filter software for the IF filter and final audio downmix. I enabled the following filters: a 48 dB amplifier, a 10 kHz filter with 38 Hz bandwidth, and an SDR digital-down-conversion mixer plugin described here: http://lists.linuxaudio.org/pipermail/linux-audio-dev/2010-February/026828.html . The software user interface looks like this: http://img231.imageshack.us/img231/5465/sdrg.png .
Item (3) should be implementable with simple all-analog hardware. I'm considering a phase-shift RC oscillator held just below oscillation at 10 kHz for the sharp IF filter, an 11 kHz RC oscillator for the BFO, and some simple active mixer for the final downmix.
That's it. From a physical-component perspective, we're just adding a BFO and some audio processing to an existing regenerative receiver, and we can achieve single-signal reception, if we are willing to put up with the multiple-knob tuning required.6454 2011-07-17 22:00:56 victor Re: Single-signal regenerative receiver
One hand the first BFO (Which is really your LO) which you tune with it what frequency to receive, second is the regenerative pre-amplifier frequency tuning to reject the image frequency and third the pre-amplifier regeneration control. Good for setting up on one receive frequency, but lets see you scan a band from edge to edge, you need a third hand and you are never shure if the regenerative tuned preamplifier is6455 2011-07-17 22:23:22 qrp.gaijin Re: Single-signal regenerative receiver 6456 2011-07-17 23:23:37 qrp.gaijin Re: Single-signal regenerative receiver 6457 2011-07-18 07:31:23 Tim Re: Single-signal regenerative receiver
There is a very strong bias in almost all the high-performance receiver articles, that "single knob" tuning is the holy grail. I sometimes have a hard time determining if this is a QST editor bias, or if it is a very real wish among all the authors.
There is an attitude that multi-knob tuning (not just for receivers but also transmitters) is very much a non-professional approach.
In the timeframe of these articles, the single knob holy grail was often implemented mechanically and there are some interesting mechanical arrangements used to achieve it... some of the most elegant use chain drive when putting everything on a single shaft wasn't possible. The simpler cases simply had multiple variable capacitors on the same shaft with tracking between sections (e.g. trimmer caps) like you'd find in any non-ham radio.
Looking back at the previous century it seems to me that the single knob holy grail was nothing more than a red herring. But it was awful important to the editors back then. Sometime in the 60's I think the editors or maybe the ham community at large was beginning to appreciate that there were multiple solutions to the price-complexity-performance-ease of use-band coverage tradeoff matrix, and that all of them had their place, that none were automatically superior to the others.
Of course there are other more recent holy grails that are probably red herrings too. e.g. confusion about whether to build a simple one band high performance radio, or whether the radio "has to be" DC to daylight and also high performance.
6458 2011-07-18 07:46:22 Chris Trask Re: Single-signal regenerative receiver
> It is actually quite interesting to read, say, QST homebrew
> receiver articles from the 30's through the 50's and maybe
> even the 60's.
>
> There is a very strong bias in almost all the high-performance
> receiver articles, that "single knob" tuning is the holy grail.
> I sometimes have a hard time determining if this is a QST
> editor bias, or if it is a very real wish among all the authors.
>
> There is an attitude that multi-knob tuning (not just for
> receivers but also transmitters) is very much a non-professional
> approach.
>
>
I have yet to see a single-knob receiver that gives performance comparable to that of even a mediocre multi-knob receiver. A workable compromise was to gang the RF and LO tuning and then have a trimmer adjustment for the RF, which made operation much easier and alignment less of a nightmare.
A compromise that appeares in many solid-state receivers is to have a wideband front end. Although convenient, the front end is easily overloaded even with high linearity amplifiers and mixers. And between these extremes are the receivers with bandpass filters added to the front ends, which greatly increases the cost. In all, a high-selectivity front end will invariably give better performance at a resonable cost.
Chris Trask
N7ZWY / WDX3HLB
Senior Member IEEE
http://www.home.earthlink.net/~christrask/6459 2011-07-18 10:23:41 Bill Carver Re: Single-signal regenerative receiver
PA3AKE has shown the third order intercept of a passive LC preselector degrades rather badly when it's made narrow, to the point it is lower than the intercent of (his) mixer. Narrowband is NOT always better.
EVERYTHING is a tradeoff. The wonderful thing about crafting a homebrew radio is we can pick and choose tecnhologies, power, mechanical complexity, and performance requirements to suit US, rather than accept those of a general-purpose manufacturer are dominated by bean counters and have no junkbox!
W7AAZ6460 2011-07-18 14:35:22 kb1gmx Re: Single-signal regenerative receiver 6461 2011-07-19 04:58:28 qrp.gaijin Re: Single-signal regenerative receiver 6462 2011-07-19 07:15:16 Tim Re: Single-signal regenerative receiver
I personally do not like the sound of cookbook-style single-stage active audio bandpass filters. They tend to be way too "ringy" in the face of impulse noise.
If you design an audio filter so that there are multiple poles, with careful attention made to move the region of rapid phase shift out of the passband, the audio quality is way more enjoyable and much less ringy. See EMRFD for "Gaussian to 12 dB" for example. This is true whether it's a crystal filter or mechanical filter or an audio filter. Do all this in DSP and you can do it without a lot of L's and C's.
