Marantz 510m: one channel high DC when starting

I'm working on this marantz 510m and in one channel the DC goes really high on startup before stabilizing. It swings positive to around 30V. I'm still running it on the dim bulb tester with a 60W bulb. The other channel doesn't do this. There's no speaker protection, and it pins the meter, so it needs to start up better than this. It takes 20 or so seconds to stabilize. I haven't changed much on this board. The electrolytic caps and the emitter resistors. The outputs have all been changed to onsemis. I've swapped driver boards between channels as they disconnect from the outputs/drivers and the problem follows the driver board.
When it settles it plays fine.

I've never seen something swing this hard when powering up, how can I troubleshoot this?
 
Raccoon1400-- I've got a Marantz 1200 which has very similar power amp boards. IIRC- the 510M does not have a speaker protect relay board.

The audio karma forum has a Marantz forum and lots of info on the 510/510M. They also have several people who "specialize" in the Marantz gear.

Your choice- spend the time reading all the posts, some which end without a fix, or the experts fix the gear via private email. Why do they do that? Maybe so you have to send the amp to them and pay to fix it.

Or, do what I did- spend an inordinate amount of time reading about all the fixes and doing it on your own.

About 5 or six years ago, there was a guy on DIY Audio who was a wizard at fixing solid state gear. I scanned the schematic, wrote down some measurements and he would email back to replace a component. He was right, three times that I asked him for help.

But, he got banned from this here forum for being naughty to someone.

Good luck with fixing your 510M. I really like the sound of these amps when they're working as designed. My one piece of advice I would give you is don't start removing and replacing transistors unless you know they are bad. I did the shotgun repair method and spent months. Learned a lot about old transistors, curve tracers and ended up placing the old transistors back into the boards. Some of the transistors would test good on the 9V battery operated tester, but failed when operating voltages were applied on the curve tracer. Then it took lots of research to find a replacement that was compatible with the old transistors. Otherwise the amp would oscillate like crazy.

If it helps, I also built a Leach Amp that sounds very much like the Marantz 1200. I don't use the 1200 very often, mostly because some of the traces on the PCBs have lifted and been repaired several times. It's like an old car that has been lovingly rebuilt- I don't take it out for a drive very often, leave it in a parking lot.