Philips 15RB587 - Humming, Buzzing and Harsh Audio

[drivefactor]

Ok I need to give up.

I have been mending a Philips 15RB587 which based somewhat on the European Phillips B4X26A. Two major problems that I am unable to put a finger on.

1. HUM. This one simply refuses to go away. I know it is on the AF stage. It varies with the volume, peaking at mid clock and reducing substantially when the pot is fully opened. And yes it is also accompanied with a buzzing which is accentuated when the tone control is opened.

2. Distorted audio. It’s simply fatiguing. Voices sound raspy and hoary. Again, the harshness is worsened when the tone pot is opened up.

I have put in new smoothing capacitors, coupling capacitors, cathode bypass and all those tiny little caps around the tone and volume pot. All the tubes (except the rectifier) have been replaces with NOS items.

I have ruled out all the usual suspects. Moved the set to an electrically quite place and grounded with 8’ earth stake. No round loops here.

What am I missing here?

[Steve_P]

Try a replacement speaker and try also feeding in a signal at the Gram socket. This will prove whether the fault is in the speaker, the audio stages or the RF/IF stages.

Are you sure the parts you have fitted are of the correct value and in the right places? It's all too easy to make a mistake!

I'd also look at C40 and the two small caps which are above the EL84 on your circuit. Also, is your rectifier up to scratch?

Cheers,

Steve P.

[drivefactor]

Thanks Steve

1. Loudspeaker is ok. I have tried with 3 different speakers with the same problem
2. I have tried the gram with the output from a portable CD player. Hum is there.
3. Yes the values/location of the replaced components is correct.
4. C38 and C37 (above the EL84) have been changed.
5. C40 is one of the few that I did not change in AF section. Tomorrow I will replace C40 and the rectifier. Will report back.

Thanks again !

[Steve_P]

Is the distortion there when you play a CD through it?

Cheers,

Steve P.

[drivefactor]

Yes. But not as bad though.
Thanks

[Steve_P]

OK. Take voltages on the valves and report them here.

Cheers,

Steve P.

[Sideband]

It's got an odd looking output transformer! It looks as if one winding is used as a smoothing choke. Could there be a problem here I wonder? If the choke winding has developed shorted turns, the hum level will increase. Just a thought....



SB

[Billy T]

The sound output (speaker) transformer is configured as a form of hum-bucking, and if the bucking voltage is reversed it will act as a hum booster. If that section has shorted turns as suggested by Sideband, the sound output will be quite low and possibly distorted, but the hum will still be there. In fact, shorted turns on any winding of the fransformer will have the same effect as the inductance (and impedance) of the transformer will be reduced.

However, it may have been wired wrong from the start, or during subsequent fault finding or replacement of the output transformer in ages past, so I'd start by checking the wiring from the junction of R7/C13 to the junction of S22/S23/C40, and R7 itself, and if that is all correct I'd try change reversing the connections of S23 from the actual output transformer. The latter will either make it much worse, or much better! The feedback from S24 is likely responsible for the strange action of the volume control.

Since your tests suggest that the source is confined to the sound output stage, and by logical inference, the power supply, this is your best place to start. Hopefully it will also be the last place you need to check, but if it looks OK I'd still suspect a wiring errror or accidental ground because there is very little else left at that point other than a faulty transformer.

If necessary you can replace the transformer temporarily with any handy output transformer, using the primary as the S22 winding, and if the harshness goes away, that will also point the finger at shorted turns in the output transformer. The hum will probably still be there to a certain extent since the bucking is removed, but part of the present problem is the extensive feedback circuit which would be disabled if the original transformer is out of circuit.

Cheers

Billy

[Herald1360]

Can you temporarily feed the valve heaters (except the rectifier) from a 6V dc supply?

Maybe also try temporarily replacing the rectifier by a couple of 1N4007 with 100R series resistors and 10n parallel capacitors?

(H-K breakdown possibilities)

[drivefactor]

Originally Posted by Steve_P

..
Are you sure the parts you have fitted are of the correct value ..

Must admit two mistakes when I retrofitted components.

