Graetz Sinfonia 3262 Fluctuating FM sound

[cathoderay57]

Hi all, I'm still using my Graetz Sinfonia 3262 for daily listening on FM. I have posted a couple of other threads on this radio in the past couple of years. Whilst, in the main, this radio is a reliable performer, I have in recent weeks been experiencing an odd fault that seems very gradually to be getting worse. After the set has been running on FM for at least 2 minutes, the deflection on the magic eye drops by more than 50% over a couple of seconds, and the audio sound level drops simultaneously, still audible, but quieter. No distortion, remains perfectly on-tune, no background noise, very very slight increase in hum level. After about 15 to 30 seconds the fault disappears. At first this occurred only once but has started repeating up to 3 times, all within the first 5 minutes of listening. Once the fault clears for the last time it operates normally until switched off. Since the signal level drops at the magic eye, I've mentally ruled out the EL84. Magic eye brightness stays the same, and I replaced the selenium rectifier with a 1N4007 bridge with dropping resistor when I restored the set so unlikely to be an HT supply problem. My initial thoughts are to leave it alone until the fault goes permanent, since tracing intermittent faults can be frustrating. I have a full set of spare valves so I guess if push comes to shove I could try replacing them one by one starting with the EABC80 and working backwards. Any thoughts or recommendations appreciated! Just in time to get my grey cells working during the latest lockdown.... Cheers, Jerry

[Radio Wrangler]

With an intermittent problem like that, a bit of valve swapping is a good start. It's easy and has a reasonable chance of finding the problem.

After that I'd measure a load of voltages in all the usual places, once before it goes quiet and once after. Comparison may then yield some clues.

On that set, does the magic eye read AGC voltage on both AM and FM, or does it switch to DC from the discriminator as a tuning indicator?

David

[cathoderay57]

Hi David, thanks for your reply. I have uploaded a snip covering most of the detector circuit. As far as I can make out, on FM it uses dc from the discriminator as a tuning indicator. The feed to the ME grid comes via R133, a 3M resistor from the negative (non-chassis) end of the 5uF electrolytic wired between the discriminator diode pair, anode of one diode and cathode of the other. I was wondering if a dirty contact on the pre set rheostat R130 (1k) (which I assume is a balancing adjustment on the discriminator) could be causing the fault. Maybe internal heat is affecting a dirty wiper on the skeleton pot, although it does seem unusually repeatable. I ran the set for 30 minutes this evening on LW without the fault appearing so it seems to be unique to FM. Cheers, Jerry

[Radio Wrangler]

That needs to be a bit further to the left. I can see the magic eye and part of the detector, but nothing of where the voltage comes from to said magic eye. I assume it's either got a foster-seeley detector or a ratio detector. Look out for ratio detectors because they need a large floating capacitor... often electrolytic and a known failure point.

Guess that guarantees a Foster-Sealey.

David

[Freya]

a bit more..

[Radio Wrangler]

Yes, the stuff to the left shows the diodes of the EABC80 just going off the edge of the first snippet are the FM discriminator, and the path for the discriminator DC output to the magic eye is now visible as well.

I'd check that electrolytic capacitor C132. The tuning indicator voltage is developed across it. It's one people tend to forget about.

For intermittency you're looking for bad joints, dirty switches, sometimes valves, sometimes resistors.

David

[cathoderay57]

Fault still present on FM this morning, & as usual disappeared after 10 minutes. I haven't done anything yet apart from exercising the wavechange switches. Thanks both for your replies. I will try a new EABC80 today and, if fault still present, clean R130 and change C132. Cheers, Jerry

[ms660]

It could be anything from RF through to the ratio detector, there's no AGC action to worry about for FM because there isn't any so far as I can see.

Lawrence.

[cathoderay57]

Hi folks. This afternoon I stuck in a brand new EABC80 and the fault still occurred. I removed the bottom cover, cleaned R130 pre set and changed C132. All OK for about 30 minutes at which point I switched off since SWMBO wanted to watch TV. Maybe it's fixed. I will try again tomorrow morning and see if it still works OK after an overnight kip. I tested the C132 that I had removed on the Avo ohms range (15v) and it charged up OK giving a final resistance of > 5M. Switched to dc volts and it had held a charge. The actual voltage across the replacement cap in-circuit swung from close to zero for no signal to about 25v with strong signal. Therefore my test doesn't necessarily mean the old cap is OK since I've known electrolytics to appear OK when tested at a lower voltage. On the cheap Chinese component tester it measured 7.5uF (supposed to be 5uF). Cheers, Jerry

[cathoderay57]

All OK this morning. So far so good!

[G6Tanuki]

Originally Posted by cathoderay57

On the cheap Chinese component tester it measured 7.5uF (supposed to be 5uF). Cheers, Jerry

Electrolytics generally _lose_ capacitance as they dry-out-with-age. One that reads 50% higher than its intended value is probably leaky - the cheap component-testers are notorious for 'reading high' when faced with a leaky electrolytic.

[cathoderay57]

A couple of days after replacing C132 the same fault returned. I swapped out the ECC85 and since then everything has been OK. I put the temperamental ECC85 onto the VCM163 and It tested as follows (spec values in brackets):
Triode #1: Ia=3.7mA (10); gm=2.5mA/v (5.9)
Triode #2: Ia=4.5mA (10); gm=3.4mA/v (5.9)
After 3 minutes on soak the values increased to the following:
Triode #1: Ia=4.5mA (10); gm=2.9mA/v (5.9)
Triode #2: Ia=5.4mA (10); gm=3.6mA/v (5.9)
Not sure the tests prove anything other than the fact that the valve is past its best.
Cheers, Jerry

[Joe_Lorenz]

Hello,
just not to forget about: The contacts of the valve holders can produce high resistance. Some contact cleaning spray may be a good idea.
Regards,
Joe

[cathoderay57]

Hi Joe, good suggestion, makes sense because the fault is maybe also heat related - heat plus bad contacts = trouble! Regards, Jerry