Repairnig a Samsung monitor power supply

Thread Starter

Zynch175

Joined Jul 20, 2017
11
Hello

This is my first post, so I hope that I am posting it in the right place.

So I have a Samsung lcd tv monitor which I got from someone for free without knowing it did not work. I opened it and found the TOP247YN broken in half (which enabled me to figure out what it is). I measured all of the surrounding components which seem to be fine, but just in case I replaced the input 150uF 450V cap and I also replaced the broken TOP. After plugging it back in it still did not work, so I measured the output voltages, which were below 1V, but were supposed to be 13V and 5V. After poking around some more, I figured out, that on the primary side of the transformer I get 300V (which might be off as I just remembered that I might have had the meter in AC mode), but on the other side I get absolutely nothing so 0V in any combination of the pins.

So do I have a bad transformer?
I am out of ideas so any help would be appreciated.

Attached are some pictures of the power supply with the TOP circled in one of them and a picture of the monitor info sticker.

IMG_20170311_104812.jpg IMG_20170311_104748.jpg IMG_20170311_104820_01.jpg IMG_20170311_104923.jpg
 

R!f@@

Joined Apr 2, 2009
9,734
Welcome to AAC
300VDC at the cap is OK for a 220VAC input. Did you measure the DC voltage at the primary cap ?
Your transformer is not the culprit.
Check for shorted diodes at the secondary side.
Check for shorted zener diodes around the TOP
 

Dodgydave

Joined Jun 22, 2012
9,153
Service manual here, unfortunately no psu circuit diagram on page 96..just video inputs and screen display chip.


The top chip uses an opto-coupler (white 4pin chip) as feedback, and your transformer has a dual output with separate diodes on heatsinks, i would download the datasheet for it and compare it to your pcb.
Datasheet..

https://www.google.co.uk/url?sa=t&source=web&rct=j&url=http://dalincom.ru/datasheet/TOP247YN.pdf&ved=0ahUKEwjdwMPCt5jVAhWLJ1AKHdkVDZ0QFggmMAI&usg=AFQjCNHpXKVlOdDAZxGN4HXqeysFSdGNbw


Service manual.
https://www.manualslib.com/manual/792462/Samsung-Ldo19ws.html?page=96#manual
 
Last edited:

ian field

Joined Oct 27, 2012
6,539
Hello

This is my first post, so I hope that I am posting it in the right place.

So I have a Samsung lcd tv monitor which I got from someone for free without knowing it did not work. I opened it and found the TOP247YN broken in half (which enabled me to figure out what it is). I measured all of the surrounding components which seem to be fine, but just in case I replaced the input 150uF 450V cap and I also replaced the broken TOP. After plugging it back in it still did not work, so I measured the output voltages, which were below 1V, but were supposed to be 13V and 5V. After poking around some more, I figured out, that on the primary side of the transformer I get 300V (which might be off as I just remembered that I might have had the meter in AC mode), but on the other side I get absolutely nothing so 0V in any combination of the pins.

So do I have a bad transformer?
I am out of ideas so any help would be appreciated.

Attached are some pictures of the power supply with the TOP circled in one of them and a picture of the monitor info sticker.

View attachment 131312 View attachment 131313 View attachment 131310 View attachment 131311
AFAICR: the TOP247 is a Power integrations product and they're pretty good about technical literature. The TV service manual shouldn't be that hard to find, but you can probably scrape by with data sheets and application notes from PI.

Bear in mind that most switchmode converters turn the wick up if the output voltage isn't fed back to the regulation circuit - the bang can be impressive!
 

Thread Starter

Zynch175

Joined Jul 20, 2017
11
Welcome to AAC
300VDC at the cap is OK for a 220VAC input. Did you measure the DC voltage at the primary cap ?
Your transformer is not the culprit.
Check for shorted diodes at the secondary side.
Check for shorted zener diodes around the TOP
I measured the voltages at the primary cap, which is directly connected to the transformer. I desoldered and checked probably every component on the primary side (around the TOP) and they are all fine according to my multimeter and component tester.
I will check the secondary side tomorrow after work.

Service manual here, unfortunately no psu circuit diagram on page 96..just video inputs and screen display chip.


The top chip uses an opto-coupler (white 4pin chip) as feedback, and your transformer has a dual output with separate diodes on heatsinks, i would download the datasheet for it and compare it to your pcb.
Datasheet..

https://www.google.co.uk/url?sa=t&source=web&rct=j&url=http://dalincom.ru/datasheet/TOP247YN.pdf&ved=0ahUKEwjdwMPCt5jVAhWLJ1AKHdkVDZ0QFggmMAI&usg=AFQjCNHpXKVlOdDAZxGN4HXqeysFSdGNbw


Service manual.
https://www.manualslib.com/manual/792462/Samsung-Ldo19ws.html?page=96#manual

AFAICR: the TOP247 is a Power integrations product and they're pretty good about technical literature. The TV service manual shouldn't be that hard to find, but you can probably scrape by with data sheets and application notes from PI.

