Manson HCS-3602 SMPS fault

[iperyth]

Hey folks. I recently bought broken SMPS, Manson HCS-3602. As far I know it is broken in two ways:

1. There was initial fault that made it impossible to use. PS was passing selftest, and displayed voltage around 10V, slowly dropping to around 1V.
2. During the repair owner somehow shorted MOSFET in PFC section. Fuse popped, story ends.

So firstly I checked the MOSFET, and it was indeed dead. Replaced both of them, diode seems to be ok (bought another one anyway, it's exactly the same).
Fuse stopped popping. But now it endlessly cycles power on. I can hear clicking of relay, selftest shows for a split second, then disappears.
Video:

Do you have any idea where should I start searching for more troubles, or how I can isolate the issue. I suspect that after relay kicks in something is shorting to the ground, but not really sure if it's high or low voltage side. Electrolytic caps seems to be ok. The only thing that looks suspicious to me is slightly low resistance between output around 80 oms and rising to 180 oms after few secounds.

Regards
Mark.

[iperyth]

After messing around I suspect that PFC controler IC (ICE1PCS01) might be broken. I found that resistance between GATE and GND in around 40 ohms which seems to be too low. Any thoughts?

[xavier60]

The PSU as a whole shouldn't care that the PFC isn't working so disable it completely for now. Remove Q3.

[iperyth]

Well, that worked. Sorta. After removing q3 magic smoke appeared from relay. After checking DC rail resistance dropped to zero. For some reason even without IC and transistor that was driving relay something went bad. Anyway, I removed both mosfets and diode, checked the bridge again and everything else seemed to be ok. I will sort PFC later.
After powering up the second issue appeared. PS performed selftest, and after passing it voltage dropped from 10 to around 1. Using encoders creates some artifacts, nothing more. After switching off display was still running for around 30s.

[xavier60]

Well, that worked. Sorta. After removing q3 magic smoke appeared from relay. After checking DC rail resistance dropped to zero. For some reason even without IC and transistor that was driving relay something went bad. Anyway, I removed both mosfets and diode, checked the bridge again and everything else seemed to be ok. I will sort PFC later.
After powering up the second issue appeared. PS performed selftest, and after passing it voltage dropped from 10 to around 1. Using encoders creates some artifacts, nothing more. After switching off display was still running for around 30s.

I can't see the encoder in the schematic I'm using. Anyway the only suggestion I have is to measure the parallel load input pins on the 74HC165 looking for possible leakage pulling a pin down.

[iperyth]

I am using those schematics.

Quick update: I was poking few thing to see if it is a bad connection issue since manual soldering from factory was really bad quality (not like mine is better  ).
Also, I checked if SW2 switch is even working. Guess it is, because now after performing selftest after around a second overvoltage protection trips. Interestingly, selftest goes without any issue which makes me thing that it's connected to previous defect.

Video before OVP:
Video after OVP:

[xavier60]

You could try measuring MCU C.V. to see if there is a regulator problem.
The input circuitry to U4A doesn't make much sense unless C70 is really jumpered out?
Wee need to compare  MCU C.V. voltage to output voltage.

[iperyth]

C70 and C71 are indeed jumpered.
Both output and MCU C.V. gives no reading except in the moment of turning on. Both spikes out of range and slowly drops while the cap discharges.

Update: looks like remote control modę bypasses overvoltage protection circuit.
Sometimes it works in normal mode as well, can’t figure out what is causing OV yet.
Under remote control mode it outputs 35V with corresponding 2.500V on MCU C.V pin.

[iperyth]

Another day, another news. Previous owner told me that "for some reason it stopped working". Looks like that "reason" was liquid, and had lots of sugar content. There was some residue on the front board that caused some shorts (on rotary encoders for sure, prolly some more). Cleaned with some IPA, now it's nice and shiny.
Sadly, that didn't change nothing at all.
I can't really check if encoder part of the circuit works, but I use remote control mode and found out that U8 has correct voltage.
MCU outputs 5V in both pin 13 and 14 regardless of voltage on pin 34. What could cause that? Does it look like MCU is damaged, or something might be shorted?

[xavier60]

There could be leakage holding something in the wrong state on the microcontroller or U7.
I usually use my DSO to check all input pins to see if the state is expected or invalid states where the voltage is somewhere between zero and 5V or randomly fluctuating.
This includes the encoder pins.

[iperyth]

I am going to. Since I have real experience with microcontrollers I am trying to figure out if it even look like a problem with that one, or I should start looking somewhere else. Guess that’s the best idea so far.