UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Other Discussions > Homebrew Equipment

Notices

Homebrew Equipment A place to show, design and discuss the weird and wonderful electronic creations from the hands of individual members.

Closed Thread
 
Thread Tools
Old 3rd Jul 2022, 9:39 pm   #1
E93AFAN
Tetrode
 
Join Date: Mar 2022
Location: Chippenham, Wiltshire, UK.
Posts: 69
Default Distortion in Regenerative Receivers

I hope this is the correct place to post this question.......

I have built a number of valve based regens using the likes of DF91, DL91/93, 6J7, 6K7, EF36/9, 6SN7, 6SL7, 12AT7, 6J5 etc. and on the whole I have found that triode based receivers do not offer quite such good results in terms of sensitivity.

Many of these sets have used grid leak detection and whilst most have provided good results given the design limitations, none seem to cope well with music due to distortion. I attach two examples, both made recently the first using EF41s, and the second 6K7/6J7 these are from RC, January 1954 and November 1947 they both offer reasonable sensitivity, pretty good selectivity and are OK for speech, but defeated by music.

Having done a good deal of reading on the subject I "sort of grasp" why distortion is inevitable in this type of receiver. However, my limited understanding of receiver design principles (not to mention mathematics) naively prompts me to ask if some designs are better than others and is there any way of improving results or am I simply wasting my time and should stick to superhets ?
Attached Thumbnails
Click image for larger version

Name:	Voyager TRF P 4.jpg
Views:	174
Size:	92.9 KB
ID:	260355   Click image for larger version

Name:	6K7 6J7 6V6 Receiver.jpg
Views:	179
Size:	57.9 KB
ID:	260356  
E93AFAN is offline  
Old 3rd Jul 2022, 10:19 pm   #2
G.Castle
Heptode
 
Join Date: Dec 2020
Location: Swaffham, Norfolk, UK.
Posts: 582
Default Re: Distortion in Regenerative Receivers

Hi.

First of all, I'm not an expert on this, however, as you are probably aware, a regenerative detector is by its nature very non linear, also, they are easily overloaded on strong signals.

If you interest is broadcast listening, a well designed straight TRF, with a nice linear detector is a revelation in quality, if you have a good off air signal.

Not built by me, but I did experience a home brew TRF tuner with an infinite impedance detector that was based on a design from here: https://www.cool386.com/4valvetrf/4vtrf.html
G.Castle is offline  
Old 4th Jul 2022, 3:52 am   #3
Catkins
Pentode
 
Catkins's Avatar
 
Join Date: Feb 2015
Location: Chepstow, Monmouthshire, UK.
Posts: 234
Default Re: Distortion in Regenerative Receivers

You have answered your own question. Leaky grid detection and regeneration (reaction) dates back to the 1920s when they we're much more concerned with amplification than distortion (not surprising when the horn loudspeaker was the output device).

If you don't want distortion then you would not use those designs, because distortion is built in. Superhet designs (starting from the early 1930s) paid much more attention to minimising distortion. No leaky grid detection or regeneration was ever employed to my knowledge in a superhet circuit.

I restore and listen to 1920s TRF sets, and I expect distortion to be present especially on music. To me it is part of their charm and part of their vintage design, especially coupled with a horn loudspeaker.
Catkins is offline  
Old 4th Jul 2022, 8:08 am   #4
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,059
Default Re: Distortion in Regenerative Receivers

I did some work last year building a superhet radio which used both a leaky-grid detector AND an infinite impedance detector!

The LGD was used to derive AVC control voltage, but I did evaluate its linearity, using a ramped RF generator. See first photo, where the output trace is superimposed on the input envelope. It's clear that it isn't very linear, but as has been pointed out, it works quite well!

On the other hand, the IID is far better (second photo)

Quote:
Originally Posted by G.Castle View Post
Not built by me, but I did experience a home brew TRF tuner with an infinite impedance detector that was based on a design from here: https://www.cool386.com/4valvetrf/4vtrf.html
How was sound quality? Mine is really very good, much better than many superhets with diode detectors!