Bear in mind that most switchmode converters turn the wick up if the output voltage isn't fed back to the regulation circuit - the bang can be impressive!
Thanks for the schematics I will check them and compare the circuits.
And so far nothing has gone bang, and I hope that it will stay that way, and I have a few spare TOP chips if they blow.
 

ian field

Joined Oct 27, 2012
6,539
I measured the voltages at the primary cap, which is directly connected to the transformer. I desoldered and checked probably every component on the primary side (around the TOP) and they are all fine according to my multimeter and component tester.
I will check the secondary side tomorrow after work.

.

The OP mentioned a 450V reservoir cap - that suggests a PFC front end. Its essentially a flyback boost converter and probably also handles the advertised input voltage range. The PFC circuit will try to maintain a specific voltage on the electrolytic. The rectifier should have a decent sized metallised foil cap - that's more about noise filtering than reservoir, but it should hold the AC peak voltage under the load of a DMM.
 

R!f@@

Joined Apr 2, 2009
9,734
The OP mentioned a 450V reservoir cap - that suggests a PFC front end. Its essentially a flyback boost converter and probably also handles the advertised input voltage range. The PFC circuit will try to maintain a specific voltage on the electrolytic. The rectifier should have a decent sized metallised foil cap - that's more about noise filtering than reservoir, but it should hold the AC peak voltage under the load of a DMM.
450V is the standard voltage for PFC but OP's monitor supply does not have a PFC front end.
 

ian field

Joined Oct 27, 2012
6,539
Welcome to AAC
300VDC at the cap is OK for a 220VAC input. Did you measure the DC voltage at the primary cap ?
Usually its nearer 320V - but it depends on how much cable between you and the sub station.

One PC PSU manufacturers was in the habit of putting 150V MOVs across each of the electrolytics in the voltage doubling capable rectifier/reservoir. they often came in with both MOVs shattered and a fir bit of other damage by the time the fuse let go.
 

Thread Starter

Zynch175

Joined Jul 20, 2017
11
Ok so I just pulled every component after the transformer out of the pcb and tested them.
Everything looks to be good the only thing that I am not sure about is the industor as my chinese component tester decided it was a 0.22 Ohm resistor and the other inductor is actualy just a jumper so I don't think it's bad.

But what confused me was when I had it plugged in and I was poking around and measuring voltages, and I then unplugged it and I just measured the voltages on the caps on the secondary side (to make sure ai don't get shocked) and they were exactly on the oltages they were supposed to be at (5V and 13V) but while it was plugged in the maximum I could measure was about 0.5V and if I plugged it back in whilst there was some charge in the caps they would just continue to discharge.
 

ian field

Joined Oct 27, 2012
6,539
Ok so I just pulled every component after the transformer out of the pcb and tested them.
Everything looks to be good the only thing that I am not sure about is the industor as my chinese component tester decided it was a 0.22 Ohm resistor and the other inductor is actualy just a jumper so I don't think it's bad.

But what confused me was when I had it plugged in and I was poking around and measuring voltages, and I then unplugged it and I just measured the voltages on the caps on the secondary side (to make sure ai don't get shocked) and they were exactly on the oltages they were supposed to be at (5V and 13V) but while it was plugged in the maximum I could measure was about 0.5V and if I plugged it back in whilst there was some charge in the caps they would just continue to discharge.
When I started training; I was taught the "split half" method and warned that shotgun component testing/replacement was "bad form".

All good and well for an AM radio where you can poke the volume pot with a screwdriver and listen for hum, but a SMPSU is an entirely different animal. Its essentially a servo system and a faulty component can be over the other side of the board from the loud bang it caused.

There is an element of shotgun testing - but with experience, you can target the most likely components.

You probably do need to do a bit of exploring to find out why the TOP247 went bang.

A common trick is to put a 60W lightbulb across the fuse holder when you fire up the rebuild - the PTC characteristic of the filament isn't ideal, but it has been known to stop it going bang again.
 

Thread Starter

Zynch175

Joined Jul 20, 2017
11
When I started training; I was taught the "split half" method and warned that shotgun component testing/replacement was "bad form".

All good and well for an AM radio where you can poke the volume pot with a screwdriver and listen for hum, but a SMPSU is an entirely different animal. Its essentially a servo system and a faulty component can be over the other side of the board from the loud bang it caused.

There is an element of shotgun testing - but with experience, you can target the most likely components.

You probably do need to do a bit of exploring to find out why the TOP247 went bang.

A common trick is to put a 60W lightbulb across the fuse holder when you fire up the rebuild - the PTC characteristic of the filament isn't ideal, but it has been known to stop it going bang again.
Ok, so right now I know at least that no component that I tested is completely dead.

So If you could give me any hints as to where I should look I would really appreciate it as I have a preety basic understanding of things, but I will go through the TOP's datasheet to see if I can figure out anything.

As to the light bulbs the closest I can get are the halogen bulbs, but we mostly use the energy saving bulbs and leds.
 

ian field

Joined Oct 27, 2012
6,539
Ok, so right now I know at least that no component that I tested is completely dead.

So If you could give me any hints as to where I should look I would really appreciate it as I have a preety basic understanding of things, but I will go through the TOP's datasheet to see if I can figure out anything.

