Samsung UH50EH5300FXZA, help with a transformer?

[DXseekerMO]

About 2 months ago I got a Samsung UH50EH5300FXZA that was a lightning strike victim. I'm finally getting around to looking at it.

The AC fuse was blown. I checked the board for a direct short and didn't find one so I piggybacked a fuse on top of the blown one, plugged it in, saw sparks inside this transformer and the fuse blew.

This transformer reads shorted corner to corner. There's a number on it, but a Go-ogle search comes up with nothing. I was wondering if anyone can help me make heads or tails out of this number so I can source a replacement?

I'd take one off another board if I had a way to spec. this one out and match it.

[budm]

Re: Samsung UH50EH5300FXZA, help with a transformer?

1) That is an inductor for the active PFC Voltage booster circuit.
2) You are reading the DCR (DC Resistance) of the inductor which will show very low resistance, you can only test for open circuit with simple Ohm meter, you need ringer tester (for shorted turn test), inductor tester to read inductance.
3) Since it blow the fuse then you need to test the bridge rectifer, diode, MOSFET (especially the one for the PFC circuit). So start checking the semiconductor devices for low resistance first.
4) What is the P/N of this power supply board? BN44-?????
5) Need to upload good clear straight shot pictures of the whole top and bottom side of the board, the pictures must be clear enough to read the printed texts and able to see the traces.

[DXseekerMO]

Re: Samsung UH50EH5300FXZA, help with a transformer?

Board number BN44-00668A, here's the pix

[DXseekerMO]

Re: Samsung UH50EH5300FXZA, help with a transformer?

That inductor had sparks in it when I plugged it in, of this I am 100% sure. The fuse also looks like it had identical looking sparks inside after plug in. Does that sound like a shorted turn in there?

[budm]

Re: Samsung UH50EH5300FXZA, help with a transformer?

Did you test the MOSFETS, the Diodes as suggested first?
The only way to get too much current flowing is the shorted PFC MOSFET.

[DXseekerMO]

Re: Samsung UH50EH5300FXZA, help with a transformer?

Not yet, I've been listening to news about the KOZK TV tower collapse in Fordland MO on TV. That tower is 11mi. from my house. I heard it collapse.

[DXseekerMO]

Re: Samsung UH50EH5300FXZA, help with a transformer?

Did you test the MOSFETS, the Diodes as suggested first?
The only way to get too much current flowing is the shorted PFC MOSFET.

Holy old thread revival Batman!

So I finally decided to break this set back out and give it another go. Before I put it up I installed MOSFETS QM801CS and QM802CS. End result, fuse doesn't blow but TV will not turn on and voltages at A5V and PS_on were very unstable.

Here are the voltages from this go-around, the first row of voltages is the PSU plugged in to the main board and the second is unplugged from the main board.

1 B13V…………………0…………………… .6 to 4.2V unstable
2 PDIM………………..0……………………. unknown/0v
3 B13V…………………0…………………… .6 to 4.2V unstable
4 B13V…………………0…………………… unknown/0v
5 Vamp………………..0………………….. .6 to 4.2V unstable
6 BLU_on…………….0…………………... unknown/0v
7 Vamp………………..0………………….. .6 to 4.2V/unstable
8, 9, 10 (gnd)……….Continuity present
11 B5V…………………0…………………… unknown/0v
12 A5V……1.8 to 4.4 unstable…….. unknown/0v
13 B5V…………………0………………….. unknown/0v
14 PS_on… .1 to 2.3 unstable**… unknown/0v

** Mainly stays low with quick flashes to upper voltage listed

The reason many of the even numbered pins are “unknown/0v” is because by the time I got around to checking them power readings had mostly disappeared from the terminals, including the ones I was fast enough to get readings on.

Furthermore I'm getting unstable voltages on both the hot and cold sides of the PSU, but the rectifier diode output is a solid 165V and so is the voltage at the main cap and drain pin of MOSFETS QM801CS, about 24V solid on the drain pin of QM802CS

I tried changing transistors QM807 and 808 which are EPITAXIAL PLANAR PNP TRANSISTOR and is described as a VOLTAGE REGULATOR, RELAY, LAMP DRIVER, INDUSTRIAL USE because the voltage at the emitter of QM808 was an unstable 1.8 to 4.4v. The base pin voltage was 10.1v solid and there was no voltage on the collector. I changed them because they are noted as commonly failing components. After this I re-flowed out any bad solder joints.

Unfortunately the problem still exists and appears to be getting worse, in the sense that the voltages are now almost all the way gone. So does anyone have an idea what may have crapped out here? I can't seem to find a schematic for this board, BN44-00668A. Does anyone happen to have a schematic laying around for this?

Good clear expandable pictures included!

[Unspun01]

Re: Samsung UH50EH5300FXZA, help with a transformer?

Check the 16-pin SOIC (ICM801S) located on bottom of board near QM801CS and QM802CS. It is an LED driver IC.

Your first photo looks like there is a scorch mark on it, and I can't tell anything from the second photo.

I had a UN46EH5000F TV with a similar power supply design that would turn on, but the backlight was strobing about once a second. This IC was screwed on my TV, with evidence of a round hole burnt/melted in the plastic.