Photos are from notes. Pentodes were used for both - actually directly-heated types, which required a bit of morphing of the circuit (plus a high negative supply rail) to realise the IID.
Attached Thumbnails
Click image for larger version

Name:	C795E303-DC3E-4B67-96DA-E2F1178DE31E.jpg
Views:	103
Size:	40.8 KB
ID:	260375   Click image for larger version

Name:	4EDD71AD-8038-440B-A88A-B693A5BB67FC.jpg
Views:	105
Size:	47.4 KB
ID:	260376  
kalee20 is offline  
Old 4th Jul 2022, 3:32 pm   #5
G.Castle
Heptode
 
Join Date: Dec 2020
Location: Swaffham, Norfolk, UK.
Posts: 582
Default Re: Distortion in Regenerative Receivers

Quote:
Originally Posted by Catkins
No leaky grid detection or regeneration was ever employed to my knowledge in a superhet circuit.
The regenerative superhet definitely did exist as a topology, (technique called 'regenerodyne'), where the superhet mixer fed a regenerative detector as an IF stage, (often reflexed too, just to add a little more distortion!) It was almost certainly done to save on valve count. The advantage if done correctly I suppose would be that as the regenerative stage worked on only one frequency it would be better behaved, I don't believe the realisation was that simple. Google is your friend here.

Quote:
Originally Posted by Kalee20

Quote:
Originally Posted by G.Castle
Not built by me, but I did experience a home brew TRF tuner with an infinite impedance detector that was based on a design from here: https://www.cool386.com/4valvetrf/4vtrf.html
How was sound quality? Mine is really very good, much better than many superhets with diode detectors!
Very good also Peter, especially at low input and quiet passages. The Australian designers audio stage was left off and was just fed through my Mullard 3-3.

Not sure of the valve type used, knowing the YL who constructed it, it was probably the designers first choice of EF80's: there's half a tea chest full of them in my cellar at home even now that belong to her, I think they breed!

I expect the IID wasn't favored by set makers for commercial superhet radios because they saw it as an uneconomic design.
G.Castle is offline  
Old 4th Jul 2022, 4:25 pm   #6
ms660
Dekatron
 
ms660's Avatar
 
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
Default Re: Distortion in Regenerative Receivers

Quote:
Originally Posted by G.Castle View Post
The regenerative superhet definitely did exist as a topology, (technique called 'regenerodyne'), where the superhet mixer fed a regenerative detector as an IF stage, (often reflexed too, just to add a little more distortion!) It was almost certainly done to save on valve count. The advantage if done correctly I suppose would be that as the regenerative stage worked on only one frequency it would be better behaved, I don't believe the realisation was that simple.
The Regenerodyne could work across a band of frequencies, the IF in one of the ones I built had a tunable range of 500kHz centered on 2.75MHz, another version had an IF tunable range of 1MHz centered on 2.5MHz, they worked extremely well, the regen detector tuning rate being constant for any band selected, I used a bench signal generator for the oscillator injection to the mixer to set the band being tuned by the regen detector.

I used it for SW listening, AM, CW and SSB.

Lawrence.
ms660 is offline  
Old 4th Jul 2022, 5:05 pm   #7
G.Castle
Heptode
 
Join Date: Dec 2020
Location: Swaffham, Norfolk, UK.
Posts: 582
Default Re: Distortion in Regenerative Receivers

So, using the mixer as band set, and the tuning of the IF as band spread?
G.Castle is offline  
Old 4th Jul 2022, 5:15 pm   #8
Hartley118
Nonode
 
Hartley118's Avatar
 
Join Date: Aug 2015
Location: Cambridge, Cambs. UK.
Posts: 2,196
Default Re: Distortion in Regenerative Receivers

I’ve come across one or two examples of the ‘All American Five’ superhet where the compact design has resulted in some regeneration in the IF stage caused, I guess by stray capacitance coupling. This of course can be beneficial to the set’s sensitivity, helping it to function on just a short wire aerial. I often wonder whether this regeneration is a deliberate design feature incorporated in the tight component layout, or is it merely a happy accident?

Martin
__________________
BVWS Member
Hartley118 is online now  
Old 4th Jul 2022, 6:14 pm   #9
ms660
Dekatron
 
ms660's Avatar
 
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
Default Re: Distortion in Regenerative Receivers

Quote:
Originally Posted by G.Castle View Post
So, using the mixer as band set, and the tuning of the IF as band spread?
Yes, with the one that I built if wanted to listen to say the 40 metre ham band I would set the oscillator (my RF sig. gen.) to 10MHz and then tune the regen. detector, the detector in one version I built tuned from 2.5MHz to 3MHz therefore with the oscillator at 10MHz the coverage would be from 7MHz to 7.5MHZ which covered the 40 meter ham band plus some extra, the detector tuned circuit in addition to its main tuning capacitor had a small value variable capacitor across the tuned circuit which was used to fine tune CW and SSB transmissions.

To set the RF sig. gen. spot on I used the harmonics from a Xtal marker that radiated out from a Murphy B40D receiver that I had close by.

The receiver had an RF amp, it's grid and the mixer signal grid was tuned with a twin gang tuning capacitor, so basically set the oscillator then peak the RF and tune the detector.