1. C31 (THAT capacitor) was originally 0.01uf but mistakenly replaced by a 0.1uf.
2. C39 (EL84 cathode bypass) originally .22uf but replaced with a 22uf/16v electrolyte simply because that was available.

Both items now changed to original spec and the hum is subdued. Its still there but the meat is gone. I guess both the capacitors were chosen by the designers to have a higher -3dB. So the issue remains but it is not overwhelming and I can possibly live with this level of hum. It still varies with the volume.

However, those annoying background noises, buzzing (that typical nervous, unmoored kind of noise) remain even when the gram input is shorted to ground.
And so does the harsh audio. My other valve radio has an ECL82 output stage and the audio is lush and rich. Compared to that this radio sounds like a cheap, uncared vintage PA amp.

The tone and vol pot track ok (checked in a analogue meter) and the housing is making proper ground contact. I am looking for a replacement volume pot. Hopefully that should address the noise.

[drivefactor]

Originally Posted by Billy T

…The sound output (speaker) transformer is configured as a form of hum-bucking,..

Yes this is a typical Philips OPT.

The wiring appears correct. There hasn’t been any major servicing on the set. Even the original smoothing cap (dating from 1962) is still in place.

Originally Posted by Billy T

….The feedback from S24 is likely responsible for the strange action of the volume control….

The bass boost (from S24) when activated reduces the background noise.

I have a NOS Philips OPT and compared the DC resistance of the windings. Everything is in order.

I will try the steps as suggested by you and report back the finding.

[drivefactor]

Sideband
Thanks. I have checked as above. Its ok.

Herald1360

I have inserted a new EZ80 so that should rule out H –K issues. The new rectifier did bump up the voltages a few notch though.


Thanks everybody for the detailed responses. My problems seem partially mitigated.

[drivefactor]

Just to set a benchmark these are the voltage readings with the variac set to 230V AC.

The values in brackets are the actual readings

B4 EL84
Pin 7 = 240v (233v)
Pin 9 = 210v (205v)
Pin 3 = 7v (6.1v)

B3 EBC81
Pin 1 = 80v (72v)

B2 EF89
Pin 7 = 150v (137v)
Pin 2 = 70v (82v)

B1 ECH81
Pin 1 = 70v (78v)
Pin 6 = 210v (205v)
Pin 8 = 125v (98v)

B6 EM84 Magic Eye
Pin 6 = 210v (205v)
Pin 7 = 58v? (55v)

C12 1st Smoothing cap 250v (246v)
C13 2nd Smoothing cap 210v (208v)

Prima-facie the readings look in order. All valves new with less than 50 hours except the EL84 and EZ80 which are fresh out of box and the EM84 which is original but very weak.

[Billy T]

Originally Posted by drivefactor

Must admit two mistakes when I retrofitted components.

1. C31 (THAT capacitor) was originally 0.01uf but mistakenly replaced by a 0.1uf.
2. C39 (EL84 cathode bypass) originally .22uf but replaced with a 22uf/16v electrolyte simply because that was available.

C31 is relatively insignificant, but C39 will have increased LF gain so that is one clear answer, however measuring the DC resistance of the Opt transformer windings is not really that useful as it is AC impedance that matters, and one, two or even several shorted turns will not show up at all. I suggest that you patch in the new transformer as that is the only reliable test available unless you happen to have an impedance meter or the means to set up a meaningful equivalent testing method.

Should you wish to try a dynamic shorted turns test in situ, one way would be to disconnect all the primary and secondary windings put a suitable load resistor across the primary (high voltage) side (say 10K or 22k 5 watt) and apply 6-10 volts ac from a filament transformer or similar into the speaker winding. Meter the voltage across the primary load resistor and compare the results for both transformers. If the incumbent has shorted turns the output voltage will be significantly lower. You will only need to make a momentary test, just long enogh to get a stable reading and this won't hurt a good transformer. The unused windings won't affect the test result.

