Samsung 32" S32D850T Monitor Repair story

[sausage]

I assume this is the place to share repair stories as well? If not I'll have it removed. Fixed and blogged about the repair over the weekend:

http://blog.waynejohnson.net/doku.php/s32d850t_monitor_repair

[fzabkar]

Congratulations.

The bridge rectifier and MOSFET chopper are generally the major culprits in cases involving a blown mains fuse. You could have saved yourself a handful of fuses if you had started there.

[sausage]

Yep, certainly excellent education for next time. But I am glad I went down the rabbit holes. I spent a lot of time getting very intimate with many components trying to locate datasheets and all sorts of things. But next time, yep straight to those components if it's not a simple cap

[lolimpol]

Hi Wayne,
I've been trying to reach you on your site but your contact form is broken

I have a similar 27" Samsung monitor, and I was wondering if you could help me on my way.

The monitor turns off after a random amount of time, sometimes 30mins, sometimes 30 seconds. It doesn't turn on again unless you unplug it, wait and plug it back in. I probed the a13V pad that goes to the logic board, and it slowly settles down from 5 to around 0V (from the Capacitor I assume) also when I flip the board on the table; so out of the monitor, it turns on/off sometimes.

The rectifier reads a nice 350ish volts DC when turned off, and to be quite honest I haven't dared to probe around more on the AC side.

Any idea where to look?

Thanks so much.

[sausage]

Hi lolimpol, really sorry I didn't get this notification. In fact I don't think I get notifications for post changes. I'll get that switched on.

My contact form should be working, I got a message on it the other day, but maybe there is an occasional problem. Maybe we can help each other. Keen to see if I can help you.

[andy3055]

Nice job and great documentation.

[jacamo]

You should invest in building a inline light blub short circuit tester/current limiter.  Would have saved you a bunch of fuses 

Here is a link to how to build one and they are super simple and can usually be made with parts lying around.

https://antiqueradio.org/dimbulb.htm

Or just google "lightbulb short circuit tester" or something along those lines.

Hope it helps!

[andy3055]

AKA, "Dim bulb tester."

[remaker]

Note: I had this exact same monitor and it would turn off hard after a few minutes or hours, meaning you had to remove mains power to restart it.

I replaced the KF12N60F MOSFET, which seems to be a problematic device in the power supply.

It is not clear if the device itself was bad, or if it was solder joints, silicone decay, or poor heatsinking. Since the device is known to be prone to failure, I replaced it, used CPU heatsink compound between it and the heatsink, and didn't add the white silicomne afterwards. Not sure what solved it, but it looks solved. Will write more if the failure returns.

(LOL, *literally* after posting this message it crashed again. So that wasn't the problem. Back to square one)

[0xG1L]

Hi,

I have the same monitor and I'm facing the same problem - my monitor turns off after a short time and can be powered on only after power is removed.
Have you managed to fix your monitor ? Which component should I  look into ?
Tnx

[m k]

Seems to be a bad connection when flipping a board is changing things, but is it supply chain or something smaller.
Is there any life after it goes off?

Old monitors had no default geometry, so OSD piggypacked incoming signal.
Newer ones have and small "no signal" rectangle or something similar is present even without input signal.

An internal control signal may be a single edge instead of a level.
So when something else goes wrong for a moment the situation is not recovering before that edge is recreated.

Since removing power cord is resetting the issue it must do something that other things can't.
This indicates that standby circuitry/functionality has something to do with it.
Maybe there is a rising edge of PowerGood.

If possible, measure how much current it takes.
There used to be models that consumed almost the same all the time, maybe that is in the past.

[0xG1L]

Thanks for the answer.

When it shuts off, its totally dead - power led turns off, nothing responds, and there is no "no signal" sign.

I have also noticed that if I wait long enough (a few minutes) it comes back to life until it shuts off again.
I connected the main power through energy meter -
when monitor is in standby mode (when there's no video signal), or even when its turned of manually - power consumption is 0.2 watts steadily.
when the monitor shuts off by itself, power consumption drops to 0 watts for a few seconds and then it moves between 0 to 0.3 watts repeatedly.

Does this information help focusing on the problematic component ?

Thanks

[madires]

Yesterday I fixed a Belinea (BenQ inside) computer monitor which didn't turn on anymore. Everyone would guess it must be the SMPSU. No, SMPSU is fine. Turns out it's the PCB strip with the push buttons. It has three pairs of mini push buttons, each pair shares one signal line, one push button shorts the line to ground via a 1k resistor and the other one via a 10k. One push button became conductive, around 5k when not pressed. That screwed up the MCU's button scanning causing the power on/off button to be ignored. All I had to do was to replace that bad push button.

[Veteran68]

Yesterday I fixed a Belinea (BenQ inside) computer monitor which didn't turn on anymore. Everyone would guess it must be the SMPSU. No, SMPSU is fine. Turns out it's the PCB strip with the push buttons. It has three pairs of mini push buttons, each pair shares one signal line, one push button shorts the line to ground via a 1k resistor and the other one via a 10k. One push button became conductive, around 5k when not pressed. That screwed up the MCU's button scanning causing the power on/off button to be ignored. All I had to do was to replace that bad push button.

Interesting that you mention this. My son-in-law gave me his old 32" FHD monitor that "died" on him, just because I was curious if I could fix it before we tossed it out. I disassembled it and checked the SMPS and it was fine, the 12V output voltages to the control board and both 24V backlight inverter voltages were present. But the power button wouldn't do anything. I confirmed the power button was making contact when pressed, but didn't check the other buttons on the PCB strip. When the obvious didn't pan out, I got busy with other stuff and set it aside for a rainy day. Now you've got me wanting to go back and check those other buttons.

[m k]

Does this information help focusing on the problematic component ?

Hardly a component, but maybe something.

Since it is restarting the fault is obviously not resetting its status, but consumed watts are very low.

If 0 watts is real there can't be any real higher leftover or memorized states.
Only thing that can have a sort of memory is a stand-by circuitry.

If power button is stuck and is constantly powering the thing there shouldn't be any stand-by possibilities.
For all normal situations power LED should also remain on.

Is there a setting how monitor behaves after power outage?
Setting that to stay off will reveal how power is actually going off internally.
If after that setting the monitor is still turning on the power is not really going off internally.

Power LED is many times connected to stand-by circuitry.
Measured wattage is also indicating that it's a stand-by circuitry that is acting up.

Is there any life in power LED when fault is on?

[remaker]

Some further (unhappy) updates.

At this point, I was thinking about cracked solder joints. I resoldered a number of places and even ran a heat gun all over the power supply board to reflow. Still no joy.

Often, a sharp mechanical jolt to the power board will restore it to operation, which had me focused on the solder joints.

The AC side of the rectifier is always fine when in failure mode, but the voltage on the far side drops precipitously when failing. I didn't make notes, just going from memory here.

I used a Kill-A-Watt that shows power draw 55W when working, drops all the way to zero when failing. All LEDs off when failing, front panel unresponsive.

I'm wondering is there is some hairline crack on a trace or something. The board routing tolerances look very right, with some thin traces running under very small 0201 zero ohm resistors.

Nothing obvious has jumped out at me. Lot of posts about failing this supply and whole power supply replacements are expensive and likely to meet a similar fate. Seems like a design issue.

[m k]

The AC side of the rectifier is always fine when in failure mode, but the voltage on the far side drops precipitously when failing. I didn't make notes, just going from memory here.

What you mean by far side?

Normal diode bridge rectifier can't drop DC if AC is not dropping.
Or can but then it would be a shunt of some sort.