Sansui AU101 distortion

[dieterboels]

Hi Sansui lovers,

My right channel of my Sansui AU101 has a problem. Any help is welcome.

When I turn up the volume (not so loud), the right channel only manages to drive the speakers for a few seconds and then jumps (instantaneously) to a distorted sound (right channel only). If I turn the volume down again, the distortion remains and jumps back (also instantaneously) to clean after a few seconds. it jumps faster to distortion with bass heavy tracks.

I tested with an oscilloscope and found the distortion only when speakers are connected and is introduced around component TR806 in the circuit. I tested that transistor and it seems fine (with diode testing).

To be completely honest : this project was my very first recapping project about five years ago and the soldering isn't beautifully done. Just if somebody might think this could be a soldering problem.

I could just switch some transistors between left and right, but maybe if someone would have a clue what this might be to approach this practically.

 

[tom_tm]

Do you have the same problem with headphones attached?

 

[dieterboels]

Well, when the speakers are disconnected no (so when the Loudspeakers switch is OFF), when the speakers are connected yes ...

 

[tom_tm]

It sounds then that it may be a problem with the soldering at the final output stage. If it works like my 222 (without looking at the schematic) then the headphones are connected to the power transistors. Dry joints are very difficult to spot, even with a multimeter. Probably best to prod around while feeding it with a sine wave and the amp under load connected to your scope. If the trace jumps, you know where there may be a problem. Reflow the solder and check it again. Great amp btw. Worth the time repairing it.

 

[tom_tm]

Also, check the switch. These are notorious for bad behaviour. You may have to open it up and clean it properly. But be very careful as there is no easy replacement.

 

[Hyperion]

Dry joints are very difficult to spot, even with a multimeter.

Might be the wrong tool, I use the Mk1 Eyeball, often assisted by a watchmakers eyeglass.

The symptoms the OP describes does sound like a dry joint. However, the fact that there is no distortion until speakers are involved, suggests a problem with the output transistors, or their supporting components.

Keep looking...

 

[sregor]

Outputs: The connection to the collector is through the screws - it could just be one of the screws are loose. Also, can be intermittent output transistor it self. I'll second dry solders as possible problem too. A picture of the scope, with a sign wave input could be informative. Good luck with it.

 

[dieterboels]

Also, check the switch. These are notorious for bad behaviour. You may have to open it up and clean it properly. But be very careful as there is no easy replacement.

Which switch do you mean?

Thank you all for your help already!

Well now I discovered the following: If I shortcut TR806 Collector and Emitter while playing the problem is solved (happened by accident). Would this be a signal of a dry solder? Still, at even higher levels low frequencies are still distorted (only right), but this time the distortion grows gradually. I will get some scope pictures this weekend.

 

[sregor]

Well now I discovered the following: If I shortcut TR806 Collector and Emitter while playing the problem is solved (happened by accident). Would this be a signal of a dry solder? Still, at even higher levels low frequencies are still distorted (only right), but this time the distortion grows gradually. I will get some scope pictures this weekend.

First a correction - outputs are not 2sd315, but still are suspect. Second, is drivers are the infamous CDC8000 and 9000 "flying saucer" type which are common failure items. (including tr806). And, the center lead is the base - shorting base to emitter on an open driver would probably give you the results you described. Second, is that the lead configuration is different from most standard Japanese transistors, so if they aren't original, they could be in wrong. Sounds like the transistor is open. If you put in replacements, make sure base collector and emitter are in the right place. Good luck.

 

[dieterboels]

Thanks for the fast reply, I'll start by replacing the TR806 (CDC8000). Do you have a suggestion to replace this one?

 

[sregor]

Thanks for the fast reply, I'll start by replacing the TR806 (CDC8000). Do you have a suggestion to replace this one?

There are probably lots of possible, one recommendation (with the caveat of getting the leads right):
CDC 8000 → KSC2690
CDC 9000 → KSA1220
2SC871 → KSC1845

 

[dieterboels]

I had still a BC639 around which also was a possible replacement for the CDC8000 and that fixed the biggest problem, namely the broken mode it went in, that's not happening anymore. So the old transistor must've been funky.

At higher levels (which I generally don't listen to) I'm still having very small distortion in the right channel, almost not audible when listening to music, but when sending a sine wave I do. Here's what I get on the scope: (after power amp entrance of power amp is clean sine). And as said when disconnecting the speakers, the sine becomes clean.

Already sounding great again at normal volume haha

 

[sregor]

If problem shows up only with load, output and driver are most likely suspect. Also check resistors in output area. My best guess would be the driver, (tr808?) but could be output or one of the resistors....

 

[skippy124]

One of the problems I found with the CDC transistors is low gain ( measured at typically <30 ), this may also have an effect on distortion when a speaker load is applied.

Cheers

John

 

[dieterboels]

Hi,

I had a BC639 lying around to replace the TR806, which solved the problem of the distortion mode the amp went in, so that's solved. Woohoow

At higher Volumes I'm still getting a little bit of distortion in the right channel. When listening to music it's almost inaudible, but with a sine it definitely audible. Only happens when load (speakers) are connected. Here is what I get with the scope after power amp (clean sine at entrance of power amp).

Thanks already.
Dieter