Maker Pro
Maker Pro

Samsung TV PSB multiple symptoms/failures at the standby SMPS - own investigation hitting dead end

Dear reader/electronics' enthousiast,

My own investigation (as a layman) of the power supply board (BN44-00269A) of my Samsung TV is reaching the point where either I ask for advice or, I stubbornly start to randomly desolder parts to hope to find the root cause of what is malfunctioning.

Please, I have attached a file 'power supply fault description', together with some pictures that can be used as a reference. Feel free to ask for more pictures or other info if this may help.

What follows is a description of the sequence of events, from the first time I noticed the issue till my last attempts at grasping the problem today:

1) About a month ago, I noticed that my TV wouldn't turn on anymore. Unplugging from the mains, waiting and plugging in again did not change a thing. Also, I noticed that there was no 'stand-by' light, as most troubleshooting guides want you to look for.

2) My first attempt at troubleshooting as someone who knows nothing about TV power supply (nor electronics, apart from my classes during high school), got me to the point where some measuring with a multimeter told me that the fuse FB801S was blown. I could not see any signs of damage on any of the components of the PSB.

3) Thinking, instead of buying a new TV, that a quick fuse replacement first may do the trick, I bought a replacement fuse (same rating). This one was a glass fuse (and not ceramic as the original). The vendor told me that glass could replace ceramic without any issue and I think I may have to doubt this info now.

4) After installing this replacement fuse, I plugged the PSB (fully inserted and connected to all other parts of the tv) into the mains and immediately I could hear a fuse popping and saw some sparks flying. My initial thought was that those sparks came from the fuse itself and so I thought the burning marks on components I will describe below, were just a result of the glass fuse popping.

5) We are now several weeks later and after doing a lot of reading and measuring, I have come to the following results and (preliminary) conclusions:

Symptoms:

-Fuse'FB801S' blown
-Signs of damage due to arcing between one of the drain pins and current sense pin of the chip 'ICB801'
-Resistor 'RB806' clearly damaged and indicates 7,5kOhms (is rated at 0,5Ohms...)
-Diode 'DB804' is shorted

-Unable to find a shorted MOSFET nor diode at the PFC

Hypothesis:

-A serious overvoltage condition existed after the PFC (at the primary coil of the standby SMPS transformer) which got safely 'isolated' by the original ceramic fuse that blew.

-My replacing with a glass fuse and reconnecting to the mains, subjected the glass fuse to the same overvoltage condition. The fuse blew, but being 'glass', remained conductive for a little longer than ceramic would have.

-This resulted in serious voltage buildup near the drain and current sense pin of the standby IC, which caused an 'arcing' and a short towards positive ground. This rush of current towards ground burnt up the 'RB806' resistor which was the only component in its way.

-At the same time, the overvoltage at the primary coil of the standby transformer caused an overvoltage condition at the diodes of the secondary and auxiliary windings of the transformer. The diode 'DB804', supplying power to the Vcc pin of the standby IC, could not handle this and shorted.

I need your advise:

-Either my hypothesis is wrong and in that case, I certainly need some help because I have the feeling that more reading and learning will not do the trick for me.

-If my hypothesis is (partially) correct, I have no clue why such an overvoltage condition could exist in the first place. Is it because the FETs or the diode of the PFC are possibly faulty but I am unable to detect this? Is it because the IC of the PFC is working erroneously, thereby creating a too big a voltage at the standby SMPS? Or..?

I know as a layman, it is probably safer and wiser to just throw away this PSB and buy a new one, but I have become too interested in this PSB and its faults and backing off before at least listening to your advise does not seem to be right.

