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 26th Oct 2021, 10:29 am   #81
Christoffrad
Hexode
 
Join Date: Jun 2018
Location: Cambridge, Cambridgeshire, UK.
Posts: 260
Default Re: Philips 543A problem detector stage

Thank you Radio 1950 and Paula for all your help.

I'm hoping that I have now found this fault.
I had already tested the EBF80 but Paula's comment prompted me to test it again. This time it displays a cathode/heater partial short.
So I shall replace it.
I have little experience of the symtoms of C/H faults so would be interested to hear if this distortion behaviour is typical.
The valve otherwise measures a good gm of 3

Unfortunately none of my other sets uses this valve to act as confirmation of the culprit.

Thanks again all.

Chris
Attached Thumbnails
Click image for larger version

Name:	valve fault.jpg
Views:	64
Size:	98.0 KB
ID:	244276  
Christoffrad is offline  
Old 26th Oct 2021, 11:02 pm   #82
Radio1950
Hexode
 
Join Date: Apr 2017
Location: Buderim, Queensland, Australia.
Posts: 428
Default Re: Philips 543A problem detector stage

If correct, good find.
Well done.
Understanding and perseverance.

My distant view of your fault symptoms has many variables.
I wouldn't be surprised to find another fault.

If you are ready to throw that tube in the bin as unusable, I wonder if it is possible to "zap" that tube "partial short" by capacitor discharge?

I have ideas, but no experience.

I wonder if any forum members can advise on likelihood of success, and parameters to use please?
.
Radio1950 is offline  
Old 27th Oct 2021, 8:33 am   #83
cathoderay57
Nonode
 
Join Date: Oct 2017
Location: Bristol, UK.
Posts: 2,364
Default Re: Philips 543A problem detector stage

EBF80s are fairly easily obtainable, but I wonder whether it would be worth disconnecting the diodes and substituting a couple of semiconductor diodes? I'm assuming the problem is coming from the detector stage of the valve, as opposed to the IF amplifier section. Jerry
cathoderay57 is offline  
Old 27th Oct 2021, 10:47 am   #84
Christoffrad
Hexode
 
Join Date: Jun 2018
Location: Cambridge, Cambridgeshire, UK.
Posts: 260
Default Re: Philips 543A problem detector stage

Thanks for comments.
Paula PM'd me with the following comment (I trust she doesn't mind me posting it here)


"Thank you for the update. I wonder now if you have two faults, the valve plus something else.

The heater cathode leakage may cause some modulation of the cathode current at 50Hz and that will make the audio sound rough. The cathode is earthed, which usually keeps the problem to a minimum, however. You may be able to see the hum if you select a timebase speed of 10ms per division and examine the waveform at the normal audio monitoring point in the detector circuit.

If there is a second problem, which intermittently reduces the signal level, any hum will be proportionally higher with respect to the wanted signal. That will make it more noticeable. When the signal goes back to the normal level, the hum appears diminished, so the distortion seems lower.

A fault in the AGC capacitor will increase the level of the output if it is shorting. If it goes open circuit, the distortion would be much worse, but you've already tried a different without any improvement. I don't think the AGC capacitor is causing any issues.

From your observations, unless the click temporarily was fixing the heater cathode leak in the EBF80, I suspect the capacitor in the IF transformer may be playing up.

I ran a quick LTSpice simulation to see what happens if C47 (Philips' number) reduces in value. The audio level does decrease when that happens and signal to hum ratio will become worse. If all else fails, it may be worth checking C47."

Last edited by Christoffrad; 27th Oct 2021 at 11:07 am.
Christoffrad is offline  
Old 27th Oct 2021, 11:01 am   #85
Christoffrad
Hexode
 
Join Date: Jun 2018
Location: Cambridge, Cambridgeshire, UK.
Posts: 260
Default Re: Philips 543A problem detector stage

I have also re-examined the ECH81 frequency changer today and, on the valve tester both sections seem marginal so i plan to replace that too.

Opening up the IFT is a bit of a last resort because if these are like the one inside another Philips radio I worked on , they are rather delicate. The ferrite/iron dust cores can come adrift from the screw ends.

One observation I made is that the switch that shorts the FM secondary winding of the first FM IFT (in AM mode) has no effect. (The FM and AM windings are in series) If I open the switch there is no noticeable difference. This may support your theory that something is amiss with the first AM IFT.

