Optonica SM-3636 protection mode

[Wray & Ting]

This is a happier thread than I thought I was going to be starting when I woke up this morning. My Optonica SM-3636 that I've owned for about 3 years went into protection mode while idle last week. For several months leading up to that, it had been intermittently emitting very loud bursts of noise when powered on that had me worried for my speakers, but it was so random and infrequent I just ignored it for the most part. The instant before it died last week it gave one of those loud bursts, so they must be related.

I had the amp opened up today and was measuring some voltages on the right channel board while it was in protection, when I heard the relay switch, but the light stayed red. I continued taking measurements (TP402 and TP404, both zero VDC compared to TP401 and TP403 which were both -27 VDC), and the relay switched again, this time lighting the happy green light! I powered the unit off and on again, and the LED went green after the usual 4 seconds just like it should. Am I probably correct in assuming there is a loose connection or cold solder joint on the right channel board that may be the culprit behind both the noise and the recent protection mode? It seems my multimeter probing might have flexed the board enough to reestablish a connection.

If I can't find any suspect solder, should I just resolder anything I see that makes me look twice? Aside from a needed recap (and the electrolytics are all original, even the power caps) is there anything else I should do as preventive maintenance going forward? As a final note, there is some evidence of previous repair inside and the unit differs a bit from the schematic available on HFE. Mine has 3x 120 V plugins on the back instead of a fuse, there is no voltage selector switch on the back, only one fuse per monoblock, and no other fuses inside, and the high voltage capacitor on the power switch is a different type than what I see online.

 

[cademan]

It is hard to diagnose intermittent problems but it almost sounds like a pre driver transistor could be in the first stages of needing replacement. When they do finally go bad or get noisy, you will experience popping and crackling noises (usually in one channel) untill it goes into protection due to high DC spikes.

 

[Wray & Ting]

Thanks for replying! I just turned it back on after several hours unplugged and the malfunction seems to be less intermittent than before. The unit powered on fairly normally (maybe a 10 second delay compared to the usual 4 seconds) and then proceeded to go in and out of protection mode over the course of 15 minutes. To help me figure out what was going on, I hooked up my headphones, and sure enough, when fully powered there was a soft, random static in the left channel only, that would periodically get louder and then result in quite angry pops which would put the amp into protection again. It's basically doing exactly what you described above. This particular amp model uses two big unobtainium ICs in the power stage, so as long as those are OK the rest of the transistors look generic and hopefully are easily replaceable. I'm not sure which ones are pre-driver by looking at the schematic, but there are only 4 of them on the left channel board (2x npn, 1x pnp, 1x double pnp).

Any advice on determining which one is faulty from either the schematic or measurements?

 

[cademan]

Need to see the schematic.

 

[Wray & Ting]

pre driver transistor

I didn't realize the parts list at the end of the service manual listed Q403 as "pre driver / constant current" and Q405 as "pre driver", so I guess it's down to only two possibilities. These two are a PNP / NPN complementary pair (2SA818 and 2SC1628) and cheap, so I could just replace both I suppose.

 

[cademan]

Q401 can also be suspect.

 

[Wray & Ting]

To make a long story short, I haven't made any tangible progress on this issue in the past six months. However, last week I plugged my SM-3636 in for the first time in months, and it powered up immediately with no issues. I put it back in rotation as it was before the protection issue, and it's been running perfectly for a week with no unwanted noise in either channel or intermittent protection mode in that time. I know diagnosing intermittent issues is tricky at best, but is it at all likely that an amp with the suspected failing transistors mentioned above would return (even temporarily) to full working condition with no repairs effected? The last time I turned it on before last week was maybe four months ago, and then it was still exhibiting random, staticky noise in both channels and sporadically going in and out of protection mode. It still needs attention in the long run as the problem could seemingly recur at any time.

PS - I have really come to appreciate the quality of this amp after using a Pioneer VSX-511S in replacement these last six months. I have happily banished the Pioneer to the workshop once again, where I've been without amplification for the same stretch.

 

[ivandezande]

I didn't realize the parts list at the end of the service manual listed Q403 as "pre driver / constant current" and Q405 as "pre driver", so I guess it's down to only two possibilities. These two are a PNP / NPN complementary pair (2SA818 and 2SC1628) and cheap, so I could just replace both I suppose.

Replace them all. Will cause trouble.
Optonica also loved to use fusable resistors that go bad. Ever receiver I’ve worked on from them is in protection or not working because of this.
The dual transistor in the input stage will also be problematic.
They look a sickly hue and are off the board a bit. Also noted in the schematic.

 

[Wray & Ting]

Well. It's been almost a year and a half since I started this thread and today I finally got around to servicing this amp. I recently ordered a replacement motor capacitor for my Planar 3 (nice ~6dB reduction in 120 Hz motor noise by bringing that back into spec) so I figured I'd order what I needed for my amp based on the feedback from this thread, and the numerous threads out there on replacement options for the original 2SA818 and 2SC1628 complementary pair of transistors. My amp started crackling in the left channel again the other day after another long spell of normal operation, so today I took it to the repair bench.

I decided to go with KSA940 and KSC2073 as the replacements since they were recommended in a couple threads somewhere. I also ordered four KSA992 with the idea of building the dual transistors Q401/402 from them. I have only replaced Q403/404/405/406 so far, and have reassembled the amp and am listening to it now. Since the pinouts on KSA940 and KSC2073 are BCE and the original 2SA818 and 2SC1628 were EBC, I clad the leads in PVC and carefully bent base and collector pins into each other's former positions.

The good news - the amp is no worse off for my efforts and so far hasn't made any crackling noise or gone into protection. The bad news - since this problem is intermittent, I have no way of knowing if it's actually improved except to listen to it a lot. So that's what I'm going to do. It'll be months before I have much confidence that I've replaced the faulty parts since it could go months of sporadic use without a problem. If it acts up again, I'll be looking to do Q401/402. Maybe I'll even give this amp a signal path recap sometime!

Many thanks, cademan and ivandezande.

 

[Wray & Ting]

Maybe this will be the last time I necro this thread, haha.

It's been a year and a half since I replaced Q403/404/405/406 in my last post. For whatever reason, that repair gave me 6 months of problem free operation until once again falling into the old pattern of crackling channels and protection mode. It's been on the shelf for a year - I had forgotten until I reread my last post that I already ordered four KSA992 to replace the two 5-legged 2SA798 (Q401/402). Well, I got back on it yesterday. Amp works again! Here are all the replaced parts in the left channel (Q401, Q403, Q404):



And the replaced parts in the right channel (Q402, Q405, Q406):



No widening of the emitter hole was necessary - I tacked the two leads together with almost no solder and they fit with a little wiggling.

Thanks again, AK. Time will tell if these two repairs will get me some more good years out of this amp. I was just "gifted" a 1991 Yamaha RX-V850 that doesn't get past the protection mode checks. Who knows, in 5 or 10 years maybe I'll get that one up and running too