Sansui 5050 tuner-amp fault

[petervk2mlg]

Hi all
I was looking at my Sansui tuner-amp as it had a rustle or crackle in the left hand channel. Thinking this might be caused by a noisy transistor I gave the transistors on the driver board a squirt with freezer spray. This just caused the crackle to get much, much worse - louder and more dramatic. Then I noticed smoke pouring out of the left hand speaker. The voice coil must have been fried.
The protection circuitry doesn't seem to have helped. What might cause this to have occurred. I realise there are many possible causes. The power transistors seem OK.

Peter

[petervk2mlg]

A scan of part of the schematic

[Herald1360]

Oops! Sounds like hypothermia in the driver stage may have turned hard on one of the output devices. DC output coupling then fried the speaker.

The circuit clip isn't a lot of help- it only shows the output stages and part of the PSU.

It's a bit late now to suggest a dummy load and 'scope approach- the horse is somewhere in the next county!

Disconnect the speakers and try comparing voltages in the driver and output stages of the good and bad channels (if powering up doesn't let out more of the magic smoke). This may help, but I suspect that the dc feedback blues may be playing in the background.

More of the circuit would be useful.

Edit: For reference- the service manual is readily available online as a free pdf, though the version I found was clear enough circuit wise, specimen voltages were illegible.

The amplifiers look like classic long tailed pair-voltage amplifier-darlington output configuration though the circuit diagram layout isn't exactly helpful. An hour spent redrawing in a logical fashion would be time well spent!

[petervk2mlg]

Many thanks Chris

[paulsherwin]

That is a very handsome receiver.

From what I can make out there isn't any protection circuitry to speak of, just a speaker isolation relay to suppress the switchon thump. If there is a DC fault in the power amp then it will put a lot of nasty current through the woofer. Usually this would just blow the cone, but if you kept the power on for any time then I can imagine magic smoke would emerge.

Don't feel too bad about using the freezer spray - I would probably have done something similar given the symptoms. However, this experience is a reminder to pursue power amp faults using a cheap pair of headphones rather than valuable speakers.

I think you probably have a bad driver transistor somewhere. As has been said, comparing the RH and LH voltages should be informative. If you're lucky a new transistor will fix things, but it's possible there have been cascading failures.

[Radio Wrangler]

It seems to have been failing anyway. you likely only accelerated it.

The amp has an anti-thump relay to disconnect the speakers at turn-on/off. It looks like it senses the voltage to the speakers. If so, it didn't work. Maybe it only senses on turn-on?

David

[tri-comp]

Originally Posted by Radio Wrangler

Maybe it only senses on turn-on

No, it's an active circuit that senses both negative and positive DC leaving for the speaker-relay. Any such will cause the relay to disconnect the speakers.
It's a fairly simple circuit that basically robs the relay switching transistor of base-current when DC is detected causing the relay to fail connecting speakers.

Could the smoke come from not the bass speaker but possibly the midrange or tweeter driver?
In case of severe self-oscillation at high frequency the powerspectrum shifts upwards and far too much energy is delivered into these drivers.
As always: Re-cap, re-cap AND re-cap ! That includes the speaker protection circuit and the stabilized power supply circuitry.
Other than that I've had problems with the input differential amp. transistors (Q1/Q3 and Q2/Q4, 2SA726). They could be causing the instability and crackling noise.

[Radio Wrangler]

The only approach is to fire it up into a safe dummy load and check for DC and high frequency oscillation.

An autopsy of the speaker should back-up your findings.

I'm not really a recap everything nut, but the same part values get used all over through a piece of equipment and most of them have done the same hours and the same temperatures. There may be a difference in ripple current in some parts, but once you start to see a few with elevated ESR, then it's time to take a shotgun approach.

There is a huge number of electrolytic capacitors in a Sansui 5050, but if a significant number are running dry, then that's the way it is and unless you get pre-emptive, you wind up doing small fixes far too frequently.

Anyway, theorising only goes so far. Some measurements will open the door to reality and the best approach can be planned.