Many thanks for any input or advise you wish to share,

Wouter
 

Attachments

  • samsung bn44-00269a power supply fault description.pdf
    782.4 KB · Views: 3
  • BN44-00269A.JPG
    BN44-00269A.JPG
    291.7 KB · Views: 7
  • ICB801 - drain and CS pins damaged.JPG
    ICB801 - drain and CS pins damaged.JPG
    171 KB · Views: 7
  • RB806 damaged.JPG
    RB806 damaged.JPG
    233.6 KB · Views: 7
  • Standby circuit.JPG
    Standby circuit.JPG
    289 KB · Views: 8
I'm by no means an expert when it comes to repairing SMPS, but I pulled up the datasheet for that IC and gave it a quick read. This part jumps out at me:

CS (Current Sense)
The Current Sense pin senses the voltage developed
on the series resistor inserted in the source of the
integrated CoolMOS® If voltage in CS pin reaches the
internal threshold of the Current Limit Comparator, the
Driver output is immediately switched off. Furthermore
the current information is provided for the PWM-
Comparator to realize the Current Mode.

If I had to guess, I'd say the internal reference voltage went high and caused the IC to fail while trying to reach that threshold. That would explain the current flow through RB806 as well as DB804 and RB805. I haven't looked in to what TB801S is or does, but it must have affected RB809 in some way. It doesn't really explain the arcing, but the Drain pins are connected directly to the Drain of the CoolMOS®. I wonder if pins 3 and 4 were shorted internally and the current draw of the already established link caused it to jump the gap.

Looks like you've been pretty thorough hunting for bad parts. I bet if you replace the ones you've found, it'll come back to life. Again, though, I rarely work on SMPS equipment, so take what I say with a grain of salt
 
Thank you for shining your light of knowledge and experience on this matter! Even though last couple of days, my feeling has drawn me more and more towards the PFC circuit as the likely culprit and more specifically its FETs, I'll take your inputs into account and maybe they'll lead me to a final solution :).

If I am not mistaken, TB801S is the transformer that eventually provides standby power for the television (on its right side of the diagram). However, through some auxiliary windings of this transformer, also the power supply for the chip 'ICB801' which is regulating that same transformer (pardon my wording, since 'regulating' is probably not the right term to use) and the power supply to the 'ICP801', the chip 'regulating' the PFC circuit, are provided.

I cannot think of anything else being able to build up such an overvoltage, except the transformer of the PFC circuit. Its MOSFETs must have been 'closed'/conducting for way too long so that energy was able to continue to build up in the transformer's core...finally to release all of this energy through fuse 'FB801S' and transformer 'TB801S', damaging all of the affected parts. I hope though the 'ICB801' has not been hurt in the process but you may be right when saying that damage may have been inflicted on this IC as well...

Greets!
 
Would someone please be able to point out what the role is of these capacitors I boxed in yellow on the attached partial diagram?

More specifically, I am interested in how the 'CP817' (rated 450V, 22uF) 'interacts' with the stand-by circuit that follows, protected by the 'FB801S' fuse (rated 250V)?

Somehow, I am starting to fall back to one of my first suspects, being these capacitors. I do not watch tv on a regular basis, but it still strikes me as suspicious that the device had stopped working the first time I wanted to watch tv after some technicians of our network operator had come by to change meters for our natural gas supply. Nothing special, except I remember them saying they had used the big circuit breaker for the whole building at one time to switch off/on the power supply of the gas boilers of our central heating systems...Somehow, I am linking this together with a (worn out?) capacitor as a possible culprit. Could this make sense?

I am not yet in the possession of a capacitance meter but am willing to go and look for one to troubleshoot.
 

Attachments

  • Capacitors in PFC circuit.pdf
    812.9 KB · Views: 3

Harald Kapp

Moderator
Moderator
These smooth the rectified AC to a steady DC voltage (PFC_DC) which is then used by the switch mode power supply (smps) to generate the lower operating voltages.
CP817 is decoupled from PFC_DC by a diode- Thus the voltage across this capacitor will not drop when PFC_DC experiences a drop e.g. when the power MOSFET in the SMPS turns on. This decoupling serves to create a more stable supply for the circuit powered from this capacitor.

