Samsung PS51D8000FV power supply clicking

[BoringName]

Hi,

This TV worked and then just went off one day and started clicking. Usually 5 or 6 times and then it would stop and the standby light would do 5 double flashes and repeat. If it was left off for a few hours or overnight it might work again in the morning but only for about 20 minutes. Now it doesn't work at all, just does the clicking.

The power supply board is a BN44-00446A

I found a couple of capacitors with high ESR and replaced them, that didn't fix it so I replaced a bunch of others. The only ones I haven't replaced are the high voltage (filter capacitors?). 450V 63 uF x 4 and 250v 180uF x 2 but they all tested good with an ESR meter and seem ok.

I unplugged all the boards from the power supply and shorted PS_ON to ground. It only clicks once and all the voltages seem to be ok. Standby at 5.21V along with all the other 5 volt points and 15.39V for all the 15 volt points. Rock steady with no fluctuations.

I bridged VS_ON to a 5.3 volt pin and this makes it do the clicking thing. From everything I have read this should make the board power up and supply VS and VA voltages. That's why I think the power supply is the issue with the TV. When it is doing the clicking thing, the VS and VA test point voltages seem about right. 205V and 58V. But they pulse on an off with the clicking.

I've tested all the resistors and diodes on the top of the board, they are ok. I haven't gone through the underside yet as there is heaps of surface mounted elements. I've looked over the board and can't see anything blown up or bad solder joints. I've tried looking for a short but some of the rails are connected to ground with a resistor so it's been tricky. A schematic would be great!! I refreshed the solder on all the transformers. I was hoping someone could point me in the right direction. It seems to work ok until it needs to supply the high voltages.

There is a replacement board at mxmee but once they tack on ridiculous postage charges and fees it will be over $200 and I would like to try and avoid that.

Thanks.

edit: Just found a service manual so I might be alright.
edit: Nope, service manual was pretty much useless.

[BoringName]

Just trying to get my head around something.

There is a transformer at the bottom right of the board if you google BN44-00446A.

It has 5 pins on each side. Currently all 5 pins on the right are shorted to ground. 3 of them go to some rectifiers MBRF20100CT and SDURF1030CT. They both do the same thing but with different voltage ratings. Pins 1 and 3 are diodes to pin 2.

From top to bottom on the right of the transformer -
Pin 1 is on a small trace with just a diode on it connected to another trace. The transformer side of the diode is shorted to ground, the other isn't.
Pin 2 is on the same trace as pins 1 and 3 of SDURF1030CT. Pin 2 of SDURF1030CT is not shorted to ground. There is a capacitor between the trace and the trace pin 2 is connected to.
Pin 3 has a path to ground.
Pins 4 and 5 are on the same trace with pins 1 and 3 of MBRF20100CT. Pin 2 of MBRF20100CT is not shorted to ground. There is a capacitor between the trace and the trace pin 2 is connected to.

I can't see anything else on those traces that could be shorted to ground. The 2 capacitors seem ok but even if they weren't, they are not connected to ground.

So I guess the transformer is confusing me. It looks like there are 4 pairs of wires on each side which is 4 primary and 4 secondary windings right? Pins 2, 3 and 4 have two wires connected, other 2 pins only have one. I'm assuming there would probably be some continuity between some of those pins but it's confusing me.

Surely the whole right side of that transformer shouldn't be shorted to ground right?

Could it be possible that MBRF20100CT and/or SDURF10300CT are shorted to ground through the package?

[BoringName]

Ok it seems I may have been misled by a video claiming you can just short VS_ON to a 5 volt line and the TV will give VS and VA voltages. I think it also needs a signal from the logic board so that could be causing the clicking because that signal is missing. One of the pins to the logic board is VS_Con so that's probably it. I already replaced the Xmain board so my focus is on the Y board now.

The TV does turn on with the main and logic boards connected but the logic board shows an error code with the LED due to the missing X and Y boards. I got sidetracked with the power supply clicking with the VS_ON bypass. Frustrating.

Anyway.... lets see how the Y board goes.

[BoringName]

I'm focusing back on the power supply. I'm 99% certain it should power up with PS_ON jumpered to ground and VS_ON jumpered to a 5 volt supply.

While I couldn't find a schematic for BN44-00446A, I have found a schematic for BN44-00333A which is a very similar supply.

It has a PIC16F882 (ICQ802) microcontroller that seems to be the brain of the power supply. The schematic details what each pin is so I'm confident I should be able to test each pin and narrow down what part of the board is failing.

[BoringName]

I realise I'm talking to myself at this point but I'll keep going for posterity.

