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

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Radio (domestic)

Notices

Vintage Radio (domestic) Domestic vintage radio (wireless) receivers only.

Closed Thread
 
Thread Tools
Old 10th Nov 2017, 2:05 am   #41
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Mt. Maunganui, New Zealand
Posts: 2,089
Default Re: Oh No! Not another howler.

Quote:
Originally Posted by Boater Sam View Post
Diode to pentode grid is probably the pathway that this set suffered from, it oscillated in isolation whenever there was HT.
There is no neutralisation that I can recognise, my grid stopper works perfectly.
How would you neutralise this stage in the conventional sense?
That AWV Radiotronics article I mentioned earlier shows how to do diode-pentode neutralization, including how to determine the capacitor sizes needed. Here is the basic circuit from that article:

Click image for larger version

Name:	from AWV Radiotronics 118 194603 p.33.png
Views:	110
Size:	175.2 KB
ID:	152036


Cheers,
Synchrodyne is offline  
Old 10th Nov 2017, 9:43 am   #42
Boater Sam
Dekatron
 
Boater Sam's Avatar
 
Join Date: Dec 2013
Location: Middlewich, Cheshire, UK. & Winter in the Philippines.
Posts: 3,534
Default Re: Oh No! Not another howler.

Ht is high but it still howls with the lamp limiter in, when the HT is low.
AGC is only on the Mixer.
If I get another I will try neutralising, there is nothing like it around this IF valve, in fact if you look at the circuit, there is NoThInG!
__________________
Boater Sam.
BVWS Member
Boater Sam is online now  
Old 10th Nov 2017, 9:48 am   #43
ms660
Dekatron
 
ms660's Avatar
 
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 6,885
Default Re: Oh No! Not another howler.

The IF amp uses AGC too, or are you saying that the AGC to the IF has been disconnected?

Lawrence.
ms660 is online now  
Old 10th Nov 2017, 11:42 am   #44
turretslug
Nonode
 
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 2,478
Default Re: Oh No! Not another howler.

ISTR the Quad AM2 instruction book has a diagram and explanation of the neutralising bridge arrangement used around the IF stage- though a modern (then!) screened EBF89 pentode, there was still enough stage gain to make this necessary for consistent performance. This was a set with an RF amp, too- the implication being that a set without RF stage would have placed even more emphasis on IF gain.

Can you imagine the instructions for a modern piece of consumer electronics going into such depth of technical detail? Even the plug fuse is pretty much regarded as forbidden mystique!
turretslug is offline  
Old 10th Nov 2017, 9:14 pm   #45
Skywave
Dekatron
 
Skywave's Avatar
 
Join Date: Jun 2006
Location: Chard, south Somerset, UK.
Posts: 5,989
Arrow Re: Oh No! Not another howler.

Quote:
Originally Posted by Boater Sam View Post
AGC is only on the Mixer.
Really? I am surprised. Moreover, it was (and probably still is) considered unwise to apply AGC to the mixer stage, since the application of AGC to a mixer can cause detuning effects.

Quote:
Originally Posted by Boater Sam View Post
If I get another I will try neutralizing: there is nothing like it around this IF valve.
Which, of course, may help. However, I assume that the original design did not have such a feature and, presumably when new, was not unstable, either - as it is now . . . .

In conclusion, it seems that you either have a common fault that even with upmost due diligence, you've missed the cause of (no disrespect implied: I've been there myself ), or that some strange & bizarre age-related fault has arisen. Perhaps the original design was so marginal that a very slight change in some 'condition' - which could be of a mechanical nature - has now caused the current fault.

Al.
__________________
"Your idea is crazy! But the question is this: is it crazy enough . . . . to be true?" (Niels Bohr)
Skywave is offline  
Old 10th Nov 2017, 9:30 pm   #46
ms660
Dekatron
 
ms660's Avatar
 
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 6,885
Default Re: Oh No! Not another howler.

