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 > Specific Vintage Equipment > Vintage Radio (domestic)

Notices

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

Closed Thread
 
Thread Tools
Old 18th Oct 2021, 4:06 pm   #41
Christoffrad
Hexode
 
Join Date: Jun 2018
Location: Cambridge, Cambridgeshire, UK.
Posts: 260
Default Philips 543A detected AM audio

I've been trying to locate the problems with A.M. audio distortion (and occasional low sensitivity) for some time with this Philips 543A
Currently the sensitivity is fine but the audio distortion after the detector filter is bad.
The strange thing is that if I radiate a modulated signal (on a clear RF freq) from my Marconi TF144H generator to create a similar or lower signal to noise ratio I find that the generator modulation level has to be set at almost zero to produce a similar audio level as the broadcast signal. Strangely the radiated signal from the generator with 400Hz mod is not distorted at any signal or mod level.
I attach a thumbnail of the filtered detected audio (scale 50mV/div).
I have tried replacing AGC capacitor C49 (C24 Trader sheet)
C66 (C52) has been replaced. Is this some quirk of a faulty detector valve.
HT is only marginally low. HT decoupling and electrolytics checked.

I have taken a range of readings through the IF and mixer having first disabled the oscillator and isolated the input circuitry by isolating the SWF1 switch connection and feeding 470kHz (unmodulated) from the generator into V2 g1 via a capacitor.
The results are: (reading are P-P voltage)
Generator EMF- 5mV, V2anode-60mV, V3 grid- 50mV, V3 anode - 7V, Detector (anode)-2V
Generator EMF 50mV, V2anode - 600mV, V3 grid -420mV, V3 anode -37V, Detector (anode) -7V
Generator EMF 500mV, V2anode - 2.7V, V3 grid -2V, V3 anode - 37V, Detector (anode)- 7V
For reference I carried out the same test with a similar working Ferranti set at 5mV generator setting. This gives:
V2 Anode 50mV, V3 grid 180mV, V3 anode 5.4V, Detector 5.8V

Any comments appreciated
Chris
Attached Thumbnails
Click image for larger version

Name:	Philips 543A audio.jpg
Views:	69
Size:	24.4 KB
ID:	243683  

Last edited by Christoffrad; 18th Oct 2021 at 4:15 pm.
Christoffrad is offline  
Old 19th Oct 2021, 3:38 pm   #42
Christoffrad
Hexode
 
Join Date: Jun 2018
Location: Cambridge, Cambridgeshire, UK.
Posts: 260
Default Re: Philips 543A detected AM audio

I have made some further tests:
The audio distortion is still present if I:
1) disconnect the feed to the audio amplifier
2) use a germanium diode as a detector instead of V3 detector.
3) use the unused diode section of V3 instead.

This set occasionally also suffers from dramatic loss of sensiivity. This does not affect the distortion.

However when exhibiting this low sensitivity the electrical disturbance caused when attaching the scope probe to V3 grid resored full sensitivity.

Ideas please !!!

C
Christoffrad is offline  
Old 19th Oct 2021, 8:55 pm   #43
Radio1950
Hexode
 
Join Date: Apr 2017
Location: Buderim, Queensland, Australia.
Posts: 428
Default Re: Philips 543A detected AM audio

To enable us to perhaps comment further, can I please suggest the following.

Set Sig Gen to 1000KHz, 20 uV, 30% mod, feed to 543 RX antenna terminal, and ensure the Sig Gen earth lead is connected to RX earth
Tune in with RX and slightly re-tune SG and RX, to avoid possible BC station interference if required.

Using 10x CRO probe, monitor anode and then cathode of detector diode.

Increase Sig Gen RF level to 200 uV, and repeat.

Post four photos of above results.
Radio1950 is offline  
Old 19th Oct 2021, 11:44 pm   #44
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: Philips 543A detected AM audio

I assume your measurements are all made with the scope and are peak to peak? Even a x10 scope probe will impact the signal.