So, even though BN44-00333A has a very similar layout, it uses different microcontrollers. Referencing some other schematics and through some tracing I've worked out the pinouts.

Probably goes without saying that some of these might be incorrect, it's just my best guess with my limited experience.

MCV14A - This one seems to do most of the work.
Pin - Description
1: VDD - powers the chip. 5.2 volts. This is constant even when the relay is clicking.
2: VS_ON - Supplied by the logic board, I currently have it jumpered to one of the 5 volt rails.
3: AC_DET - While the board is powered it shows 1.5-2.0mv but once the power is switched off it sits at 4.35V, supplied by capacitors I assume. Sits at zero without VS_ON jumpered.
4: PS_ON - Normally connected to the mainboard. I currently have it jumpered to ground.
5: PFC_OK - Is what it's supposed to be, on this board it just goes to a 1k ohm resistor and a spot for an LED that is not present. This sits at 4.38V while the relays are clicking and without VS_ON jumpered
6: Main_On - Pulses between 0 and 9 volts while the relay is clicking. Seems a little high. It goes to an optocoupler linked to the hot side, pretty sure this is what turns on VS and VA voltages.
7: B15V_ON - Forgot to measure this one....
8: B5V_ON - pulses between 4.6v and 9 volts while the relay is clicking. 4.87v constant without VS_ON jumpered.
9: Relay_ON - Pulses between 4.7 and 8.7 volts while the relay is clicking. 4.7v constant without VS_ON Jumpered
10: A5V_ON - 4.91V this stays constant the whole time.
11: D15V_DET - Sits at 2.7V
12: VS_DET - Fluctuates between 3.7V-3.8V while the relay is clicking. The Capacitors probably smoothing that one out.
13: A5V_DET - 2.59 volts.
14: GND

2A20118B - Manages the PFC.
Pin - Description.
1: ZCDB - Goes to pin 1 of first PFC via a 31k ohm resistor. Sits at 0.552v
2. ZCDA - Goes to pin 1 of the second PFC via a 31k ohm resistor. Sits at 0.465v
3: VINAC - Goes to pin 2 of the first PFC via a bunch of resistors. Sits at 4.66v
4: No idea - Goes to a grounded capacitor. Sits at 4.99v doesn't fluctuate at all with the clicking.
5: No idea - Goes to a grounded capacitor via a 82k ohm resistor. Sits at 4.93v
6: No connection
7: No idea - Goes to ground via a 32k ohm resistor. Sits at 1.64v
8: No idea - Goes to a grounded capacitor. Sits at 3mv
9: AGND
10: No idea - Goes to a capacitor that goes to ground via a 12k ohm resistor. Also a second grounded capacitor and a 1M ohm resistor to ground. Sits at 4.3v doesn't fluctuate.
11: VSENSE - Not sure if this is HVSENS or VSENSE. Sits at 2v doesn't fluctuate. Capacitors keep it constant.
12: HVSENS - Not sure if this is HVSENS or VSENSE. Sits at 2v doesn't fluctuate. Capacitors keep it constant.
13: No idea - Goes to a grounded capacitor. Rises to 400mv, goes OL then back to zero and repeats. Same with VS_ON Jumpered or not.
14: No connection
15: No idea - Goes to ground via a 220k ohm resistor
16: No idea - Goes to ground via a 220k ohm resistor
17: GDB - Sits at 0.9mv
18: PGND
19: GDA - Sits at 0.6mv
20: VCC - I think this supplies power to the chip and is also the collector for the transistors where GDA and GDB connect to the base. Sits at 16.38v doesn't fluctuate

From what I can see, MCV14A is the only thing that controls the relay so when it's clicking on and off, that is this chip doing it. Given that the chip never loses power and AC_DET/PFC_OK behave the same way regardless if VS_ON is jumpered, I'm assuming this chip is either broken or performing a reset due to VS_DET not hitting a threshold or another chip cutting the power ie) the DET pins drops to zero, it's going to be 2A20118B or the PWM chip (SQT8011K).

I have found instructions on how to bypass MCV14A but I think I will leave that as a last resort and keep checking 2A20118B/SQT8011K first.

edit: Added voltages for the 2A20118B chip. That's the voltage they sit on without VS_ON jumpered. Unless it's stated they don't fluctuate, when VS_ON is jumpered they generally fluctuate a bit but are around the listed voltage.

[BoringName]

I ignored this for a few days so the board has just been sitting in a spare room doing nothing.

Plugged it in today and got one click. Thought I must have dislodged the VS_ON jumper but I hadn't!!!

