SMPS repair - primary side controller not starting up/driving gate

[MateV]

Hello!

I've been trying to repair a blown SMPS, and got stuck.


The original fault seems to be Q9101(TK13A65U) going short, burning out R9101 in the process. C9104(OCP filtering) and R9108(gate resistor) were also bad, along with U9102(feedback optocoupler) having very low emitter-collector resistance.
Measured and replaced everything that was bad in this area, along with U9101.

Powering on the PSU results in the following on the Vcc pin of U9191:


There never seems to be any gate drive output, except a few picoseconds ~2V pulse, which I assume is just some sort of noise(no voltage detectable on any output on the transformer either).
Removing power from the PSU causes the following oscillation on Vcc until Vbus drops under ~20V:


U9101 looks to be this with a custom part number:
https://www.semicon.sanken-ele.co.jp/sk_content/ssc1s311a_ds_en.pdf

Full schematics:
https://elektrotanya.com/715g7831-p0e-000-0020_philips_psu.pdf/download.html

Any help or suggestions on what to measure would be greatly appreciated!
Thanks.

[Gyro]

Welcome to the forum, MateV.

It's a bit of a long shot, but maybe C9103 has dried out, particularly if it's near hot parts. These small switcher VCC reservoir caps are notorious for causing startup problems in SMPSs. It's worth a check anyway.

[wraper]

Check ESR/replace C9103, C9128. As optocoupler was bad, I'd expect ZD9107 and probably ZD9101 and C9102 being bad too.

[MateV]

Woo, it seems to work! (at least for primary side switching, PSU output is still not quite good)

Don't have an ESR meter, so couldn't measure the caps, but their capacity is good. Replaced them anyway, along with all the zeners mentioned.
I suspect it was ZD9107, the board is very very slightly discolored under it.

Thanks to both of you for the help!