The device I was trying to fix is a SONY CRT TV model KV-32FV16. The power supply was dead; specifically the MCR5102 controller IC.
Here are some pictures: schematic, psboard (removed IC), and the IC on heat sink.



My first attempt was to use this what assume is a fake HP psu:



The FET was dead. I replaced it. The new FET and its snubber resistor blew up. Maybe it's still fixable, but left it for later.
So I grabbed this instead:


And... it was kinda working? The supply reached the desired voltage but in pulses. Every second a pulse of the expected 135V came out of the supply. Nice that these IC's are hard to kill. Didn't get stable power, don't know exactly why; I wish I had an oscilloscope.
I don't remember if I tried increasing the output load; maybe I did, maybe didn't. Dumb me if I forgot about that. Because of that now I want to repeat the experiment.
I tried many arrangements of demagnitizing circuits for the transformer, but adding nothing at all is what worked best.
I tried reducing the value of the IC's current sensing resistor, little by little expecting some change. This happened:


Maybe that psu still works if I replaced the FET. But left that aside as well.
So then I grabbed this:

It has a strange IC: MIP2E4DMY. Repeated all the same steps above, got same results as before: power in pulses, and idiot me doesn't remember if tried having a proper load on the output.
Nothing blew up though.
Then I grabbed an old CRT TV model KV-21R22/5. It has an interesting psu:

So I removed the main transformer, replaced the transistors with 13009, added 18k on the TL431 divider, and the KV-32FV16 is working again.





I don't know. Maybe it'll die again soon. Obviously I hope not.