I measured VS voltage and it was sitting on 205 volts like it should be. Before I could check anything else it started clicking 

All up it probably lasted about 30 seconds before the clicking kicked in. So at least I can be 100% sure now it is definitely the power supply. I wasn't fully sure jumping VS_ON would power up the supply when it wasn't connected to the other boards in the TV, now I know it does.

Now it's back to clicking at soon as it's turned on.

I feel like it has to be a capacitor causing problems. Leaving it off for a while allowed them all to discharge fully. I'm going to try discharging every capacitor I can find and see how that goes. I've discharged all the electrolytics before so I'm leaning towards a ceramic being the issue.

edit: Tried discharging all the electrolytic and ceramic capacitors. no change. Discharged all the SMD capacitors and it powered up for about 10 seconds, clicked, stayed on for another 10 seconds then went back to the clicking multiple times. So looking like it's an SMD capacitor. Now I just need to trial and error to find the little bastard!

[BoringName]

So I had some success messing around with the VS circuit. I found if there was an amount of residual voltage in the circuit it would click. There are 3 capacitors. 180uf 250v x 2 and a 100nf 250v. I discharged the capacitors but could never get the residual voltage to completely disappear. If I got it to around 10mv or less the board would power up and hold voltage for a good 30 seconds or so and then start clicking again. Very odd. I was able to reliably repeat this phenomenon.

I replaced the 2 x 180uf and that didn't fix it (they had tested good with an ESR but I figured it was worth a shot). I replaced the 100nf and it stayed up for over a minute before clicking. The longest it had stayed up since it began clicking all the time. I was able to measure the VS/VA voltages and noticed they were 206 and 59 when they should be 205 and 58. So I adjusted the potentiometers (note the 100nf capacitor was connected to VS and the potentiometer).

I had tried turning the potentiometers as low as they would go weeks ago and it didn't help. But this time it worked. I don't know if changing the cap helped or what is different now but supply stayed up with VA set to 58 volts and VS set to 198 volts. I turned it back and forth a few times and confirmed 198 was stable, 205+ clicked.

So I thought I would just play around and try and get the highest voltage I could without it clicking and stick it back into the TV to see how it went. I got up to 200 and then while I was lifting the board up to adjust the potentiometer, I didn't notice the jumper wire I had for VS_ON got caught on part of the board, when I put it back down and turned it on nothing happened. That's when I noticed the wire. There was no smoke or smell and I can't see any burnt components but all I get now is standby voltage

The microcontroller is still getting power and sending Relay_On and A5V_On voltages. The relay has voltages showing on the contacts but it's not clicking so I think I fried it. I'll test it out tomorrow. The relay is 5 volt and it's possible the VS_ON wire may have bridged a 15 volt line to it. I'm feeling pretty dumb right now.

Edit: Relay is ok. Looks like I've blown the transistor that Relay_On goes to the base. It gives a path to ground for the relay. Positive on Base and negative on Emitter is giving me OL when it should give me around 0.5v. Negative on Base and Positive on Emitter is giving me 0.8v when it should be OL. I've compared it with transistors with the same part number elsewhere on the board just to confirm. I also bypassed the transistor to ground and the relay clicked on.

[wraper]

Try replacing small electrolytic capacitors on the primary side.

[BoringName]

Try replacing small electrolytic capacitors on the primary side.

I've replaced a few already and what is left all test ok on the ESR but I think you are probably right. I should have just done them all from the start regardless of the ESR values and it probably would have saved me a lot of troubleshooting.

Hopefully Jaycar have a transistor that's suitable so I don't have to wait for mail over xmas I'll grab some caps at the same time.

[BoringName]

I replaced the transistor I blew up and got it to stay up with under 200v just like before. All good.

So I turned the potentiometer up a bit until it clicked and replaced all the small electrolytic capacitors one at a time. Checking each time if it stopped it clicking.

After changing 7 capacitors it still just clicks. But now, even if I turn the potentiometers all the way down and discharge the board to under 10mv like before, it just clicks all the time. So I've gone backwards... I checked all the capacitors and I used the correct sizes and haven't stuffed up anything with the soldering. All the capacitors I replaced were showing good ESR so I don't know what's going on. I have a feeling putting all the old ones back in will be a waste of time.

I'm fairly sure it's something in the VS circuit of the board. Considering it works fine when VS_ON isn't jumpered and just it's behavior the last few days. Changing the capacitor in circuit with the potentiometer improved things. Maybe the potentiometer is playing up....

Frustrating!

[BoringName]

The plot thickens.

So I couldn't get it to come back on. Just clicking every time no matter what I did. I had an idea that maybe the VS_DET pin on chip MCV14A might also measure the voltage and cut out if it's too high or something. Which would explain why higher voltages previously would click when lower voltages didn't.

