Samsung LN40A650A1F, no backlight

[CraigParton]

Hello everyone! I'm a little new to TV repair and I was looking for a little help with this Samsung LCD TV. It will power up (red LED goes out) and you hear the startup sound as well as sound from the HDMI input I had connected to it. But the LCD screen stays dark. I opened it and noticed a blown 3.15A fuse on the power supply board (BN44-00198A). It wasn't the main fuse (obviously) but after looking over the board a bit I saw it protected the control circuitry for the high voltage transformer that powers the inverter board.

After a little testing I found two MOSFET transistors (FQPF9N50C) that were shorted source-gate-drain. I replaced these and the fuse and powered the TV up. Fuse stayed intact but still no backlight.

Now to show my newbiness. I'm an automotive tech with 13 years of experience troubleshooting and testing low voltage DC circuits. I have some understanding of how electronics work and I have a DSO and know how to use it. But I'm not exactly sure what I'm looking for. I did test the power source for the fuse that was blown. It comes directly off the bridge rectifier and the two 450V 2200uF filtering caps. I was reading close to 400V but it seemed to have a lot of AC riding on top of it. Again this may be normal I'm not sure.

I diagrammed the main circuit for the two mosfets (minus the small SMD control circuitry) but I'm still a little unsure what they do. From the diagram and what I've read it looks like they convert the DC into a high frequency AC for the transformer. There's also a small SMD 30K resistor and a zener across the gate and source. Any information on how this works or how to test it would be much appreciated. Unfortunately I don't have a high voltage probe to test the transformer output directly.

Thanks in advance!

[budm]

Re: Samsung LN40A650A1F, no backlight

Please post good clear pictures both sides of the boards you are trying to fix using "attachment" in the advance posting.
This will help us a lot to guide you, thanks.

[CraigParton]

Re: Samsung LN40A650A1F, no backlight

Here are the pics. Sorry I should have done this in the first post...

[budm]

Re: Samsung LN40A650A1F, no backlight

OK, when the that high voltage MOSFET's shorted out like that, more likely it will take out the Gate resistor and Gate draining diode (there should be a low Ohm resistor in parallel with the diode which will be connected to the Gate of the MOSFET) for each of the MOSFET and it more likely taken out the SMPS IC output sections that drive the MOSFET also. If you travs out the gate conections, it should be connected to that 20-pin surface mounted IC on the bottom of the board.

[CraigParton]

Re: Samsung LN40A650A1F, no backlight

Thanks for the info Budm! The marking on that small surface mount resistor connected to the gate of the mosfet is 3002. Would that be 30,000 ohms? Also is the small diode is marked WL. Is it a regular diode or a zener and do I need to test it live? Can I use the scope to check the gate of the MOSFET to see if it's being controlled? If so what should the waveform look like?

[budm]

Re: Samsung LN40A650A1F, no backlight

That does not sound right, it should be a lot lower than that, see example of how the MOSFET is being driven (single end setup), in you case it is a push-pull type setup so it needs two MOSFET's. See attached PDF.
I cannot find any SCH for this power supply yet.

[CraigParton]

Re: Samsung LN40A650A1F, no backlight

Thank you for the schematic I'm trying to decipher it and understand what's going on. I'll try to diagram the circuit tomorrow and trace it back to the ic. Any tips on identifying those small smd components?

[Scenic]

Re: Samsung LN40A650A1F, no backlight

Any tips on identifying those small smd components?

Go by the silkscreen markings

• C - capacitor
• D - diode
• F - fuse
• IC - integrated circuit / controller
• J(P) - jumper wire
• L - inductor
• LED - duh..
• Q - transistor
• R - resistor
• S(W) - switch
• TP - testpoint
• U - packaged integrated circuit (for example linear voltage regulators)
• Z(D) - zener diode

This wikipedia article has a bunch more of those..


edit: lookie what I found in my Samsung I/P-Board archive
It's pretty sparse-looking, but better than nothing I guess..

This I/P board looks one heck of a lot like the BN44-00199A I recapped for my LE40A536. Shopjimmy even lists the 199A as a substitute for the 198A
I've attached the 199A pdf too, just in case.

[CraigParton]

Re: Samsung LN40A650A1F, no backlight

Lemme just say, you are THE MAN! Thank you so much. That's going to help tremendously. I'll check it this afternoon and post results.

[budm]

Re: Samsung LN40A650A1F, no backlight

I look at the diagram, this is Power supply/Inverter combo board. The 30k and the ZENER are for the Gate protection.

[budm]

Re: Samsung LN40A650A1F, no backlight

By the way, Scenic, where did you manage to find all these SAMSUNG SCH from? Thanks.