If you don't have a suitable low voltage ac source, another 10 or 15k 5w resistor in series with the winding and supplied from the mains will do but in that case it is preferable to use an isolation tranformer to supply the mains input voltage. However, it is perfectly safe to make the test with direct mains connection if you connect it all up with the supply off and unplugged, then switch it on at the wall or plug into the outlet for the test, read the meter then switch off again and unplug before disconnecting the leads etc.

If you get significantly more output from the new transformer then the old one has shorted turns. It may also buzz if faulty, in fact if you substitute a resistor for the speaker and wind up the bass and volume, the transformer may talk to you if it has shorted turns. This depends on how tight the windings and laminations are, so it is not that definitive, but is an interesting effect.

Cheers

Billy

[drivefactor]

Originally Posted by Billy T

C31 is relatively insignificant, but C39 will have increased LF gain so that is one clear answer...

Thanks. But I guess that the designer would have taken care to ensure hum is not generated rather than blocking it in a fashion in the last stage

Thanks also for that excellent information on setting up and testing the OPT's. It is not something that I need to do everyday, but do know about a variation of 6 volts AC trick to determine winding ratio.

I think its much easier to simply drop-in the NOS OPT. Will do so at the soonest and report back.

This radio has proven to be more than what I expected. One of the IFT had an open winding. I tried re-winding on my own but then it wouldn't fit back into the housing. It needed the expertise of an old man who used to do such things in the past. Nowadays he makes a living by rewinding the CRT deflection yokes. The repaired IFT worked ok in a manner, but luckily a replacement was located and installed. The OM helped with alignment. In between the IFT from a Philips Solid State receiver was tried. It sort of worked but that is another story.

Thanks again !

[drivefactor]

I tried the replacement OPT (same Philips part number) this evening. No improvement in hum situation and the audio quality is actually worse off.

Would changing all the resistors make any worthwhile difference? Till now I have only changed the ones with a runaway value. The original look so beautiful; it will be a shame to put in modern MFR's.

[Billy T]

Originally Posted by drivefactor

I tried the replacement OPT (same Philips part number) this evening. No improvement in hum situation and the audio quality is actually worse off. Would changing all the resistors make any worthwhile difference? Till now I have only changed the ones with a runaway value. The original look so beautiful; it will be a shame to put in modern MFR's.

It is not that sort of problem. From your descriptions, which are very detailed and clear, the problem seems to be centered in the power supply and audio output stages, so there is little point in attacking elsewhere using a shotgun approach.

The sound output circuit is complex (typical Philips ) and there are multiple feedback paths for hum and distortion factors.

IIRC, you said there was hum with the volume control at minimum and that the hum was least apparent with the volume control centered. That suggests that the hum is being injected via the C31/C38 taps and in my opinion there will probably be a primary cause that once rectified will eliminate most of the problems at one fell swoop.

To get to the heart of the matter, try disconnecting the feedback to SK7/C38 from S24 and see what effect that has. I suspect that it will at least eliminate or minimise the center hum issue. If the hum is still present, disconnect one end of C31 (AF feed to the volume contol tap) and this will eliminate the full signal path from the front end of the receiver right up to the detector. That should definitely leave you with a humless and silent set. If not, then you almost certainly have excessive HT ripple.

Logical analysis is usually the best approach, though there would not be a tech alive who had not seen logical analysis fail miserably, the impossible realised as reality, and the ultimate cause of the fault to be the one component or circuit that was considered completely blameless! Many years ago I spent hours on an auto-tuning car radio with an unstable autotune circuit. I had changed everything and it still wouldn't work. My then boss (and mentor) came along, put the original transistor back in and away it went. I'd fixed one fault, but introduced another, and from that experience I learned the kiss principle: "Keep it simple stupid!"

In this case you appear to have eliminated the transformer as a potential cause, but I still have strong reservations about the S23 hum bucking feature, and that makes C12 & C13 suspect, as not only are they the principal HT filters, any hum there can and will find its way into the volume control via S23/S24. I see you have replaced them, but were they the right values? As a diagnostic aid, you could also try temporarily bridging C13 with another cap just to see if it alters the situation. Adding more capacity on C12 is not recommended as it might give the rectfier conniptions. The humbucking circuit is still in my sights as well!

