You may have read my post in another thread about having a Sansui BA60 on the bench. Well in it's stock state one channel's THD measurement was within spec of the manual, <0.3% at 20 watts. While the other channel was higher at just over 1.0%. None of the original caps were bad but this amp was made in 1970 and anything that old is now suspect.
I replaced all the electrolytic caps with new Nichicons and measured the voltages and they all checked out against the schematic. However the distortion issue is still there. Any thoughts on what could be the cause? I left the input coupling caps as they are they the green dipped film type. My gut feeling is a carbon comp resistor but how to track it down? Perhaps take distortion readings at the various stages? Here are schematics for those interested.
I replaced all the electrolytic caps with new Nichicons and measured the voltages and they all checked out against the schematic. However the distortion issue is still there. Any thoughts on what could be the cause? I left the input coupling caps as they are they the green dipped film type. My gut feeling is a carbon comp resistor but how to track it down? Perhaps take distortion readings at the various stages? Here are schematics for those interested.
Slow progress on this as it's been busy around here. I tested the components rayma suggested and the cap was fine and the resistor had drifted a bit higher to 360 ohms. The boards were still coated with lots of ancient flux so I cleaned all that off and it fixed the bit of oscillation but THD is still reading more than double the good side. Looking at the output from my analyzer just shows higher third harmonics. Wonder if it was maybe always like this as you could probably never tell by listening?
From the speaker outputs. Good point about bypassing the phase switches but the results didn't change. Most of the resistors can't be measured in circuit so one leg will have to be lifted. It's such a straight forward dual mono design. They used the same board layout for each side instead of being fancy and mirroring one.
Success! I have found the offending component causing the problems. The instability never showed up again after giving the boards a thorough cleaning. On a whim I decided to change R806 and didn't notice any difference. Finally I started poking around with the distortion analyzer and found one side of R808 measured much higher than the other. Removed it from the board and found it had drifted to 5.2k. Replaced it and distortion is much better now. In fact it was better than the "good" side. I replaced the same two resistors on the "good" side and now both channels measure much better. 0.015% at 20 watts.