[CraigParton]

Re: Samsung LN40A650A1F, no backlight

Now I see what blew the fuse. When the MOSFETs shorted it gave diodes DI221 & DI223 a straight path from ground to power. I did notice the solder trace on one of those had been hot. I forgot to mention that I tested the larger power diodes DI221, DI222, DI223, and DI224 with one leg unsoldered and they tested fine with the diode test funtion on my multimeter. I'm wondering if they're failing under power.

So from what I can see (with my limited knowledge) it appears that UI101 uses the AOUT and BOUT terminals to control the transformer TI202 through a series of mosfet and bipolar transistors. I'm assuming those are used to step up current. The transformer amplifies the voltage of the control signal (AC at this point?) into the MOSFETS I replaced (QI201 & QI202). These MOSFETS control the high voltage transformer TI201 that powers the inverter backlights.

So now for more newby questions. Will the transformer isolate the damage from the shorted mosfets to DI103, DI104, DI110, DI111, QI101, and QI202? I'd like to check these control signals with my scope, so if anyone could give me a little operational theory on what's going on here it would help me a lot. For example, what is the whole circuit trying to accomplish? I know the high voltage transformer is stepping up the voltage to light off the backlights, but is the control circuitry amplifing the frequency as well? Any idea which components might be damaged?

I have to leave for a couple hours but I'll test this stuff when I get back. Again, thank you budm and scenic for the excellent info.

[Scenic]

Re: Samsung LN40A650A1F, no backlight

@ budm: They're from all over the place. Most of them came from some russian website, hosting the files on a russian filelocker (sort of like Rapidshare etc.)

Was pretty tricky to get them, as the websites were completely in Cyrillic and I couldn't use google translator on the download page (told me it already was in english for some reason), so I had to figure out where to click to start the download by randomly clicking almost every link

I've re-uploaded a bunch of those to a filehoster, but they're censoring the search results, so the files don't show up when you search for them

[CraigParton]

Re: Samsung LN40A650A1F, no backlight

Scenic I found a service manual for the european version of this tv on a similar site. I had to manually dump the text into google translator to figure out how to register and download. No schematic for the power board in that manual, but I really don't need it now do I?

[Scenic]

Re: Samsung LN40A650A1F, no backlight

I just took a look and found the service manual for the A65x series (LE__A65_) LCD TV's (the LE indicating it's the european model)
It's probably the same thing you found, except I didn't have to register to download it.. instead I had to deal with a shitton of popups -.-
It's a ~30MB zip with the SM split up into several PDF's

[budm]

Re: Samsung LN40A650A1F, no backlight

One caution, when work on the Line (hot side) of the circuits, make sure to use Isolation transformer to power up the board and the whole thing should be powered through the GFCI outlet to protect yourself.

[CraigParton]

Re: Samsung LN40A650A1F, no backlight

One question on that isolation transformer: The scope I'm using is battery operated and doesn't plug into the wall at all. Will I still need that?

[PlainBill]

Re: Samsung LN40A650A1F, no backlight

One question on that isolation transformer: The scope I'm using is battery operated and doesn't plug into the wall at all. Will I still need that?

You can observe waveforms, but be very careful. I wouldn't even touch the scope when the TV power supply is getting AC voltage. The voltage on the scope is potentially as high as 400 volts.

PlainBill

[budm]

Re: Samsung LN40A650A1F, no backlight

Well the scope will become part of the Line side of the AC, if you touch the metal part of the scope, the AC current will flow through your body to ground, it will be hard not to accidentally touch the high side of the circuits, especially when you have have to attach the ground side of the scope probe to circuits. I deal with line voltage circuits everyday, you only get to it wrong once.
These are part of my main tools:
http://s807.photobucket.com/albums/yy352/budm/Tools/

[CraigParton]

Re: Samsung LN40A650A1F, no backlight

Sorry I should have specified which scope I was using. It's an automotive scope/scan tool. Its a Snapon MODIS. It's handheld 4 channel scope and the entire case is plastic. No exposed metal. I checked the voltage rating and it's rated at 400V AC/DC. The highest voltage I've seen on this supply (except the HVT) is 380V. Close but just inside the limit.

But I may not even need to test it live. Not that I could, most everything I need to test is on the bottom of the board. I checked between the gate and source of QI201 and the resistance is 9 ohms. There's a definite short somewhere. Gate-source resistance of QI202 is 30K (the gate protection resistor). These use identical control circuits so something is definitely shorted. It has to be either the MOSFET itself (QI201, but its new), the zener diode (ZI101) or QI101. RI130 tested 68 ohms and the resistance between the base of QI101 and ground was 235 ohms (parallel resistance of both 470 ohm resistors). My guess is ZI101 but I'll desolder and test these and post the results.