Samsung SMPS BN44-00339A

[Loretto]

Hi all,
this unit comes from a samsumg LCD TV LE32C530F1W which stopped working after making a strong noise like something was exploded inside of it.
I opened it and found that fuse FP801S was blown, capacitor CP814 was completely disintegrated and MOSFET QP801S was shorted.
I replaced all of them and instead of the fuse I connected a 100 Watt bulb. Then, without connecting it to other TV boards, I switched it on, but the bulb begun to blink.
I thought that maybe the bulb was somehow interfering with the normal operation of the power supply, so I replaced it with a fuse, ... and boom !!! I burnt again the MOSFET QP801S I just replaced.
I had another one, so I replaced it again and did some more tests.

I checked all the electrolytic capacitors and noone seemed to be bulged or leaking: also the measure of the capacitance was good (don't know about ESR).
I checked the photocouplers and they seemed to be good, at least for the section of the infrared led: do not know about the phototransistor on the other side.

I found out that connecting PS_ON to cold ground, I had a stable 5 Volts on A5V (stand-by voltage) and the bulb stopped blinking.
In this situation, I measured the voltage on the legs of the main filter capacitor CP809 and I had a stable 324 Volts (our mains in Italy is 200 Volts).
But if I connected PS_ON to A5V, all the output voltages of the power supply became unstable and the voltage on the legs of CP809 was about 300 Volts, which was not good at all.

So I ordered a new ICP801 (FAN7530M) and a new ICM801 (FSFR1800US) and replaced both of them.

At the beginning, with PS_ON connected to A5V, everything seemed to be working: I even measured stable 13 Volts and 24 Volts and the voltage on the legs of CP809 reached about 415 Volts.
But then I begun to hear strange noises and didn't understand where they were coming from.
I rapidly switched everything off and noticed that CP809 (120 uF / 450 Volts) was warm and now seemed bulged, on both sides (top and bottom).

I replaced it temporarily with something similar (150 uF / 400 Volts) but the story is the same, if not worse: the bulb is steady on and all the output voltages are unstable.
Now, even the voltage on the legs of CP809 is fluctuating, from 100 to 300 Volts ... while, if I connect PS_ON to cold ground, everything works like before.

I replaced resistor RP822: nothing changed.

I measured CM801S without disconnecting it from the board and it seems good.

Resistor RM803 seems good too: anyway, when PS_ON is connected to A5V, as soon as I plug the power supply, the voltage across it is e few millivolts (less then 10) and then goes to 0. Is it normal ?

I really don't know what else to do: I played all my cards. Any suggestion ?

Thanks.

[Loretto]

I know you guys couldn't wait anymore to see how things ended up with this power supply, could you ?

So finally, here I am for the epilogue ...

To sum up, it was necessary to replace FP801S, QP801S twice (as you know, something went wrong while troubleshooting), CP814 (all blown) and ICP801 twice: the first time just to check if it was the reason of the failure, the second time because it measured only 6 Ohm between Pin 1 (INV) and Pin 6 (hot ground).

Then, during the tests, CP809 began to bulge, so I replaced it as well ...

At the end, when I had already lost all my hopes and didn't know what else to do, I replaced even ICM801, but perhaps this was not necessary ...

But the real fix consisted in replacing all the SMD resistors from RP813 to RP817 as they were completely out of value. Furthermore, even though I disconnected the side of RP817 that connects to RP818, CP804 and Pin 4 of ICP801 (there was a jumper which I could easily desolder from the board), the total value of the 5 resistors was different depending on the polarity of the multimeter: in one direction it seemed a capacitor was charging ... Absurd !!!

After removing all the 5 resistors, the tracks on which they were soldered, although apparently separated/isolated from everything else, showed a resistance towards other components around, between 200 KOhm and 1 MOhm (?!?) so, I decided to remove those pads and mount the new SMD resistors in a different position ...

Unfortunately, at that moment, I didn't know yet that finally the power supply was working, because actually, the 100 Watt bulb connected in place of the fuse wasn't staying off as I thought it should, but it was lighting up shortly, then was going off shortly again, to turn then back on forever.

By chance I noticed that if I connected the digital multimeter (ampere mode) in series with RP822 (the PFC block current sense resistor), the light bulb, after turning on and off, lighted up dimly and then turned off permanently ... The multimeter was reading between 10 mA and 20 mA ...

So, I removed the multimeter and replaced RP822 with a 2 Ohm resistor: the second flash of the light bulb was at full brightness but then it turned off forever ...

Increasing RP822 to 5 Ohm, the second flash of the light bulb was dimmer than before, and then it turned off forever ...

Remove, or not remove (the bulb): that was the question !!!

If you still remember, the first time I replaced CP814 and QP801S, the bulb begun to blink: hoping it was somehow interfering with the normal operation of the power supply, I replaced it with a fuse and ... damn, I burnt QP801S again !!!

Now, you have to know that when I lost all my hopes to fix this power supply (more than once, indeed), I gave in and purchased a working second hand one ...

As soon as I received it, after checking that it was working also with the PS_ON connected to A5V, I replaced its fuse with the 100 Watt light bulb, which remained turned on just like with my BN44-00339A under service.

So fingers crossed, I decided to replace the light bulb with the fuse on my bad power supply as well and magically, even with the PS_ON connected to A5V, nothing blew, the 395 Volts appeared on CP809 (*) and all the output voltages were correct and stable.

(*) 318 Volts with PS_ON connected to cold ground, for the record ...

Closing with, the lesson is: don't blindly trust the bulb since sometimes it can stay on even if everything is fine, especially if the power supply has a PFC block. In that case try to increase the current sense resistor and if the bulb turns off, restore the old resistor, say a prayer and try removing the bulb ...

Hope this can be helpful to somebody else.

Cheers.