Kenwood KA-907 oscillating when preamp is hooked up

I’m repairing this amplifier for a friend, and this one has given me several headaches. When I received it, the input board had multiple blown out traces, as well as multiple dead transistors. I did a full electrolytic recap as well as replaced all of the black flag capacitors with silver mica of equal value. I also replaced several zener diodes. I am now at a place where I want to do a bit of testing before I go further with any work.

On power up (on DBT) I found the bulb went bright as the bulk capacitors charged up, and then went dim and then a second later went slightly bright again. Also ans I was monitoring the bias current it went up quite a bit ans well. After a bunch of trial and error as well as input from others, I found that the amplifier was oscillating. I am seeing a 2 MHz oscillation at the input of the protection relay.

IMG_6549.jpeg


Trying to figure out where it was originating from I found that when I removed the pre out/amp in jumpers on the back of the amp that the oscillating went away and no more brightening of the bulb and bias was stable.

So that told me the oscillating was originating in the preamp, right??? Measuring at the preamp output with it disconnected from the amp input I was seeing nothing. No oscillating whatsoever.

IMG_6573.jpeg


If I turned the volume pot all the way up, I would get oscillating, but only at the very end of turn. And the oscillating that I was seeing was not increasing as I turned the volume pot, it wasn’t there, and then all of a sudden appeared at the end of the turn, plus this oscillating was 24 MHz.

IMG_6574.jpeg


I’m hoping to get some suggestions on where to look, I am confused as to why the amplifier goes into oscillation when it’s input is connected to the preamplifier output.

Really appreciate it!

Dan
 
Does oscillation stop if you reduce the volume control position? If so, this suggests a global feedback path, i.e. amp output somehow couples back into preamp; reducing gain via the volume control is sufficient to reduce the feedback loop gain enough to quench oscillation.

Are you using phono input or do other input sources have similar susceptibility? If you short phono inputs, is the problem resolved?

Post a schematic for improved advice.

Good luck!