David

[petervk2mlg]

I have the luxury of having another of these amps, so I removed the driver board from it and installed it in the suspect amp. I thought it had fixed the problem, but after a while the crackle in the left channel returned.
I replaced all the electrolytics (apart from the non polarised ones as I don't have any in stock) on the original driver board and reinstalled it. After the amp had been playing for 15 minutes or so, the bad crackling returned.
Is it possible for the output transistors to have an intermittent fault?
One thing: if I turn the balance control to only have one channel operating, the crackle is present in the faulty channel independent of the position of the balance control.

P

[Biggles]

Could it be the output transistor's insulating washers to the heatsink breaking down, if it has them? I am not familiar with this particular amp but I have had similar problem with whiskers of heatsink metal poking through. I did have a crackling problem with my Trio tuner amp which turned out to be a small signal plastic transistor breaking down. Took me quite a while to track that one down. It doesn't seem to be a very common problem though. I haven't had many like that. The general rule seems to be that the chances of a device being faulty is inversely proportional to the number of legs it has!
Alan.

[Radio Wrangler]

Can you repeat the test with the input cap of the driver board lifted at one end, that ought to eliminate the earlier stuff. The balance control suggests it's OK, but when tracing awkward ones, I go into trust-nothing mode.

How about a bad connection on one power rail to that channel power amp? Could do crackles and a big DC offset.

David

[thermionic]

Just a tip that I always do when repairing DC coupled amps, insert a 1000uf capacitor between the + output terminal of both speaker outputs and your speaker leads. This will prevent any DC currents damaging your speakers!


Good luck!

SimonT

[petervk2mlg]

I tracked the crackle to a noisy transistor. With the crackle only in one channel, I thought I might swap transistors one at a time from one channel to the next to see if I could make the crackle move channels. The last transistor I moved did the trick. It's an SA 726.
My local supplier doesn't list this transistor, but I see plenty on offer on ebay.

[ajgriff]

Don't replace with another 2SA726 (pnp) as they are famous for becoming noisy with age (not always dependent on use). 2SA992 is a good replacement and doesn't suffer from the same problem. Note, however, that the lead configuration is different (reversed if I remember correctly). Check data sheets if in doubt.

It's worth replacing any other 726s at the same time on both channels to avoid problems in the future. Also look out for 2SC1313 transistors (npn) which can suffer from the same issue. 2SC1845 is a good alternative but check lead configuration as before.

Fairly sure that BC327 and BC337 would work too and there may be others.

Hope this helps.

Alan

[petervk2mlg]

Many thanks for the advice on replacement types Alan

P

[ajgriff]

Might also be useful to be aware that 2SA, KSA, SA and A prefixes seem to have been used interchangeably in the past. Same applies to 2SC, KSC, SC and C. You may aleady know this but it caused me some confusion when repairing a Sansui AU 2900 a while back. No idea if there's any logic behind this.

Alan

[Herald1360]

Well, less ink is cheaper......

[petervk2mlg]

Originally Posted by ajgriff

Might also be useful to be aware that 2SA, KSA, SA and A prefixes seem to have been used interchangeably in the past. Same applies to 2SC, KSC, SC and C. You may aleady know this but it caused me some confusion when repairing a Sansui AU 2900 a while back. No idea if there's any logic behind this.

Alan

Thanks, Alan. I didn't know that. It makes sense. I'll post here again when I get the replacement transistor and the amp is up and running properly again.

Many thanks too, to all those who were kind enough to offer suggestions.

Peter

[petervk2mlg]

I’ve now replaced the noisy 2SA726 with a 2SA992 and the amp is playing fine.
Peter

[Michael Maurice]

Originally Posted by ajgriff

Might also be useful to be aware that 2SA, KSA, SA and A prefixes seem to have been used interchangeably in the past. Same applies to 2SC, KSC, SC and C.

Alan

Not necessarily, I've been servicing a couple of Dynatron systems which use a C1115 and other C.... transistors. They are a mixture of NPN & PNP devices and bear no resemblance to a 2SC1115 or whatever