Roberts R303

[ajgriff]

The 9.48V across C14 is just the on load battery voltage and is as you'd expect with a fresh battery. In the absence of a 9.0V supply everything looks fine to me. The TR1 voltages might be considered a little low and this is probably down to resistor value drift in that part of the circuit. Again I'd leave well alone if the radio is performimg satisfactorily. Shame about the scope.

Alan

[DonaldStott]

Originally Posted by ajgriff

The TR1 voltages might be considered a little low and this is probably down to resistor value drift in that part of the circuit.

Thanks again Alan.

Easy enough to check resistor values in the TR1 part of the circuit.

I suppose the only thing of concern is that the set is a bit bright and sibilant but maybe I'm expecting too much?

[ajgriff]

Difficult to say without hearing it but the radio may well be performing within its original design parameters. It doesn't sound as if there are any issues with the RF/IF side of things or with tin whiskers, for the moment at least. The only other thing that might be worth thinking about is the fact that TR1 (BC148) is a lockfit type and these can sometimes be problematic. The non-lockfit BC108 is a direct replacement if needed. Personally I wouldn't go beyond the work you already have in hand. As you've said the owner knows where you are.

Alan

[DonaldStott]

As expected most of the resistors associated with TR1 are measuring about 12 -15% over their stated values although a couple of them are difficult to access properly!

TR1 is of course a lockfit type transistor and this set could benefit by replacing this with a BC108.

I am, however, erring on the side of caution and don't want to start messing with the tracks on the PCB unless I have to. The set is performing well although probably sub-optimally and I am going to walk away and leave well alone until such times as an actual fault occurs!

So I've just got to finish off cleaning up the case etc. then reassemble this neat little transistor radio - I'll Post some pictures when I'm finished.

Meantime plenty of other stuff waiting on my ToDo list and I have a faulty Farnell 30-4D 'scope to investigate.

Keep safe, keep well and keep healthy!

[DonaldStott]

Roberts R303 fully cleaned and refurbished - reassembled the set, switched on and then my heart sank. Another victim of Murphy's Law ...

What had been a really lively and sensitive set on both LW and MW, albeit a bit bright and sibilant, has now become a mush! Constant background hiss, very faint stations on LW and MW and all round very, very murky. I have to turn the volume control all the way up to hear anything but that also increases the background hash!

I'm assuming I must have disturbed something during reassembly - any ideas on where I should go looking first ?

So close ...!

[Davewantsone]

Did you replace those AF117s inside the Mullard can? If not use BC557 of other suitable types.
Earlier you mentioned an extra transistor. In later Roberts R200 sets they had 7 transistors even though it was a 6 transistor set. The "7th" transistor was used as a diode to produce a voltage drop across a variable potentiometer to bias the output transistors.

[Boulevardier]

Originally Posted by DonaldStott

Roberts R303 fully cleaned and refurbished - reassembled the set, switched on and then my heart sank. Another victim of Murphy's Law ...

Did you spray a lot of contact cleaner around when cleaning/reassembling - especially in RF sections and switches? If so, things may return to previous working state once it has had time to fully evaporate.

Also, have you checked whether giving it the works has disturbed the positioning of the coils on the ferrite rod?

[David G4EBT]

The fact that you can hear stations on both MW & LW if only faintly, suggests that the front end, IF and detector stages are working, if only after a fashion. Everyone has their own approach to fault finding, but my own approach would be to ascertain whether the fault lies in the audio stage or front end/IF stages. If you apply a signal tracer probe to the slider of the volume control and get clear tunable signals, that would suggest that the RF/IF/Detector stages are working as they should.

If you then inject an AF signal - 1kHz or so into the volume control slider and the audio is 'faint and mushy' that would point to a fault/faults in the audio stage.

If on the other hand, with the tracer on the slider of the volume control you only hear faint mushy signals, but with the AF injector probe on the slider you hear a loud clear tone from the speaker, that indicates that the audio stage is working as it should, and the fault is in the front end/IF stages - maybe the 'switchery', a disturbed soldered joint or some such thing.

If it is the audio stage which is the problem, you can perhaps narrow down the fault by injecting a 1kHz audio signal at various points. For example, if injected at the collector of TR3 and you get a nice tone from the speaker, that suggests that TR4 & TR5 are working correctly. If you then inject a signal at the base of TR3 and the output is faint and mushy, that points to TR3 being faulty. If however there's a nice clear tone at the base of TR3, move the injector probe to the base of TR2 & TR1 to see if that pinpoints the fault.

Also, assuming that at some point when the set was working correctly you noted the voltages at various points indicated on the circuit, you can check those again for any anomalies.

It's much better to try to discover faults through a methodical diagnostic approach then speculation based on subjective folklore, as often happens in sets that use AFxxx transistors, which - in a dead set might be fine. It could for example be a detector diode, as if often the case of Bush TR82s. The last thing you want to be doing unless you're confident that's where the fault lies, is to have to faff around with the IF module.