The plan was to use a SLF tuning capacitor for the detector tuning and a suitable scale with equal graduations but I never got round to it.

Have to say it was the best performing "regen" that I ever built.

Lawrence.
ms660 is offline  
Old 4th Jul 2022, 9:13 pm   #10
E93AFAN
Tetrode
 
Join Date: Mar 2022
Location: Chippenham, Wiltshire, UK.
Posts: 69
Default Re: Distortion in Regenerative Receivers

Thanks everyone for your comments and for confirming my understanding.

My question was not quite as dumb as it might seem, my father (born 1893) was a radio engineer by profession and as a boy in the 1950s I had access to a number of 1920s/1930s commercial regens which he acquired when they became obsolete and unwanted after WW2. My recollection is that music from these sets sounded just fine, if "warmer" in tone than we would expect these days. But I have recently come to suspect that this is because I had no reference against which to judge fidelity. Clearly I am not "missing a trick" and so my query has been answered.

G Castle - thank you for your kind recommendation, I have visited this site a couple of times and its interesting, may give this one a go I think as I have plenty of EF41s that would sub quite well for EF80s I'm hoping.

Hartley 118 - I recently restored my mothers Roberts RP4, which had laid unused since the 1980s. It was always super sensitive on SW, the reason I discovered when I restored it is that it is on the edge of regeneration over much of the band. I feel certain this is not a design feature but it does make for a great little receiver .
Attached Thumbnails
Click image for larger version

Name:	Roberts RP4 - 06.jpg
Views:	90
Size:	55.5 KB
ID:	260427   Click image for larger version

Name:	Roberts RP4 - 03.jpg
Views:	90
Size:	45.9 KB
ID:	260428  
E93AFAN is offline  
Old 4th Jul 2022, 9:26 pm   #11
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,059
Default Re: Distortion in Regenerative Receivers

Quote:
Originally Posted by G.Castle View Post
Quote:
Originally Posted by Catkins
No leaky grid detection or regeneration was ever employed to my knowledge in a superhet circuit.
The regenerative superhet definitely did exist as a topology, (technique called 'regenerodyne'), where the superhet mixer fed a regenerative detector as an IF stage, (often reflexed too, just to add a little more distortion!) It was almost certainly done to save on valve count
Agree - an example being the Marconiphone T43DA. It uses UCH81, UF89, UL84 and UY81, and the UF89 was, not an IF amplifier, but a leaky-grid detector (with fixed regeneration).
kalee20 is offline  
Old 4th Jul 2022, 10:00 pm   #12
G.Castle
Heptode
 
Join Date: Dec 2020
Location: Swaffham, Norfolk, UK.
Posts: 582
Default Re: Distortion in Regenerative Receivers

Quote:
Originally Posted by ms660 View Post
Quote:
Originally Posted by G.Castle View Post
So, using the mixer as band set, and the tuning of the IF as band spread?
Yes, with the one that I built if wanted to listen to say the 40 metre ham band...
...Have to say it was the best performing "regen" that I ever built.

Lawrence.
Ingenious set Lawrence! Certainly a cut above a three valve domestic receiver, nevertheless would need more than the average punters understanding to drive it.

Quote:
Originally Posted by kalee20 View Post
Quote:
Originally Posted by G.Castle View Post
Quote:
Originally Posted by Catkins
No leaky grid detection or regeneration was ever employed to my knowledge in a superhet circuit.
The regenerative superhet definitely did exist as a topology, (technique called 'regenerodyne'), where the superhet mixer fed a regenerative detector as an IF stage, (often reflexed too, just to add a little more distortion!) It was almost certainly done to save on valve count
Agree - an example being the Marconiphone T43DA. It uses UCH81, UF89, UL84 and UY81, and the UF89 was, not an IF amplifier, but a leaky-grid detector (with fixed regeneration).
Thanks Peter, that's a good example of what I was thinking of.

I'm custodian of several sets that wouldn't be quite as good receivers as they are, were it not for a bit of unofficial regeneration as E93AFAN says of the RP4...
G.Castle is offline  
Old 5th Jul 2022, 10:42 am   #13
ms660
Dekatron
 
ms660's Avatar
 
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
Default Re: Distortion in Regenerative Receivers

Quote:
Originally Posted by G.Castle View Post
Ingenious set Lawrence! Certainly a cut above a three valve domestic receiver, nevertheless would need more than the average punters understanding to drive it.
It only had one more valve, it was very easy to drive, a schematic of a prototype is attached ('scooze the scrawl) Note that it was cobbled together only from what I had in my scrap box so the values shown may not be optimum, nevertheless it worked extremely well considering and had very smooth regen. action with very little drift when advancing the regen. control to resolve CW and SSB.