The IC was shot, but even after replacing the power supply, about 20 out of 44 backlight LEDs were burned out as well...


Incidentally, there is a repair kit from shopjimmy for less than $10 for this board with that IC included and a whole bunch of other parts.

https://www.shopjimmy.com/samsung-bn...repair-kit.htm

[DXseekerMO]

Re: Samsung UH50EH5300FXZA, help with a transformer?

Here's a close-up of ICM801S, can't tell if the last 3 numbers are 420 or 4Z0. Either way I'm having a hard time finding a datasheet for it. I'm pretty sure the dot at the lower left indicates pin1. Not sure if I see any scorch marks, but the other pic wasn't as detailed so maybe this is a bit more conclusive?

[R_J]

Re: Samsung UH50EH5300FXZA, help with a transformer?

Fe5760

[budm]

Re: Samsung UH50EH5300FXZA, help with a transformer?

So right now of you run the power supply board by itself, is the A5V steady?
Your report does not make any sense:
second is unplugged from the main board.
12 A5V……1.8 to 4.4 unstable…….. unknown/0v
BTW, that 16-IC is the SMPS IC, the LED driver circuit (boost converter circuits are in the cold side, not in the hot side).
The power transformer is being driven otherwise you will not have any Voltage on the cold side, but it is not stable right now.
BTW, I still cannot see the location of the so called sparks on the board, and that board also has many through hole rivets which are known for cold solder joints.

"I installed MOSFETS QM801CS and QM802CS" were any of them tested as shorted? To blow 5A main fuse, something was drawing lots of current.

[DXseekerMO]

Re: Samsung UH50EH5300FXZA, help with a transformer?

Fe5760

Thank you R_J! I was trying the other number lol. Datasheet still hard to find

[R_J]

Re: Samsung UH50EH5300FXZA, help with a transformer?

If QM801CS and QM802CS were shorted, I would check the resistors/diodes on there gates, when those fets short they can blow the gate driver parts also.
I would also check that white square cap CM803s they like to go open (usually they bulge)
It does look like it is trying to work if you have some voltage on the A5V

[budm]

Re: Samsung UH50EH5300FXZA, help with a transformer?

"I would also check that white square cap CM803s they like to go open (usually they bulge)"
+1
The GATE drive outputs can also be damaged when MOSFET shorted out.

[DXseekerMO]

Re: Samsung UH50EH5300FXZA, help with a transformer?

So right now of you run the power supply board by itself, is the A5V steady?
Your report does not make any sense:
second is unplugged from the main board.
12 A5V……1.8 to 4.4 unstable…….. unknown/0v
BTW, that 16-IC is the SMPS IC, the LED driver circuit (boost converter circuits are in the cold side, not in the hot side).
The power transformer is being driven otherwise you will not have any Voltage on the cold side, but it is not stable right now.
BTW, I still cannot see the location of the so called sparks on the board, and that board also has many through hole rivets which are known for cold solder joints.

"I installed MOSFETS QM801CS and QM802CS" were any of them tested as shorted? To blow 5A main fuse, something was drawing lots of current.

At this time A5V is low and not steady, from .9v up to 3.3v and swings wildly.

I explained in my update that the voltages on the first row are with CNM803 hooked up to the main board and the second row is with connector to main board at CNM803 unplugged. By the time I got around to checking for voltage at the even numbered pins for CNM803 all the voltage had disappeared at the connector.

The sparks from the board you are referring to were seen during the initial diagnostic. I piggybacked a fuse on to the blown main fuse and plugged it in. The sparks were seen inside of transformer LP802CS until the main fuse popped again.

MOSFET QM802CS was short. If I remember right QM801CS was OK but I replaced it anyhow.

[R_J]

Re: Samsung UH50EH5300FXZA, help with a transformer?

There is a resistor, RP821 off the (-) of the bridge (between LP802CS coil and JP811) check that it is ok

[DXseekerMO]

Re: Samsung UH50EH5300FXZA, help with a transformer?

If QM801CS and QM802CS were shorted, I would check the resistors/diodes on there gates, when those fets short they can blow the gate driver parts also.
I would also check that white square cap CM803s they like to go open (usually they bulge)
It does look like it is trying to work if you have some voltage on the A5V

"I would also check that white square cap CM803s they like to go open (usually they bulge)"
+1
The GATE drive outputs can also be damaged when MOSFET shorted out.

CM803 has about 16Kohms in one direction and open in the other. I didn't see a bulge on the case anywhere.

All items off the gates of QM801CS and QM802CS appear to be of the appropriate resistance and not shorted, save for resistance across CM810 which seems low at 101ohms?

[DXseekerMO]

Re: Samsung UH50EH5300FXZA, help with a transformer?

There is a resistor, RP821 off the (-) of the bridge (between LP802CS coil and JP811) check that it is ok

Indicates zero resistance on ohms scale. Looks like this one has to come out of circuit for an accurate assessment?

[R_J]

Re: Samsung UH50EH5300FXZA, help with a transformer?

No, its likely ok, it should be close to 0Ω

[DXseekerMO]

Re: Samsung UH50EH5300FXZA, help with a transformer?

No, its likely ok, it should be close to 0Ω

Cool.