Of course having stuck my neck out by making all these prognostications, the true cause will probably be found to be something absolutely outside of the frame that you stumble over quite by accident, but that's vintage radio for you! (On that point, make sure that there are no dodgy screwed or clamped connections to ground in the power supply, or anywhere else for that matter.)

Keep the faith.

Cheers

Billy

[drivefactor]

Originally Posted by Billy T

It is not that sort of problem.

Thanks for such a detailed explanation. I appreciate the effort and time you (and others) have spent in diagnosing this in a call centre manner

I will try the steps suggested by you, hopefully over this weekend, and report the results. Please standby till then.

Meanwhile:

- C12 & C13 (main smoothing caps) are new 47uf/350V high quality electrolytes. They are wired below the original dual section can. I don’t expect them to be bad, but will check nevertheless.
- A new problem has emerged. This morning, upon power-on, the radio came alive for a couple of minutes, made a few low intensity crackling, spikey kind of noises and then went silent. It is not completely dead as I could hear some very low back ground noises. I am suspecting either the EL84 or its cathode resistor (both new) is the culprit here. I did not have time for a detailed inspection before I left for work.

Originally Posted by Billy T

Logical analysis is usually the best approach, though there would not be a tech alive who had not seen logical analysis fail miserably, the impossible realised as reality,

Sir, it is quite an assurance to see that you bring a rather philosophical approach to this From where I come from and what I do for a living; your statement summarises the general situation brilliantly!

[drivefactor]

Originally Posted by Billy T

... disconnect one end of C31 (AF feed to the volume contol tap) and this will eliminate the full signal path from the front end of the receiver right up to the detector. That should definitely leave you with a humless and silent set. If not, then you almost certainly have excessive HT ripple.

I did that and there was no residual hum. I have also changed almost all the capacitors (exepct some very small value PF's) and the hum is considhe erable less. I guess more due to loweing EL84 cathode bypass to its original value.

I am coming around the view that the existing amount of hum is acceptable. But the buzzing problem remains. It varies with the volume. Zero at minimum position, peaks at mid point and reduced almost 90% at max position.

Originally Posted by drivefactor

A new problem has emerged. This morning, upon power-on, the radio came alive for a couple of minutes, made a few low intensity crackling, spikey kind of noises and then went silent. It is not completely dead as I could hear some very low back ground noises.

That was the plate resistor for EBC81 (R15/220K) spluttering and burning out. A replacement solved the matter and we are alive again. Earlier the grid resistor for EF89 suffered similar fate. Strengthens the case for a wholesale replacement of the resistors notwithstanding their sexy appearance.

Billy T:
PM replied. I agree. Lets save the Adminullahs from ulcer

[drivefactor]

Resurrecting this thread to report a progress. Well short of.

To summarise, the radio produced 50Hz Ac hum with its typical buzzing tone. This hum varied with the volume pot and peaked at mid-point.

All known tricks were tried to mitigate this issue. The origin of the buzzing was narrowed down to the AF section. The area around there is a bit messy and defies logical wiring techniques. Multiple feedback and fold back loops travel all over making several return trips for a simple function.

I made an effort to bring some order by changing over to shielded wires for AF signal and routing them away from noise sources. All C’s and R’s in AF section, and some in the RF stages, were changed although to be fair most of them measured within spec.

Still no luck! Although, the buzzing diminished a little but it was still annoying.

I had vaguely suspected that the vol pot with the on/off switch could be the culprit but never got around to investigate it. It was expected that the original design took care to avoid any interfacing of AC hum/buzz. In the end I finally bypassed the on/off switch and directly wired the mains lead in to the power transformer and voila the offending hum/buzz was gone. It is completely silent now. With the disappearance of buzz, the tonal quality of the audio has blossomed and it finally sounds the way it should.

At this stage this should have been a happy ending, but I know this is NOT the way and I am keen to re-employ the integrated on/off switch.

Why the pot is picking up AC buzz? Do these have an internal shield which is somehow damaged in this case? The body of the pot is grounded and I have tightly twisted the AC wires. The switching section is of push pull design as against the more common rotary type.

Thanks.