Acer XB321HK monitor power board voltage issues?

[Tim88]

I picked up a non working Acer XB321HK gaming monitor. These were about $1600 back when they came out so thought it would be a fun project. The monitor will not power on though there is a blue LED on the touch switch that is on. The monitor evidently got hot as there are black smoke trails on the shields for the boards. Anyway, the voltage going out of the pwer board to the main board is 12.1V and I originally thought this was good. However, after much reading and research, 19 volts seems much more common. 19V does indeed come out of the chopper transformer going into pin 8 of IC701. The MOSFET Q701 at the drain reads 12.1V and I am stumped as to what the issue is. I found a 90 W TEA1752T and TEA1791T demo board pdf that has a similar circuit that puts out 19.5V. That power button LED seems a tad dim which also leads me to believe the main voltage is under powered. I attached that demo board schematic pic and what I have mapped so far of the power board. A little help and direction would be greatly appreciated as I am at a dead end...

[Tim88]

My power board cable connector to the main board does not have the voltages marked unfortunately. After looking at dozens of interweb pics of similar Acer predator boards, I finally found one with the output voltage of 19V labeled. So, 99.9% sure this should be putting out 19V and it is only dealing 12.1V nice and steady. When you plug it in, kilowatt meter shows it draws no amperage, hit the power button, goes up to 70 mA. All caps were replaced though none shows sign of strain. So what is preventing this from coming up to 19V???

[Kurets]

Both of the example schematics you show are regulated by a TL431 on the secondary side. If the feedback circuit works (with some other fault causing the erroneous output voltage) you should be able to measure around 1V on the cathode/output pin, corresponding to full-on or loss of regulation.

If instead you see a voltage close to 12, it would inducate that the feedback is faulty, resistor network or TL431.

You can further try to confirm that the optocoupler is on/conducting.

But all of this is moot if you have not been able to identify the source of any soot/smoke you have seen.

A simple question is: How have you confirmed that the output is "solid"? Measuring with a DMM would give you the average voltage and ignore crazy ripple which could cone from repeated restarts of the DCDC.

[fzabkar]

If the supply is regulating correctly, then the Ref pin of the TL431 should be sitting at 2.5V.

[MathWizard]

In that schematic anyways, Q4 is a mosfet-rectifer, and the TEA1791 controls and drives the gate, and monitors the current and input voltage. Yeah another thing to check would be R33/C31. They might get pretty hot, and drift or worse.

A lot of PSU's would use a diode, or dual diode, as the output rectifier, but they usually still have some RC snubber circuit like R33/C31 (perhaps to GND tho), and they often take a pounding and get hot, and are only priced/sized to last so long.

Do you have an oscilloscope?

[Tim88]

Hi, thanks for the response. My thoughts were also on the regulation side. That 4870 and 18700 I cannot confirm without removing, may have to do that, they are those tiny SMD (I hate those). I measured the ref pin on the TL431 and it is 2.49V on the button. Maybe the optocoupler may not be conducting? I wish I could identify where the smoke was coming from, so many soot tattoos on the shielding. As for the 12V output, steady should have been used, not solid. No fluctuation on the voltage. Will look into how to verify the optocoupler is working. 

[Tim88]

Measure 2.49V so am assuming we are good on the low voltage side.

[Tim88]

I do have a scope, what do you want to see?

[Tim88]

Did some more testing and I think I narrowed down the issue. The voltage across the LED in the optocoupler is only 1 volt. It is a 817B optocoupler (mis labeled on schematic) and the spec sheet has the typical forward voltage of 1.2V. So either the TL431 is weak or one of those precision SMD resistors is out of spec?

[fzabkar]

Everything is working correctly. The design voltage of your regulator is in fact 12V.

  4870 / (4870 + 18700) x 12V = 2.48V

[Tim88]

You guys aren't gonna believe this, but it was the one capacitor I did not change! A freakin poly cap, 270uF. I never worked with a poly cap, read a little on them and they were supposedly very durable so I figgered it would be ok. Well, son of a gun... Learned something new. Thanks to all that helped!