Triode hexode/heptode mixers are usually under AGC control, some pentagrid mixers weren't.

Lawrence.
ms660 is online now  
Old 10th Nov 2017, 9:32 pm   #47
Boater Sam
Dekatron
 
Boater Sam's Avatar
 
Join Date: Dec 2013
Location: Middlewich, Cheshire, UK. & Winter in the Philippines.
Posts: 3,534
Default Re: Oh No! Not another howler.

The AGC is, of course, applied to the IF valve & mixer G1. It is via 1M5 and 680K resistors respectively, which are within spec.

The only components that have not been tested thoroughly are the IF transformers and their padding caps, they are still sealed from the factory and the gain seems to be very good, not twiddled.
I avoid unnecessary alignment efforts, had too many cores fall off the carriers.

The set is cleaned, complete and finished with its extra component in place. It is working as well as every other example of this model I've done.

I've pondered and worked on it long enough.
__________________
Boater Sam.
BVWS Member
Boater Sam is online now  
Old 10th Nov 2017, 9:35 pm   #48
Radio Wrangler
Dekatron
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 10,741
Default Re: Oh No! Not another howler.

It might be easier to stop using the diode section and add a solid state diode if diode anode capacitance to the pentode is problematical.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 10th Nov 2017, 9:44 pm   #49
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 4,569
Default Re: Oh No! Not another howler.

Quote:
Originally Posted by Skywave View Post
Quote:
Originally Posted by Boater Sam View Post
AGC is only on the Mixer.
Really? I am surprised. Moreover, it was (and probably still is) considered unwise to apply AGC to the mixer stage, since the application of AGC to a mixer can cause detuning effects.
That's true, AGC on the frequency changer can cause slight change in oscillator frequency as AGC voltage varies. So, from a design point of view, safest not to take AGC to this valve.

Also, the final IF valve, if biased hard back by AGC, signal-handling capability is reduced. And of course, that's just when you want it, when dealing with a strong signal generating lots of AGC. So, it's safest to not apply AGC to the final IF valve.

Trouble is, if you follow both these bits of advice, in the standard 4+1 valve superhet, there's not exactly much left to apply AGC to...

So, in practice, valves intended to be used together have similar grid bases so they both reduce gain gracefully together. With gain control of two stages, you get enough control without the FC being biased hard enough to wildly affect frequency, or the IF to affect inability to drive the detector.

Like Al, I am suprised it s only on the mixer. As a design improvement, I'd be inclined to try applying to the IF valve too. It doesn't reveal whether the tendency to oscillation in this radio is a design fault or not, but you may end up with a better-performing radio for your trouble!
kalee20 is offline  
Old 10th Nov 2017, 9:49 pm   #50
Boater Sam
Dekatron
 
Boater Sam's Avatar
 
Join Date: Dec 2013
Location: Middlewich, Cheshire, UK. & Winter in the Philippines.
Posts: 3,534
Default Re: Oh No! Not another howler.

Please see post 47.
__________________
Boater Sam.
BVWS Member
Boater Sam is online now  
Old 10th Nov 2017, 10:23 pm   #51
turretslug
Nonode
 
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 2,478
Default Re: Oh No! Not another howler.

The Chapman S6BS here applies full AGC to the mixer (ECH81 heptode) on MW, then progressively pots it down as the bandswitch goes higher in frequency, presumably in the interest of frequency stability- whilst the whole point of the triode-hexode was to reduce frequency -pulling at HF in consumer sets, the change in effective inter-electrode capacitance with signal grid bias can still have an effect. (It's not shown on the circuit diagram, where the coil-pack and switching is shown as a block, but it's obvious on inspection). This set does have the luxury of 1x RF, 2x IF and AGC amplifier stages, though.
turretslug is offline  
Old 11th Nov 2017, 11:40 pm   #52
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Mt. Maunganui, New Zealand
Posts: 2,089
Default Re: Oh No! Not another howler.