In response to Paula's comment about C47 (C38 Trader) I have tested it. It measures 43pf so it's OK

Last edited by Christoffrad; 27th Oct 2021 at 11:08 am.
Christoffrad is offline  
Old 29th Oct 2021, 1:05 pm   #86
Christoffrad
Hexode
 
Join Date: Jun 2018
Location: Cambridge, Cambridgeshire, UK.
Posts: 260
Default Re: Philips 543A problem detector stage

Whilst I await replacement valves I decided to try an experiment of replacing the diode portion of the EBF80 detector (which I now know has the H/C leak) with a separate diode in the form of an EB91...... The result is the same. i.e. I still have low detected audio with some bass related distortion.
Christoffrad is offline  
Old 2nd Nov 2021, 4:16 pm   #87
Christoffrad
Hexode
 
Join Date: Jun 2018
Location: Cambridge, Cambridgeshire, UK.
Posts: 260
Default Re: Philips 543A problem detector stage

Today the two replacement valves arrived V2 - ECH81 and V3 - EBF80
My hopes wre not high so I was not surprised when there was no improvement.
Admittedly these are not new valves but I am now completely out of ideas except to think that Paula's thought that there may be a problem with one of the IFTs (or the associated C)

Ideas please!

Chris
Christoffrad is offline  
Old 2nd Nov 2021, 6:53 pm   #88
Mr 1936
Heptode
 
Join Date: Mar 2017
Location: Romsey, Hampshire, UK.
Posts: 511
Default Re: Philips 543A problem detector stage

Hi Christoffrad

I have been trying for a long time to follow this torturous trail ! I'm sure you have the sympathies of a whole group of forum members.

Since most of the obvious checks have already been suggested, I am going to put forward some more offbeat ideas which may or may not help.

To summarise your problems, you have distortion on AM only with a Sig Gen at >80% modulation level or on broadcast signals, plus an intermittent loss of sensitivity but still with distortion.

It seems to me that the AGC system is working OK in the sense that it establishes a constant (ish) carrier level at the detector input. The AGC reference level should be chosen by the designer such that there is enough headroom to cope with 100% modulation peaks, i.e. double the level of the steady carrier. I wonder if it's possible that for whatever reason the pre-detector stage cannot quite supply this much peak level without compression? It's also possible that at least some broadcast stations are using sophisticated transmitter compression that effectively gives over 100% modulation on the positive peaks, the carrier being dynamically nudged up momentarily to avoid negative peak clipping.

Is there anything that might be hampering the ability of the pre-detector stage to give sufficient IF drive to the detector ?

If it's fault free but still can't give enough drive, it might be necessary to modify the AGC servo loop so that the pre-detector carrier sits a bit lower, effectively giving more headroom. I haven't thought through how to achieve this, mind you. Maybe before diving in to do this you could prove a point by running the AGC line from an external variable negative DC supply (to give manual gain control) and see if you can clear the distortion albeit at the expense of recovered audio level.

Another minor point is that the IF passband response of the radio will naturally reduce the modulation percentage of the higher audio frequencies (a little), making them slightly less problematic than the lower ones.

I'm assuming that your off air signal is coming from one dominant transmitter, and not a synchronised network. These suffer from all sorts of odd distortion effects when you are in the geographic "mush zone" between two or more transmitters.

Turning to the intermittent fault, I know how maddening these can be. If it's a bad metallic mechanical joint (bolt and tag, corroded rivet, dirty valve pin etc etc) the transient DC shock of connecting a probe can be enough to change the behaviour of the "diode junction" thus formed. This is more of a problem with dissimilar metals. Soldered joints aren't above suspicion, but in my experience are less of a problem unless made by the phantom bodger with a 5 watt iron and plumber's solder !

Good Luck.
Mr 1936 is offline  
Old 3rd Nov 2021, 8:48 am   #89
Radio1950
Hexode
 
Join Date: Apr 2017
Location: Buderim, Queensland, Australia.
Posts: 428
Default Re: Philips 543A problem detector stage

Hi again

Deja Vu.

In your post #72 you mention that with prodding around the IF wiring, you achieved a time when broadcast audio distortion disappeared.

I think you have to return to that stage of fault finding now and see if you can provoke something beneficial. This must be a high probability area to investigate thoroughly.

Have another look at my posts #78 and #79.
You haven't replied specifically to my suggestions with any results, and I think you maybe were diverted with the suspect detector/first audio valve.


Go deep.
Suspect everything.
Expect the unusual, like a valve socket pin broken inside the moulding or wafer.


Use a heat gun (low heat, and at a distance) or hair dryer to push and provoke.