BTW, December 1922 QST has a interesting tidbit on page 11, I think in Armstrong's original idea the IF was much closer to audio frequencies:
In December, 1919, Major E. H. Armstrong gave
publicity to an indirect method of obtaining
short-wave amplification, called the Super-
Heterodyne. The idea is to reduce the incoming
frequency which may be, say 1,500,000 cycles
(200 meters), to some suitable super-audible
frequency which can be amplified efficiently, then
passing this current through a radio frequency
amplifier and finally rectifying and carrying6463 2011-07-19 08:15:32 Tim Re: Single-signal regenerative receiver
"Good mechanical construction is too expensive for mass produced
or even kit radios, but is just a matter of planning, care,
some worthwhile mechanical skills, and time for a designer-
builder of a single radio. This is one area where a
designer-builder can far exceed the mechanical quality and
electrical integrity of a mass-produced receiver built under
severe time and budge constraints, for example, a Collins 75S3C."
In a slightly different direction chosen on the price-performance-complexity trade-off matrix, the sound of a miniR2 or binaural IQ receiver at wide bandwidth and listening to a variety of CW signals simultaneously6464 2011-07-19 15:59:05 Bill Carver Re: Single-signal regenerative receiver
I "grew up" with a Hallicrafters S-85 that had zero rejection of the opposite sideband. A Heath QF-1 Q multiplier improved it a great deal, as a regenerative detector would. But I'll never forget that audio image biting me once in a while, and I don't build anything without 100 dB of opposite sideband rejection at 1 KHz. "Holy grail"??? No, that's too grand a term for a "personal idiosyncrocy" (sp?)
Now upward/onward......
Are you aware of the "dual tank oscillator" that Colin Horrabin, G3SBI, described in Radio Communications? He found that coupling two resonant tanks together could almost double the phase noise rolloff rate in an LC oscillator. That was about ten years ago, and that's been commercialized in a super high performance SWL receiver designed in the UK.
So I'm wondering....could you make a two tank regen detector? Would that produce approximately twice the rolloff rate just like Colin's LO, and give you 60 dB of rejection of the audio image? I'm on the road in Colorado. Maybe Dan Rae of someone that has a good set of RSGB literature could give you a better jumpstart on this idea.
Bill - W7AAZ6467 2011-07-20 05:27:06 Chris Trask Re: Single-signal regenerative receiver
> I'm very interested in and would like to ask a question about
> regenerative receivers. The issue at hand: I've recently
> discovered a simple way to turn any regenerative receiver
> into a single-signal receiver that can suppress one sideband;
>
>
><
> Has this scheme been tried before? It's almost like a superhet,
> except the IF is an audio frequency heterodyne beat note, so
> strictly speaking it isn't a superheterodyne receiver. Since
> an audio frequency heterodyne and a regenerative mixer are
> required, I'll call this scheme an "audiogenerodyne".
>
>
>This is a very intersting idea. I'd like to see the nuts-and-bolts side (schematic) of this scheme.
> What do you think? Old trick, or new trick? I haven't seen the
> historical survey literature on regenerative receivers mention
> trying something like this, but maybe some early superhet
> developements were done in a similar direction.
>
I've used the high audio frequency IF for an earlier version of my adjacent-signal interference canceller as it was more manageable for experimentation.
Chris Trask
N7ZWY / WDX3HLB
Senior Member IEEE
http://www.home.earthlink.net/~christrask/6471 2011-07-20 10:56:49 qrp.gaijin Re: Single-signal regenerative receiver 6476 2011-07-20 19:22:08 cbayona Re: Single-signal regenerative receiver
the regen detector and for the same reasons, better frequency
rolloff. Has anyone tried it? Or for that matter, a band pass filter
in addition to the tuned circuit inside the regeneration loop.
The two circuits would need to track fairly close in order to have a
workable circuit and that is not so easy to do. The amount of
coupling could be changed to affect the basic shape of the resonant curve.
Cecil
k5nwa6478 2011-07-20 21:19:04 qrp.gaijin Re: Single-signal regenerative receiver 6482 2011-07-21 11:10:35 Bill Carver Re: Single-signal regenerative receiver
That would not apply below oscillation, would it? You would not want a regenerative circuit to pop in/out of oscillation as you tuned the radio. Although having one hand on the regneration control was required in some implementations in the good 'ol days. For a regenerative detector it might require tuned circuit tracking, and therefore be a non-starter. All cute ideas that pop into one's head do not necessarily hold water, hi.
W7AAZ6483 2011-07-21 13:35:29 cbayona Re: Single-signal regenerative receiver
>6487 2011-07-21 19:28:58 cbayona Re: Single-signal regenerative receiver
>6492 2011-07-22 04:41:26 victor Re: Single-signal regenerative receiver
Victor - 4Z4ME
6508 2011-07-22 18:04:14 Dan Mills Re: Single-signal regenerative receiver
> That would not apply below oscillation, would it? You would not want aI am sure I remember superregen sets that squegged into and out of
> regenerative circuit to pop in/out of oscillation as you tuned the
> radio.
oscillation at an ultrasonic rate.
These were effectively early direct sampling receivers!
In fact I seem to remember getting into trouble in my youth for building
a simple airband radio using that trick to let me listen to the tower at
an airshow..... I knew the show would be loud so I sort of increased the
plate current to make the 'phones a bit louder, apparently my 50mW or so
of audio came at the cost of a WATT of fairly broadband RF...... The
tower kept retuning and I would follow them....
The things you can almost get away with when you are 12 years old!
Regards, Dan.