Sony TA-1010 motorboating

This is my first post here and as Dutch is my primary language I want to apologize in advance for my Dunglish 😉

I'm busy with a Sony TA-1010 that had several problems and by now is driving me mad. When I got it there was a loud buzz so I didn’t test it thoroughly as most of the time the basic things like replacing bad caps and cleaning already solve a lot of the problems.

Some of the things I have done already:
  • Cleaned switches and pots
  • Replaced the volume pot as the old one was too far gone.
  • Replaced all electrolytic caps with Nichicon FG caps except the main power supply cap C135 (1000µF/63v) as this measured fine. Most other caps were on their way out with either a high dissipation factor or a high ESR.
  • Replaced C120/C220 .02µF tantalum caps for 0.22µF Nichicon FG caps
  • Replaced the trimpots with Piher PT15 ones as the old ones were fluctuating a lot. The pots are set at approximately the same settings as the old ones (measured them before adjusting them).
  • Replaced R152 (6800 ohm's) as it was cracked and the resistance varied

The problem I have is that the left channel is motorboating. It seems to be coming from something on the amplifier board. When I take away the left input and short it with a 1K ohm resistor the problem still persists. Probing around it looked like C125 was leaking as it measured around 600 ohms in circuit . I didn't have 370pF laying around so I replaced it with a 330pF but this didn’t solve the problem. The 600 ohms measurement is gone... As the right channel plays as it’s supposed to do I don't expect it to be a power supply problem.

Does anyone have suggestions where to look?
 
This is a part of the power amplifier schematic.
Power AMP part schematic.jpg
 
Some suggestions:
https://en.wikipedia.org/wiki/Motorboating_(electronics)
https://www.tdpri.com/threads/what-causes-motorboating.210392 (#7)
https://electronics.stackexchange.com/questions/612704/why-is-my-amplifier-motorboating

Considering this circuit, two dc feedback loops are involved.
In the front stage through R132 to C123. Replace these 100µ/3.15V caps
Near the Q108 current source, C127 as this one can have to much reversed polarity voltage on it due to the broken R152 6800Ω resistor.
Replace these 33µ/6.3V caps too!
C126 is a bootstrap cap - 6.3V is somewhat low imo, better have supply voltage hard. Modern caps are way smaller.
 
Thank you for your reply, I didn't have much time to work on the amplifier earlier but I today I have replaced C123/C126 and C127 unforunatly the problem still exists. The caps that came out of the amplifier were already replaced en measure fine (as expected). The new caps have in most of the cases already a higher voltage rating then the original caps.

Is there a way to determine the source of the motorboating by measuring it in circuit?
 
Not the usual suspects... an unusual suspect: C121, stabilising the first stage rail; it's an injection point into the feedbackloop.
C122, only 10nF is at the base of the input bjt at a rather high impedance and is within a dc-stability loop.
Then bad solderings (reflow if in doubt).
Given the nature of the problem, it's unstable at a slow pace. Something is charging and discharged.
If you have a scope try to find at the bjt's legs where the distortion shows a more spiky respons.
Lift one leg of C120, see what happens.