Use a paint brush to brush and disturb any debris, visible or not, which may be a problem.
Use a blower to blow out any possible things like steel wool remnants and which are virtually invisible.

Use a magnifying glass and torch for high lighting levels whilst searching.

When you need a break from this avenue, try slightly prodding all those switch contacts, the myriad of contacts in band change as well as in audio, whilst listening to broadcast (nice) music.


And do you have a dummy speaker, ie resistive load?
If not make one up.
Doesn't have to be too elegant.
Best if you can add a small monitor speaker (2-3 inch) across the dummy load with a switch and 20 ohm or so fixed resistor, so you can just hear the RX audio.
You may need this soon.


The strategy of fault finding an intermittent is not well defined.
Provocation and patience are key.

Apologies if I am preaching
Good luck.
Radio1950 is offline  
Old 3rd Nov 2021, 10:54 am   #90
Christoffrad
Hexode
 
Join Date: Jun 2018
Location: Cambridge, Cambridgeshire, UK.
Posts: 260
Default Re: Philips 543A problem detector stage

Mr 1936 and Radio1950 Thank you VERY MUCH for both of your in depth considered thoughts. I will further investigate accordingly.
BTW that intermittency is no longer evident. ie touching the 1st IFT with the scope probe no longer restores good reception. The volume is just low (and distorted by bass). For example Radio 4 LW needs to be almost on max for normal listening. (I am located in Cambridge so available signal level is is normal). Speech sounds OK

Last edited by Christoffrad; 3rd Nov 2021 at 11:04 am.
Christoffrad is offline  
Old 3rd Nov 2021, 12:24 pm   #91
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: Philips 543A problem detector stage

I notice that you consistently say 'low volume' and intermittently the volume improved and the distortion reduced? The distortion appears to be a consequence of an issue with RF gain so I suggest we chase this rather than the distortion.

Have you replaced all of the paper capacitors including those in the RF/IF, although the trader makes them look mostly ceramic?

Have you carefully cleaned the waveband switch with servisol or similar?

Have you checked all the valve voltages against the service data?
PJL is offline  
Old 3rd Nov 2021, 3:07 pm   #92
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: Philips 543A problem detector stage

Also clean the earthing contacts to the tuning capacitor shaft.

Once the DC conditions are confirmed OK I would then recommend a full realignment as this should showup any faults with tuned circuits.
PJL is offline  
Old 3rd Nov 2021, 5:41 pm   #93
frsimen
Heptode
 
Join Date: Jan 2011
Location: Croydon, London, UK.
Posts: 767
Default Re: Philips 543A problem detector stage

One method of tracking down poor connections, especially around valve bases is to tap the suspect joint/pin with a wooden or plastic rod. This is particularly good at spotting problems with the contact to a pin on a valve, due to a poor valve base.
frsimen is offline  
Old 3rd Nov 2021, 7:40 pm   #94
Christoffrad
Hexode
 
Join Date: Jun 2018
Location: Cambridge, Cambridgeshire, UK.
Posts: 260
Default Re: Philips 543A problem detector stage

Today things have become a bit more exciting! .... I further examined the effect I've called intermittent. Once again (but with rather more predictability) As you will see from the data sheet the first AM and FM IFT primaries are linked in series and similarly the secondaries (FM secondary is switched out in AM mode). Using my scope probe, as I had done before, I found if first I lightly touch the primary link and then the secondary link, as well as a loud pop, the set would invariably either start working normally or revert to the low (distorted) condition. Once in either state it would remain so (even if switched off and on. After further experimented I found that it would sometimes go unstable. This made me suspect the decoupling cap of the IFT primary.
I had observed (and remarked in post 60) that the 2k2 resistor (R14 Philips / R13 Trader) IFT feed had been replaced but thought that I was reading the colour code wrongly. Now I lifted and measured that this 18k resistor which should be 2k2 measures 26k. Furthermore the associated 4n7 decouplining cap has been replaced in the past with a 1n (which I've also measured).
The set now oscillates like mad (even with these values) which it did not do before. This 26k resistor puts 175v on V2 mixer anode (should be 225v).
This set obviously has a history and someone has done some sort of bodge to make it work acceptably. Now I need to locate the root cause.