It may be that distortion is less noticeable with a fixed 400Hz tone. I assume you have tested the audio stage with the gram input or VHF and there is no distortion? Also the valve DC conditions are all OK? Is the distortion better or worse on a weak station? I guess your scope image is from C39 would be good to see that with the input suggested by Radio1950?
PJL is offline  
Old 20th Oct 2021, 2:37 am   #45
Radio1950
Hexode
 
Join Date: Apr 2017
Location: Buderim, Queensland, Australia.
Posts: 428
Default Re: Philips 543A detected AM audio

I probably should have explained my reasons for these tests, at least a little.

The first is a test of the detector at a signal level which should have some noise as well the sig gen signal, and which should also be below too much AGC action which might interfere in the front end and IF of the receiver..
It should give us an idea of detection efficiency, and whether there is enough IF amplification for the detector to function.

The second is to show how the whole receiver functions at normal local broadcast station level, and if the IF amp is normal, and if AGC is working, and if the detector can handle larger signals.

The source impedance of the second IF transformer winding (ie output winding feeding the detector) is normally relatively low, and the circuit is normally broad in tuning, so the resistance and capacitance of a 10x probe shouldn't interfere with either the actual measurement nor the receiver IF or detector action.
The probe will have a very small effect however, and this should always be in the back of your mind.
If you have a curious mind, this is like a "double slit experiment", ie you cannot measure something, without having an effect on it.

Good luck

Last edited by Radio1950; 20th Oct 2021 at 2:38 am. Reason: Old age
Radio1950 is offline  
Old 20th Oct 2021, 8:01 am   #46
PJC58-Hythe
Heptode
 
PJC58-Hythe's Avatar
 
Join Date: Sep 2010
Location: Hythe, Hampshire, UK.
Posts: 626
Default Re: Philips 543A detected AM audio

Have you done the basic stuff? Tested V3 and measured the voltages in the signal chain against the ones on the service data? Checked the values of of R15, R14 and tested C40 & C41?
__________________
Peter - M0HYT - BVWS Member - GQRP Club

http://www.retroworkshop.co.uk
PJC58-Hythe is offline  
Old 20th Oct 2021, 10:21 am   #47
Christoffrad
Hexode
 
Join Date: Jun 2018
Location: Cambridge, Cambridgeshire, UK.
Posts: 260
Default Philips detector problem

I'm hoping somebody explain to me what is going wrong with this A.M. detector.
I have attached 3 thumbnails which includes the circuit of the offending part.
I have disconnected the audio amplifier by disconnecting at pin 3 of SW,F2
I have also disconnected AGC resistor R18. There is no connection to a tuning indicator from pin 3 of SW,F2
The third thumbnail shows the signal on V3 detector anode (pin8). So my question is, why is the detected audio measured across C76 distorted?
The waveform on V3 anode circuit looks normal ie a symetrical amplitude modulation
I have tried wiring to the other diode (pin7) and also using a germanium diode, and get the same result. Audio level across C76 from a normal broadcast signal as shown in image 1 is about 200mV

Chris
Attached Thumbnails
Click image for larger version

Name:	Philips543A audio.jpg
Views:	63
Size:	29.2 KB
ID:	243798   Click image for larger version

Name:	Philips543A Detector circuit.jpg
Views:	73
Size:	52.9 KB
ID:	243799   Click image for larger version

Name:	Philips543det.jpg
Views:	64
Size:	26.7 KB
ID:	243800  
Christoffrad is offline  
Old 20th Oct 2021, 11:15 am   #48
Sideband
Dekatron
 
Sideband's Avatar
 
Join Date: Oct 2003
Location: Croydon, Surrey, UK.
Posts: 7,548
Default Re: Philips detector problem

Hmm....this could be due to the increased modulation levels seen with so many modern transmissions. Some detectors distort with more than about 65% modulation. In the good old days when this set was built, modulation levels were kept at around 30%. Adding a very small amount of forward bias to the diode might help. Try adding a 10 megohm resistor between the screen grid of the IF amplifier and the diode anode. Why the screen grid? Well the current through the valve varies with the AGC level so the amount of bias on the detector via the added 10 meg between strong signals and weak signals will vary slightly. Try it and see if it makes any difference.