Just a few thoughts which I hope might help.

Good luck with it.

[DonaldStott]

Thanks for all the quick responses.

Originally Posted by Davewantsone

Did you replace those AF117s inside the Mullard can? If not use BC557 of other suitable types.

The AF117s were discussed earlier in this Thread but as there were no apparent problems (yet) it was decided not to replace them?

My understanding from reading numerous Posts is that the AF117 problem manifests itself as a "dead set" that springs to life when the I.F Module is tapped?

Originally Posted by Davewantsone

Earlier you mentioned an extra transistor. In later Roberts R200 sets they had 7 transistors even though it was a 6 transistor set. The "7th" transistor was used as a diode to produce a voltage drop across a variable potentiometer to bias the output transistors.

The "extra transistor" is a Mullard OC45 with an ORANGE covering (see picture in Post #7 abaove) - as it's not shown on the circuit diagram in the Trader Sheet I'm not sure how it's connected. Also not sure of what the colour coding standard is for transistor legs (Emitter/Base/Collector) but I've measured the following: -

RED covered leg: 4.94V

Bare leg: 5.15V

WHITE covered leg: 5.23V

Originally Posted by Boulevardier

Did you spray a lot of contact cleaner around when cleaning/reassembling - especially in RF sections and switches? If so, things may return to previous working state once it has had time to fully evaporate.

The volume pot and wave switches were spayed with contact cleaner before final reassembly - so that was when the set was working satisfactorily?

Originally Posted by Boulevardier

Also, have you checked whether giving it the works has disturbed the positioning of the coils on the ferrite rod?

I've checked very carefully and there is no evidence of the coils having been disturbed.


NOTE: The only thing I did do was to un-solder the External aerial and Earphone sockets to facilitate cleaning of the perspex fascia panel. Both these sockets were re-soldered as per original wiring.

[DonaldStott]

Originally Posted by David G4EBT

The fact that you can hear stations on both MW & LW if only faintly, suggests that the front end, IF and detector stages are working, if only after a fashion. Everyone has their own approach to fault finding, but my own approach would be to ascertain whether the fault lies in the audio stage or front end/IF stages...

Thanks David - as you know I have my Velleman K7000 Signal Tracer/Injector which will be helpful in tracing the problem?

Your systematic approach has worked in the past and will be adopted here.

As you say I don't really want to start faffing around with the IF module!

[Boulevardier]

"The volume pot and wave switches were spayed with contact cleaner before final reassembly - so that was when the set was working satisfactorily?"

I seem to remember a recent case of a forum member where putting contact cleaner on the wavechange switches caused weird problems. After a time (possibly about 24 hours) the liquid dried off, and normal working was resumed. So may be worth giving it some time in a warm place! I always think that if it recently worked OK, but after doing something it stopped working, then it's most likely the result of something that you did in between those two times, rather than a brand new fault.

Mike

[Davewantsone]

The "7th" transistor is used as a diode and usually only uses 2 legs to produce the diode function which causes a voltage drop of around 0.2V.
I have repaired lots of Roberts Radios. If you are going to repair others you need a "test" receiver. I use a old RT1 radio. I have made two connections to this radio, one in the middle of the IF between TR2 and TR3,, lead goes to a base of TR3 via a capacitor. The second goes to the volume control centre wiper again via a capacitor. It then becomes very easy to fault find on a radio. For example "test radio" set to R5 and playing normally. Faulty radio connect volume wiper to test radio with both chassis connected together. R5 comes through OK then fault is either IF or RF. Disconnect leads, Reconnect output from IF of "test radio" to bases of each IF transistors or if IF (Mullard) can to input of faulty radio. If functioning R5 should be clearly heard. If still nothing fault is in IF stage. R5 OK fault in RF stage. To check RF stage in faulty radio off tune "test radio" adjust tuning on faulty radio and at some point you should hear a station.playing on "test radio" indicating RF stage is working.
For reliability you really need to replace these AFxxx transistors especially if you are doing this as a repair for someone. They will only bring it back if it stops working and maybe blame you. (It worked for 50 years without a fault and has only worked for 3 months since you fixied it!)

[DonaldStott]

Progress so far :-

1. Applying the signal tracer probe to the slider of the volume control I do get clear tunable signals from the speaker of the Velleman K7000. As David has stated that would suggest that the RF/IF/Detector stages are working as they should.

2. When I then use the Velleman K7000 to inject an AF signal into the volume control slider I would expect a loud beep but I am getting nothing whatsoever! As David has stated that would point to a fault/faults in the audio stage!

I've cut my teeth over the last few years refurbishing mainly vintage valve radios and Dansette record players and this is my first foray into the brave new world (for me) of transistor sets! I'm learning as I go so be patient please if I sometimes revert to asking novice type questions such as the one in my Post #29 above about the colour coding standard for transistor legs?