Reduced AGC on the mixer for the HF bands is an interesting aspect of the Chapman S6BS that as you say is not shown in the schematic nor mentioned in the brochures. I am not so surprised though, as that model was CTC’s pièce-de-résistance with much attention to detail, and which, with some changes although not to the core design, stayed in production from 1953 to c.1969. The AGC IF amplifier sidechain allowed the application of full AGC bias to the final IF stage without getting into modulation rise distortion, and also avoided differential distortion. In that case the EBF80 did not need neutralization, perhaps because the demodulator diode anode was fed from the 2nd IF stage whereas the pentode grid was fed from the AGC sidechain IF stage.

Re the Quad AMII, the detailed circuit description was also included in the sales brochure. This was something of a “textbook” design from John Collinson. It had fractional AGC (about a quarter) on the IF stage to minimize modulation rise distortion, but given that it had an RF stage, there was room to do this without unduly compromising the AGC range. I suspect that the small case size mitigated against having a large valve count for heat build-up reasons, hence the single high-gain (and neutralized) IF stage rather than two lower gain stages. Still, there was more than enough gain for the primary purpose of high-quality MF band listening. Here is the AMII EBF89 neutralization circuit:

Click image for larger version

Name:	Quad AMII Neutralization Circuit.jpg
Views:	48
Size:	50.0 KB
ID:	152132

Basically, C32 neutralizes the EBF89 pentode anode-to-grid capacitance, and C37 neutralizes the diode anode-to-pentode grid capacitance. It follows the AWV Radiotronics circuit, which was also shown by Langford-Smith.

As kalee20 has said, the designer of a simple MF/LF receiver with just a mixer and a single IF amplifier was between a rock and a hard place when it came to AGC, and the usual approach seems to have been to apply full AGC bias to both stages. Sliding screen bias on the IF stage evidently helped somewhat in reducing modulation rise distortion. A high-gain IF stage would be preferable, but then this would likely require neutralization, perhaps not welcome when cost reduction was the objective. Degeneration by omitting the IF stage cathode resistor bypass capacitor was another way of reducing modulation rise, and it saved a capacitor, but that doesn’t seem to have been much used; the loss of gain might have been a concern.


Cheers,
Synchrodyne is offline  
Old 12th Nov 2017, 8:10 am   #53
peter_sol
Octode
 
Join Date: Aug 2003
Location: Welwyn Garden City, Herts. UK.
Posts: 1,485
Default Re: Oh No! Not another howler.

Have you tried Bridging the smoothing caps? I know you say there is no hum but I was fooled with one of these once.
peter_sol is offline  
Old 12th Nov 2017, 10:17 am   #54
Boater Sam
Dekatron
 
Boater Sam's Avatar
 
Join Date: Dec 2013
Location: Middlewich, Cheshire, UK. & Winter in the Philippines.
Posts: 3,534
Default Re: Oh No! Not another howler.

All caps have either been changed, or temporarily substituted, except the micas in the front end.
Reservoir and smoothing caps have been reformed but they were fine even before. They usually are in these sets provided they have been used in the last 10 or so years!

The lack of any neutralisation on the stage is probably just a cost cutting measure, they got away with using as few components as possible.
__________________
Boater Sam.
BVWS Member
Boater Sam is online now  
Old 12th Nov 2017, 10:47 am   #55
peter_sol
Octode
 
Join Date: Aug 2003
Location: Welwyn Garden City, Herts. UK.
Posts: 1,485
Default Re: Oh No! Not another howler.

I would still try the smoothers.
peter_sol is offline  
Old 12th Nov 2017, 10:55 am   #56
Argus25
Octode
 
Argus25's Avatar
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 1,656
Default Re: Oh No! Not another howler.

I don't know if these remarks are going to be of any help, but maybe.

I once had a valve radio that was entirely original in its circuit configuration. All the tubes were normal on individual tube testing. All the waxies had been replaced and new electrolytics also and all the connections and earths were fine. Yet it still had instability with RF feedback issues causing the IF to oscillate under some circumstances, making it a "howler".