So I soldered the ends of a 56k resistor where a metal wire is used to bypass some traces which goes to VS_DET. Being a dumbass I forgot to remove the metal wire so really the resistor wouldn't do anything but I figured I would turn it on anyway before turning the iron back on. The thing stayed on. I measured the resistance across it and it was zero so the resistor was doing nothing.... I removed the resistor and added some more solder to the wire.

It now stays on with VS at 203 volts.... if I turn it up any higher it will click off but it seems to be stable at 203 and VA at 58. So weird. The solder didn't look bad on the wire before I messed with it. I can't get my head around how that could have made it better. Obviously it's not fixed if it clicks at higher voltages but it's very strange.

I'm tempted to put it down to 200 volts to give a buffer and see if it works in the TV. Maybe not turn the brightness up to high and it will be ok?

edit: It powers up in the TV alright but now there is a lot of red where there shouldn't be. From what I have read that is probably due to the voltage being too low I tried getting the voltage higher but it just started clicking. The TV picture was fine before it started the clicking rubbish, hopefully something else hasn't broken....I guess I'll keep playing around with the board and try and get it stable at 205+

[BoringName]

Inspired by the small amount of progress I made re-soldering a bridging wire, I figured I would try that some more. I re-soldered all the bridging wires in the VS/VA section of the board. Including the trace to the VS_DET pin on chip MCV14A. That trace was mainly bridging wires but also included half a dozen SMD resistors.

I can now get up to 215 volts now before it clicks which is well above the 205 required by the TV.

Power supply repaired. (well, repaired enough for me)

But the increased voltage didn't fix the red shadows on the TV, I tried upping the voltage to 210 and it made zero difference. I tried re-seating all the ribbon cables to the logic board and now the shadows are yellow instead of red..... I'm sure it's just an issue with the ribbon cables so I will keep working on that.

Anyway, the issue this thread was started for is fixed.

I couldn't find a schematic for BN44-00446A anywhere but the board seems to be a mash up for the following 2 schematics that are readily available - BN44-00333A and BN44-00508B

The values for the components don't match but they helped me work out the layout of the board. Especially with the microcontrollers.

In the end this is effectively all I did.
- Replaced all the smaller electrolytic capacitors - The board used to stay on for a little while if it hadn't been turned on for a day, after changing these it just clicked all the time.
- Replaced 1 small blue MKT capacitor near the VS pot - I don't think this really did anything.
- I didn't replace any of the 450 volt 68uF or 250 volt 180uF capacitors. - They tested ok with an ESR meter and most of the smaller caps I replaced did too.
- Refreshed the solder on all the transformers - I did this very early on and it didn't do anything.
- Refreshed the solder on all the bridging wires in the VA/VS section and along the trace to the VS_DET pin on MCV14A - This seemed to be what fixed it.

I don't know why that fixed it. One thing I did notice though is after refreshing the solder in the VS section, the VS Pot was more sensitive. Smaller turns increased the voltage a lot more.

Hopefully my suffering helps someone else in future. Although this TV is that old there probably isn't many left.

I'm just fixing it for a mate to put in his shed  He is going to owe me a few beers if I can sort out this red/yellow shadow problem!

edit: I cleaned the ribbon cable connectors between the main board and the logic board and the shadows are now black :-D

[BoringName]

I can now get up to 215 volts now before it clicks which is well above the 205 required by the TV.

Just a final update on this.

After a week or two my mate told me the TV started clicking again. It took me a while to get around to looking at it.

I started off turning the VS voltage down to minimum and it powered up ok. I noticed that adjusting the voltage would result in it clicking at different voltages the more I moved it around. So I gave the trimpot a clean and I was able to get it up to 222 volts (highest it would go) and it stayed on for a quite a while. Eventually it started clicking again but it highlighted the trimpot being the issue. Looking back, every fix I tried involved adjusting the voltage levels to test it. So any improvements I got were from manipulating the trimpot and not whatever else I did.

The strange part is it appeared to be working perfectly fine, it measured the correct resistances and turning the trimpot adjusted the voltage smoothly. There were no skips or jumps. I had previously refreshed the solder on it.

So I replaced the trimpot with a new one and now everything is working perfectly. Just bizarre really. I guess it was doing something I couldn't see on the multi-meter but it was enough for the power supply to go into error mode and reset.

[BoringName]

And it's died again. Lasted a week with a lot of use.

Guess that explains why the trimpot tested just fine.

Only thing I can think of now is a component in the area of the trimpot is intermittently failing and whenever I put heat through that area with the soldering iron it's bringing it back to life for a little while.

It's mainly just a bunch of SMD resistors. I think I'll just leave it off for a couple of days and then test all the components in the area.