As to the light bulbs the closest I can get are the halogen bulbs, but we mostly use the energy saving bulbs and leds.
AFAIK: halogen bulbs have more turn on surge - so less protection.

If the TOP247 burst, it probably indicates core saturation.

There is probably a small electrolytic nearby that stores a sample of the output voltage for the regulation - dry joints, high ESR electrolytic are the most common faults. Check whether there's an opto coupler and possibly a TL431 - some of the PI chips do primary side regulation which makes those unneccessary.. Primary side regulators tend to use higher resistances to sample the voltage, which are more prone to failing open.

If its secondary side sensing, it requires a supply which is also the source for the sample voltage - it has to be smoothed, and the elecro' can fail.

The main output reservoir/filter electrolytics can allow enough ripple to confuse the regulator.

Usually - SMPSUs just stop if a secondary side rectifier shorts - but I have had to rebuild a few that went bang because of it.
 

Thread Starter

Zynch175

Joined Jul 20, 2017
11
Check whether there's an opto coupler and possibly a TL431
Yes there is an opto coupler and a 431 chip

high ESR electrolytic are the most common fault
The caps seem to be ok, but I will check them again in case I did not read the measurements correctly

Usually - SMPSUs just stop if a secondary side rectifier shorts - but I have had to rebuild a few that went bang because of it.
The two diode packages (MBR20100CT) seem to be ok, as all four diodes have approximately the same forward voltage of about 250mV, but if I read the datasheet correctly this seems a bit low IMO.
 

ian field

Joined Oct 27, 2012
6,539
Yes there is an opto coupler and a 431 chip


The caps seem to be ok, but I will check them again in case I did not read the measurements correctly


The two diode packages (MBR20100CT) seem to be ok, as all four diodes have approximately the same forward voltage of about 250mV, but if I read the datasheet correctly this seems a bit low IMO.
Any fault between the resistor divider sensing voltage, all the way back to the TOP chip can go bang when you fire up after repair.

Poking around a non starting SMPSU is another way to make it go bang.

Dodgy electrolytics have usually lost electrolyte and feel noticeably lighter when to take them out of the board.

They tend to run hot when failing - sometimes deterioration of the plastic sleeve is obvious. They can leak past the rubber bung and even corrode through one of the legs - usually makes a nasty smell when you try to unsolder it.
 

R!f@@

Joined Apr 2, 2009
9,734
What I do if the supply starts at power input and shuts down, is replace the TOP to confirm a bad TOP. This has happened.
But before that I would isolate the supply form mobo and check if the mobo is causing the shut down

Second thing is to isolate the inverter stage and check the PSU
 

Thread Starter

Zynch175

Joined Jul 20, 2017
11
What I do if the supply starts at power input and shuts down, is replace the TOP to confirm a bad TOP. This has happened.
But before that I would isolate the supply form mobo and check if the mobo is causing the shut down

Second thing is to isolate the inverter stage and check the PSU
The supply is not starting the only time I can measure the correct output voltage (I did not try this on the inverter) is imedeately after powering the board down.

Any fault between the resistor divider sensing voltage, all the way back to the TOP chip can go bang when you fire up after repair.

Poking around a non starting SMPSU is another way to make it go bang.

Dodgy electrolytics have usually lost electrolyte and feel noticeably lighter when to take them out of the board.

They tend to run hot when failing - sometimes deterioration of the plastic sleeve is obvious. They can leak past the rubber bung and even corrode through one of the legs - usually makes a nasty smell when you try to unsolder it.
None of the caps has leaked or is showing any other signs.

I'm being very careful about poking around so lets hope that I don't blow it up.

So the resistor divider should be around the 431 chip?

I will try to figure out if there are any bad resistors or capacitors around there (they are mostly smd).

Could the problem lie somewhere in the invering side, as I have not really looked around there roo much.

Oh and how likely is it that any bad smd components (resistors, capacitors, transistors) could be indentified visualy, I really do not want to desolder all that stuff just to measure them and smd capacitors don't even have markings so how am I going to know if they are bad.
 

dendad

Joined Feb 20, 2016
3,514
Just out of interest, can you run it on lower i/p volts? Do you have a Variac? It almost sounds like it is shutting down on over volts as when you turn it off, the supply drops and the TOP starts up for a bit. And if you can pick up a "normal" light globe or two somewhere they are a very useful addition to the test gear as mentioned above.
 

ian field

Joined Oct 27, 2012
6,539
Just out of interest, can you run it on lower i/p volts? Do you have a Variac? It almost sounds like it is shutting down on over volts as when you turn it off, the supply drops and the TOP starts up for a bit. And if you can pick up a "normal" light globe or two somewhere they are a very useful addition to the test gear as mentioned above.
Some people do that - but some SMPSUs have UVLO - so nothing happens till you hit the start threshold, then you find out if it goes bang or not.

PFC front ends are flyback boost converters and usually take care of universal input voltage range. It should try to start somewhere around 85V. It can still go bang - but you've given it a lot less energy to go bang with.
 
Top