Somehow, I am linking this together with a (worn out?) capacitor as a possible culprit. Could this make sense?
Typically worn out electrolytic capacitors lose capacity, they do not go into short circuit.
I am not yet in the possession of a capacitance meter but am willing to go and look for one to troubleshoot.
Maybe less expensive to simply replace these capacitors?
On he other hand a simple component tester isn't that expensive. These may be not very precise, but the accuracy is sufficient to verify the capacitance of the electrolytic capacitors.

R806 is supposed to be 0.5 Ω, as it measures 7.5 k ist is definitely broken.
R805 is o.k. (6.8 Ω expected, 7 Ω measured.
D804 may be defect. However, I assume you measured the diode while it was still in the circuit (on the PCB). Therefore other components may make your readings invalid. Remove the diode from the circuit, at least one leg, then measure again to verify it's the diode that is faulty.
 
Thank you Harald,

Your explanation about the capacitor's role in the circuit is helpful.
Probably I am missing something obvious, but any idea why a 250V fuse (FB801S) is placed in a circuit where the capacitor is rated at 450V and where the PFC is probably a 'boost' PFC, boosting the voltage up to well above 300V? How does this work out (so that the voltage at the fuse stays below 250V)?

Maybe less expensive to simply replace these capacitors?

This also crossed my mind as a much more budget-friendly option :). I am right now in the learning mindset and hoping to find the exact root cause of what caused all this havoc and I am afraid I will lose information by replacing components I am not sure of whether they are faulty or not. I'll maybe see if one of the neighbours has one but probably the cheaper meter will do, as you say.


Remove the diode from the circuit, at least one leg, then measure again to verify it's the diode that is faulty.

Thanks again
 

Harald Kapp

Moderator
Moderator
250 V is the AC rating (RMS) of the fuse. It can withstand a peak of 250 V × 1.4142 = 350 V.

I am right now in the learning mindset and hoping to find the exact root cause of what caused all this havoc
Thumbs up. A laudable attitude. Many inexpensive multimeters come with a capacitance range and the simple component testers I linked to are inexpensive alternatives.
 
250 V is the AC rating (RMS) of the fuse. It can withstand a peak of 250 V × 1.4142 = 350 V.

Great! I've come across that square root of 2 already a couple of times when AC is being rectified to DC, this makes sense :).

I am curious how this investigation will develop over time but all your inputs have definitely helped me already a bunch.

Greets
 
Update: the TV is up and running again.

After the last message I came up with a final hypothesis and did repairs accordingly:

1) Hypothesis:

-The 'stand-by IC', 'ICB801', due to age, failed internally: possibly, the circuitry that regulates (the frequency of) the built-in MOSFET of the IC failed, causing the MOSFET to 'power close' continuously, connecting 'drain' to 'current sense' anytime the IC is powered.
-In turn, this leads to the drain pins continuously being connected to the current sense pin whenever the IC is powered, whereas normally, they would only intermittently (with high frequency) be allowed to 'drain' via the current sense pin.
-This condition was safely isolated the first time by the ceramic fuse.
-The second time however, the glass fuse blew but remained conductive, just long enough for current to continue to drain through the drain and current sense pins. In the meantime, this increasing flow of current through the primary winding of the stand-by transformer 'TB801S' had been charging the core of the transformer with energy.
-When eventually the glass fuse stopped being conductive, energy had been building up so much in the core of the transformer, that any sudden interruption of the circuit would require a spark to release that energy -and the faster the circuit would be interrupted, the higher volts would be generated. The most convenient place for this spark to form, would be between one of the drain pins and the current sense pin.
-The resistor RB806 got damaged in the process.

2) Repairs:
-ICB801 replaced (both 'Vcc' and 'S/S' pins were shorted internally to 'ground' pin on the faulty IC)
-RB806 replaced
-Ceramic fuse 'FB801S' installed

note: the diode 'DB804' indeed measured ok outside of the circuit (thanks Harald Kapp).

I hope this thread may someday help someone in troubleshooting his or her bad TV power supply to some extend.
 
Top