This might be why your earlier check using an external modulator produced a cleaner signal....the modulation level is probably set for around 30 - 40%.

If you look at most alignment instructions for AM, they will usually advise using an AM generator modulated at 30% for the injection of signals.

The general idea with AM signals these days is LOUDER! Some locals are using nearly 100% and the old detector circuits just can't cope. There is also an awful amount of compression used and this can really show up as distortion on older sets.
__________________
There are lots of brilliant keyboard players and then there is Rick Wakeman.....
Sideband is offline  
Old 20th Oct 2021, 11:38 am   #49
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,799
Default Re: Philips detector problem

I'm not sure that there is anything wrong. You may be misunderstanding the circuit.

The detector diode anode conducts if the IF tries to drive it positive. Effectively it shorts any positive swing at this point. Consequently the IF signal voltage gets to provide current into C47, charging it in a negative direction.

When the IF tries to drive the detector anode negative, it succeeds! The anode swings negative, taking no current. The IFT secondary now has no ground connection from which to drive current into or out of C47, so the IFT thinks its load has disappeared. C47 gently discharges through R39, R17. This allows the audio voltage to decrease at a speed slow compared to the IF freq, fast enough for audio.

So the IF should show nice symmetrical AM modulated waveform.

The diode anode should show an AM modulated waveform, but with the top severely compressed.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now  
Old 20th Oct 2021, 12:00 pm   #50
ms660
Dekatron
 
ms660's Avatar
 
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
Default Re: Philips detector problem

Multiple threads don't help.

Lawrence.
ms660 is offline  
Old 20th Oct 2021, 12:41 pm   #51
unitelex
Hexode
 
Join Date: May 2005
Location: Greater Manchester, UK.
Posts: 492
Default Re: Philips 543A detected AM audio

Is it possible that the IF amp is unstable/oscillating?
This set has RF stopper ferrite beads on the heaters so stability might have been an issue found by Philips during the design stage.

If so then I would check the ferrite beads are in place and uncompromised, and that the HF decoupling on the HT is low impedance.

Chris
unitelex is offline  
Old 20th Oct 2021, 1:16 pm   #52
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,799
Default Re: Philips detector problem

Quote:
Originally Posted by ms660 View Post
Multiple threads don't help.

Lawrence.
Thanks, Lawrence I hadn't spotted that. Multiple threads create havoc as replies get scattered around and the OP hasn't a hope of tracking them all. This way he still has to do the work of sorting them out to understand what's being said, but at least the forum software puts them all in one place and sorts them by date/time.

Threads merged

The varying titles remind me of "you are in a maze of twisty little passages, all the same"

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now  
Old 20th Oct 2021, 7:29 pm   #53
Christoffrad
Hexode
 
Join Date: Jun 2018
Location: Cambridge, Cambridgeshire, UK.
Posts: 260
Default Re: Philips 543A problem detector stage

I would like to thank you all for your most helpful comments.
As I'm tied up this week (on an archaeological dig) I will carry out the tests you suggest ASAP.

Chris
Christoffrad is offline  
Old 21st Oct 2021, 8:46 am   #54
Christoffrad
Hexode
 
Join Date: Jun 2018
Location: Cambridge, Cambridgeshire, UK.
Posts: 260
Default Re: Philips 543A problem detector stage

Here are the results of the measurements suggested by Radio1950.
Note, that the detector anode is at ground so no measurements for anode.
I should also mention that even a x10 probe on the cathode has a noticeable loading effect. ( a voltage factor of >10 on the audio)

Comments please!