I searched for the feedback pathway that was causing this and came up empty handed, until I fitted 0.1uF capacitors on both the heater terminals to ground of every valve in the set. This cured it, it was microscopic heater cathode leakage or coupling causing the trouble.

On the other hand it is possible, that a single IF stage will still oscillate, even with what appears to be perfect screen grid bypassing as the screen grid as a "shield" is close to, but not 100% perfect. So with tuned circuits in the anode & plate, they can still exchange energy with each other & encourage oscillations. Try putting the screen bypass caps in the IF stages from the screen to the cathode rather than to ground and put a 1K resistor in series with the screen supply if there is not one already. Watch out for gassy tubes too which can misbehave in this manner.

Lowering the IF stage gain with a 47k or 100k resistor across the IF coil secondary is not a bad idea either, though it will broaden the IF bandwidth a tad.

One trick to settle an IF stage instability down, if you want to give it a punt, wrap a single turn around the IF transformer (usually in the space between the two coils) and put that turn in series with the IF tube's cathode. (if the polarity is wrong it will oscillate) if right polarity it will suppress oscillations.
Argus25 is offline  
Old 12th Nov 2017, 11:00 am   #57
ms660
Dekatron
 
ms660's Avatar
 
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 6,885
Default Re: Oh No! Not another howler.

There's always the old idea of trying to find what the fault is, surely all AC41's can't be doing this.

Lawrence.
ms660 is online now  
Old 12th Nov 2017, 12:03 pm   #58
Argus25
Octode
 
Argus25's Avatar
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 1,656
Default Re: Oh No! Not another howler.

Probably most AC41's don't do it, but with a borderline condition such as a subtle amount of tube gas or heater cathode leakage, its enough to trip it off. I agree though, all the usual suspects should be checked first. It might be though that in this set, one of the IF coils was made with accidental reverse polarity, could be just enough to do it in conjunction with other factors.
Argus25 is offline  
Old 12th Nov 2017, 12:30 pm   #59
ms660
Dekatron
 
ms660's Avatar
 
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 6,885
Default Re: Oh No! Not another howler.

I haven't got a receiver to experiment with but I wonder if TATG oscillator criteria might apply to this, eg: IF's slightly off tune...anode load inductive WRT grid circuit?

Lawrence.
ms660 is online now  
Old 12th Nov 2017, 12:55 pm   #60
Boater Sam
Dekatron
 
Boater Sam's Avatar
 
Join Date: Dec 2013
Location: Middlewich, Cheshire, UK. & Winter in the Philippines.
Posts: 3,534
Default Re: Oh No! Not another howler.

It is possible Lawrence that the IF may of drifted slightly with the passing years. However as the slugs are known to fall off the brass threads and they are still locked with the Bush white goo I am not going to twiddle.
There may be all sorts of reasons for this radio to misbehave. I have satisfied myself that it is not a component that I can change that is at fault, they have all been checked one way or another.
As the 8k2 grid resistor has stopped the oscillation and the set is as sensitive and selective as the 2 others of the same model that I have, it has been moved to the "done" pile. I am not going back to spend more time on a £20 radio that is working fine.
The next one, same model, is a later serial number, no paxolin B9A socket, otherwise the same. It doesn't howl, but it is so insensitive that it probably won't no matter what I do to it.
The IFs have been twiddled, and possibly the aerial and oscillator coils and trimmers.

After this second AC41, if there is time before we go away for winter I have a choice, either a McMichael 135 or an 808 or even battery Cossor.

All good comments from you all, thank you. There is a wealth of experience out there.
__________________
Boater Sam.
BVWS Member

Last edited by Boater Sam; 12th Nov 2017 at 1:01 pm. Reason: added
Boater Sam is online now  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 5:38 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 - 2018, vBulletin Solutions, Inc.
Copyright ©2002 - 2018, Paul Stenning.