Last edited by Christoffrad; 3rd Nov 2021 at 7:49 pm.
Christoffrad is offline  
Old 4th Nov 2021, 3:29 pm   #95
Christoffrad
Hexode
 
Join Date: Jun 2018
Location: Cambridge, Cambridgeshire, UK.
Posts: 260
Default Re: Philips 543A problem detector stage

It would be nice to be able to check the tuning of this Philips IF transformer as a stand alone entity. In my working career I had access to equipment to design and tune RF filters using either network analyser or spectrum analyser + tracking generator. These devices used 50ohm termination so whilst I may be able to access a network analyser it would not provide a suitable termination. I wonder if there is a sensible method to do this.?
Perhaps the only way is to feed V2 grid (with oscillator killed and front end disconnected at SW.F1.) and scope V3 grid.

Is this a wacky idea?
BTW changing V2 anode load R14 to the correct value (2k2) makes no noticeable audio level difference.

Chris
Christoffrad is offline  
Old 4th Nov 2021, 7:59 pm   #96
frsimen
Heptode
 
Join Date: Jan 2011
Location: Croydon, London, UK.
Posts: 767
Default Re: Philips 543A problem detector stage

That's a reasonable idea. You probably don't need to disconnect the front end even for this test as the signal generator will damp the input tuned circuit with its 50 (or 75) ohm output impedance. I wouldn't worry too much about the local oscillator either, for that matter unless interference proves to be a problem.

I'm sure I don't need to tell you that the oscilloscope probe will detune the secondary of the first IF transformer, unless you can arrange some very loose coupling. An alternative approach is to measure at the anode of V3, making sure that you apply a 10k ohm damping resistor across the primary winding of the the second IF transformer. That should then allow you to see how well the first IF transformer has been aligned without the second IF transformer confusing matters.
frsimen is offline  
Old 4th Nov 2021, 11:37 pm   #97
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: Philips 543A problem detector stage

The AM IF alignment instructions are quite specific about alignment order and if someone has previously peaked everything the IF may be too narrow and the set could be marginally unstable. I suggest you follow the instructions to the letter and see if it shows up any issues or with a bit of luck it might even fix the fault.
PJL is offline  
Old 5th Nov 2021, 10:33 am   #98
Sideband
Dekatron
 
Sideband's Avatar
 
Join Date: Oct 2003
Location: Croydon, Surrey, UK.
Posts: 7,549
Default Re: Philips 543A problem detector stage

Quote:
Originally Posted by PJL View Post
I suggest you follow the instructions to the letter and see if it shows up any issues or with a bit of luck it might even fix the fault.
I agree with that. I had a problem with my Philips 470A going slightly unstable. It was my fault as I had quickly peaked the IF's. When I read through the instructions however, it was important to damp the previous IF with a resistor across the primary whilst aligning with the generator. Once I did this and followed the instructions properly, I had a stable set and much better audio.
__________________
There are lots of brilliant keyboard players and then there is Rick Wakeman.....
Sideband is offline  
Old 8th Nov 2021, 2:12 pm   #99
Christoffrad
Hexode
 
Join Date: Jun 2018
Location: Cambridge, Cambridgeshire, UK.
Posts: 260
Default Re: Philips 543A problem detector stage

I have carried out a check of 1st IF tuning using the method suggested by Paula (Frsimen) on post 96 ie with 10 k across the V3 anode load IF transformer (S32 Philips) (L23 Trader)
I killed the oscillator and open circuited the switch feeding V2 grid for this test
Feeding in 50mV emf onto V2 grid (via 0.01uF) I measured the two 6dB points for the p-p signal on V3 anode = 456kHz and 483kHz So mid point = 469.5kHz
The p-p signal on V3a is 650mV under these conditions.
So it looks like the 1st IFT is performing OK

Views please...
Chris
Christoffrad is offline  
Old 8th Nov 2021, 4:05 pm   #100
Christoffrad
Hexode
 
Join Date: Jun 2018
Location: Cambridge, Cambridgeshire, UK.
Posts: 260
Default Re: Philips 543A problem detector stage

Turning thoughts to the alignment between front end and oscillator I have made another test:
First, to check the resonance of the front end I fed a 1332kHz signal from the signal generator through 3 turns loosely coupled around the ferrite rod and placed the scope probe on X10 (unconnected) near V2grid and tuned the radio (switched off) for max pick up. I then removed the signal generator signal and switched the radio ON. I fed the scope probe into my frequency counter and placed the probe in the area of V2 but about 10mm away to pick up the oscillator without influencing it. The counter measured 1800KHz exactly. 1800 - 1332 = 468kHz which is close enough to the 469.5kHz estimated above to make me think there is not a front end - oscillator alignment problem.
Again... views please...

C
Christoffrad is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 5:46 am.


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.