Chris
Attached Thumbnails
Click image for larger version

Name:	V3 cath 20uV emf x10probe.jpg
Views:	54
Size:	51.4 KB
ID:	243853   Click image for larger version

Name:	V3 20uV emf audio out across C76 x1 probe.jpg
Views:	55
Size:	51.8 KB
ID:	243854   Click image for larger version

Name:	V3 cath 200uV emf x10 probe.jpg
Views:	57
Size:	53.2 KB
ID:	243855   Click image for larger version

Name:	V3 200uV emf audio out across C76 x1 probe.jpg
Views:	55
Size:	52.3 KB
ID:	243856  
Christoffrad is offline  
Old 21st Oct 2021, 10:02 am   #55
ms660
Dekatron
 
ms660's Avatar
 
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
Default Re: Philips 543A problem detector stage

I think you've got your anode and cathode mixed up.....

The audio waveform looks ok to me.

Lawrence.

Last edited by ms660; 21st Oct 2021 at 10:11 am. Reason: extra info
ms660 is offline  
Old 21st Oct 2021, 5:56 pm   #56
Christoffrad
Hexode
 
Join Date: Jun 2018
Location: Cambridge, Cambridgeshire, UK.
Posts: 260
Default Re: Philips 543A problem detector stage

You are right Lawrence! Anode (not Cathode)

However the question remains.... Why does the broadcast audio sound so bad.

In answer to earlier comment. I have used the GRAM input to check the audio amplifier and it sounds OK. I have a question about why there is little difference in bass when the bass boost switch operates but that's another matter for another day. Other than to say that the audio distortion has the characteristic of the bass being both distorted and modulating the higher audio frequencies. And this seems to be evident on one of the imagas I posted on 20th.
Christoffrad is offline  
Old 21st Oct 2021, 7:19 pm   #57
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: Philips 543A problem detector stage

You could check the value of R19 and R29. What is it like on VHF? Are weak stations better than strong ones?
PJL is offline  
Old 21st Oct 2021, 7:25 pm   #58
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,799
Default Re: Philips 543A problem detector stage

There are lots of things to check, A useful tool in a workshop is a little audio amplifier and speaker. Something good quality but not silly. Something you can steal a bit of audio from a set and listen to without using the set's amplifier and speaker. It can eliminate a lot of territory quite quickly and gets a fair amount of use.

Beyond that you're going to be needing either distortion measurements or else a low distortion sig gen for substitution tests.

David

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now  
Old 21st Oct 2021, 7:40 pm   #59
frsimen
Heptode
 
Join Date: Jan 2011
Location: Croydon, London, UK.
Posts: 766
Default Re: Philips 543A problem detector stage

All the photos you have included showing the waveform at the output of the IF amplifier stage have had a rather lop sided looking modulation envelope with the upper part compressed. One in particular, on post 16, shows this very clearly.

Could you connect your oscilloscope to the output of the signal generator to check that the odd waveform isn't being generated there?

If it looks symmetrical, as it should, the problem is being created in the radio. That being the case, the problem is most likely around the bias arrangements for the ECH81 or the EBF80. Do check the components mentioned in post 57 by PJL, as R19 is associated with the bias of the two valves. So too are R18 and, for the ECH81 only, R10. (Trader references used)

Paula
frsimen is offline  
Old 21st Oct 2021, 10:51 pm   #60
Christoffrad
Hexode
 
Join Date: Jun 2018
Location: Cambridge, Cambridgeshire, UK.
Posts: 260
Default Re: Philips 543A problem detector stage

Thank you all.

Something I have just discovered (can't believe that I hadn't noticed before)

Someone in the past has replaced resistor R14 (R13 Trader) in the V2 anode HT circuit and has replaced the 2K2 value shown on the service data with a 18K....Very strange.

Chris
Christoffrad is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 1:09 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.