I picked up on the idea from Gary Johanson.

Lawrence.
Attached Files
File Type: pdf regenerodyne3.pdf (447.7 KB, 90 views)
ms660 is offline  
Old 5th Jul 2022, 12:04 pm   #14
G.Castle
Heptode
 
Join Date: Dec 2020
Location: Swaffham, Norfolk, UK.
Posts: 582
Default Re: Distortion in Regenerative Receivers

Thanks Lawrence!

I'll download it and have a look when I'm next at the PC. It's unusual to find a regen circuit that the tunning is steady at threshold of oscillation for CW particularly.

Cheers!
Greg.
G.Castle is offline  
Old 5th Jul 2022, 12:22 pm   #15
ms660
Dekatron
 
ms660's Avatar
 
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
Default Re: Distortion in Regenerative Receivers

Attached is a link to another schematic, in this one I was experimenting with a harmonic generator to set the band at 1MHz intervals, again I was limited to the components I had to hand so the component values etc might not be ideal but the concept did work, unfortunately due to a change in personal circumstances I wasn't able to develop it further and put it all on a nice neat chassis and put it in a case.

Lawrence.
Attached Files
File Type: pdf regenerodyne4.pdf (568.2 KB, 84 views)
ms660 is offline  
Old 5th Jul 2022, 5:36 pm   #16
G.Castle
Heptode
 
Join Date: Dec 2020
Location: Swaffham, Norfolk, UK.
Posts: 582
Default Re: Distortion in Regenerative Receivers

Thanks again Lawrence. I've now had a look at the schematics, the penny has now dropped regarding the harmonic generator, (circuit "4").

Also noted that you're fond of designing with the metal envelope valves!

The audio stage, (my comfort zone of understanding), seems a very sensitive design, with the 6F6 preceded by the triode voltage amp. ISTR that the 6F6 was often used in commercial designs straight after the detector as it had sufficient gain.

Thanks again, most interesting.
G.Castle is offline  
Old 5th Jul 2022, 5:59 pm   #17
ms660
Dekatron
 
ms660's Avatar
 
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
Default Re: Distortion in Regenerative Receivers

No problem.

Lawrence.
ms660 is offline  
Old 19th Jul 2022, 11:09 am   #18
G6Tanuki
Dekatron
 
G6Tanuki's Avatar
 
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,951
Default Re: Distortion in Regenerative Receivers

This was also known as the "Super-gainer" design; basically a frequency-changer followed by a fixed-frequency regenerative detector; I built one of these for 80 and 40 Metres using an EF39 RF amp, ECH35 frequency-changer, EF39 regen detector and 6V6 output.

IF was 1.6MHz, and the front-end used Denco coils. A low-value [50pF] tuning-capacitor padded out allowed a good tuning-rate on 3.5 and 7MHz; The big thing about it compared to traditional regenrative-detectors was that I could optimise the regeneration so it slipped gently into oscillation [letting me resolve CW/SSB] and there was much reduced 'swaying-antenna-causing-the-tyuning-to-vary' or hand-capacitance effects compared to a traditional regen.

By their nature, regenerative detectors are never going to be low-distortion devices; detection is after all a 'kind' of distortion; when combined with amplification in the same valve all sorts of odd interactions occur.

[ Thought: has anyone ever tried separating the detection and regenerative-amplification functions? I'm thinking here of a regenerative RF amplifier feeding into a traditional diode-type detector? ]
__________________
I'm the Operator of my Pocket Calculator. -Kraftwerk.

Last edited by G6Tanuki; 19th Jul 2022 at 11:21 am.
G6Tanuki is offline  
Old 19th Jul 2022, 12:54 pm   #19
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,059
Default Re: Distortion in Regenerative Receivers

Quote:
Originally Posted by G6Tanuki View Post
[ Thought: has anyone ever tried separating the detection and regenerative-amplification functions? I'm thinking here of a regenerative RF amplifier feeding into a traditional diode-type detector? ]
I seem to remember John Scott-Taggart's overly-complex circuits where he wrapped regeneration around RF amplifiers, and also around leaky-grid detectors, in the same circuit, each controlled by its own knob... So you could, if you wanted, set detector regen to zero and RF regen to quite a lot, to do what you ask (except it wouldn't be a diode detector).
kalee20 is offline  
Old 19th Jul 2022, 1:58 pm   #20
ms660
Dekatron
 
ms660's Avatar
 
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
Default Re: Distortion in Regenerative Receivers

My last Regenerodyne did that by controlling the g2 voltage on the RF Amp, sharpened up the RF response nicely.

Lawrence.
ms660 is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 12:59 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.