As to other helpful suggestions: there is no evidence of excess contact cleaner on the wave switches or volume pot; all three legs of the Mullard OC45 are connected and the measured voltages are shown in Post #29 above; no evidence (so far) of AF117 problems so I'll wait until I get the set fully working again before I consider replacing these.

Next step then is to investigate the audio stage in some detail by following David's methodical diagnostic approach ... wish me luck!

[Cobaltblue]

Just a quick thought have you checked the earphone jack?

The types used on some of these Roberts (enclosed plastic) are truly awful and moving the wiring can make them go high resistance.

Cheers

Mike T

Originally Posted by G6Tanuki

From memory - chassis is held into the case by a couple of screws through the side-plates, concealed by the carry-handle [some versions use the same screws to hold the carry-handle to the side-plates].

Remove these, and the chassis drops-out from below.

On the R303, the chassis comes out through the top. There are screws in the side plates which go into wooden struts which are designed to go around the handle fixings, same as the RIC 1. You unscrew both, and the chassis comes out through the top of the case, not through the bottom like on say a R600.

[ajgriff]

As Mike says the earphone wiring often causes problems as it's a switched socket intended to disconnect the speaker when the plug is inserted.

Alan

[DonaldStott]

Originally Posted by Cobaltblue

Just a quick thought have you checked the earphone jack?

Thanks Mike and Alan

Of course I'm going to ask the obvious question - how do I check the earphone jack.

I know I've disturbed the wiring around this and the External aerial socket so it's worth investigating.

As an aside, if it's never going to be used, can it be safely bypassed?

Other areas of wiring that I have disturbed include the volume pot and the speaker. I'll go back and check/re-do those connections just in case?

[David G4EBT]

Originally Posted by DonaldStott

Of course I'm going to ask the obvious question - how do I check the earphone jack.

Given that by using your Velleman K7000 in signal tracer mode, you have determined that the front end, IF and detector are all working fine, but the audio stage is dead when a signal is injected to the audio stage, you can use you tracer to listen for audio signals at the input to the earphone socket as shown in the sketch below, with your ground clip to the - VE ground, and your probe tip to the input to the earphone socket (C13/R15 junction). Hopefully, you will hear signals there if, as is suspected, the earphone socket is open circuit, which is a known fault area. The socket is a 'switch' which - when an earphone plug is inserted, cuts the signal to the speaker and diverts it to the earphone. If at some time an earphone has been used, the insertion of the plug can weaken the spring a little, the contacts become tarnished and hence open circuit or intermittent.

If you do hear signals at the junction of C13/R15, you can try to clean the earphone socket with switch cleaner to see if it come back to life, but if that's not successful and you want to bypass the socket, you need to wire across points A and C as shown on the circuit so that the audio signals from C13/R15 go direct to the speaker.

If you don't hear any signals at C13/R15, you need to use your tracer probe to progressively check back towards the slider of the volume control at the emitters and bases of TR4&5, and the collectors and bases of TR3, TR2 & TR1 to see where the audio signals are being lost. (See two red arrows as examples).

The sketch below might help.

Good luck with it.

[DonaldStott]

Thanks again David - this is turning into the sort of mystery that Conan Doyle might have liked?

I went back and checked all the solder joints I had disturbed and redid them - those associated with the volume pot, the external aerial socket, the earphone socket and the speaker terminals. I must have improved something somewhere as most of the background hiss and hash has now gone - still very, very faint though even with the volume turned up to maximum!

Using my Velleman K7000 in signal tracer mode I am getting audio signals at the junction of C13/R15 and indeed on both sides of the earphone socket (RED and ORANGE wires) and to the speaker terminals, so I'm tending to rule out the earphone socket being the problem?. The signal strength, however, is much reduced from what I am hearing when placing my signal tracer probe on the slider of the volume pot where it is very loud!

So I traced the GREEN wire from the slider back to the circuit board where it connects with C3 and C5. No loss in signal strength on both sides of these capacitors - still very loud! When I traced the signal route across R2 I was getting intermittent responses and sometimes no response. I looked back at my notes and saw that R2 was measured at 11.33kOhms - it should be 10k so not that far out. Flipped the board over and noticed that I was getting a crackling response when touching R2 with a finger or screwdriver blade. Give it a bit of a wiggle and basically it came apart!!!



I know that David advised to use my " ... tracer probe to progressively check back towards the slider of the volume control" but I can revert to that method to identify where the signal strength drops once I fix this R2 problem. So first thing first, let's replace R2 with a new 10k resistor and see if that helps - it may not be the answer but at least we will have ruled out one weak link? As Mike (Boulevardier) said above " ... I always think that if it recently worked OK, but after doing something it stopped working, then it's most likely the result of something that you did in between those two times, rather than a brand new fault" - I tend to agree but let's see what transpires?

[poppydog]

That's really weird because when I finished repairing my 303 it decided to make loud crackling noises when the volume was turned up, It turned out it was a resistor like yours that was coming apart. I am going to have to dig my 303 out now to see which resistor it was that I replaced..

Regards Poppydog