Samsung UA43NU7020KXXU repair

[WaveyDipole]

I recently won a Samsung 43in TV panel in an auction 15GPB but it seems that my bargain was short lived. When I plugged it in after getting it home, it came on, the Samsung logo appeared and it went to the menu. I connected my laptop and the desktop appeared. Everything looking good so I ordered the feet and a cheap remote for it. Total cost all in for the TV, 50GBP.

When the feet arrived and after I attached them, I powered it up again and this time, but this time there was no display, no backlight discernible. Only the standby light comes on. Pressing the power butting makes the light flicker momentarily but it stayed red. Today I received the remote and when I press buttons the red led flickers showing its receiving the signal from the remote, but still the display does not come on and it doesn't seem to be coming out of standby. Obviously disappointing as I can't return it to the auction house, but on the other hand every reason to open it up and have a look.

The first obvious thing were two bulging 470uf, 25V capacitors which will have to be replaced. I am not convinced that I am seeing anything under a flashlight except smudges but very hard to tell. The TV does not seem to have a separate T-con board or power and main boards. Everything seems to be integrated on one large board. It has two large ribbon connectors at the bottom. I tried disconnecting these one at a time, but no change.

I don't want to spend any more on this than necessary so I am thinking about how I can test the backlight without buying one of those LED testers? There are two pairs of wires that go to the connector so I assume there most be two circuits? When I check the output from the board it shows around 180V but starts collapsing quite rapidly over a few seconds as soon as I probe with the DMM. Not quite sure whether it should doing that? The 4.7uf, 500V capacitors adjacent to the LED test points do not bulge and physically look OK.

I did see a video online where a guy used 9V PP3 batteries in series, but without a current limiting resistor, so I am not sure that is safe, hence my question about how to go about testing the backlight.

[Kean]

Start with replacing the bulging caps.  It it possible they are powering some circuitry that controls the backlight driver, so it doesn't stay on.

[WaveyDipole]

That was indeed the initial plan for today and have just done exactly that. I don't know what kind of solder was used in the assembly of this board, but I could not un-solder the old caps. The solder would just not melt. My Weller 80W TCP soldering iron usually has no problem with either leaded or unleaded solders. In the end, I had to snip them off the top. As they were so close to the board there was hardly any room to get in and consequently next to no lead left protruding. There was a bit of lead protruding on the underside and I did try to push 3them through but it was a no go. Even with adding 60/40 solder to the pad they simply wouldn't budge. Very strange indeed.

There is no space underneath the board so they had to go on top. Using plenty of flux I managed to solder the leads of the new caps on to the pads on the top and mount them at right angles to the board. I will probably need to to secure them with some hot melt.

I just did a quick test and the TV came on so it looks like you were correct and that was the issue. Since these were low voltage, they were probably related to the control circuitry of the LED backlight driver. The two 500V caps on the high voltage side look fine. Just debating whether to replace those as well.

I would be interested to understand what is going on with that solder though? Obviously it must me some kind of higher temp industrial alloy?

[Kean]

I've seen only a few higher voltage caps fail, so I wouldn't bother unless you see more issues.

I doubt any special solder was used, and Samsung probably don't use really crap solder alloys either.  If there is just the one PCB, then I assume it is 2 or more layers with plated holes, which can require a lot more heat especially with pads connected to power planes and parts with a lot of thermal mass.  An 80W iron should be up to the task though.

I always prefer when there is a separate power PCB, as those are often cost reduced and so are single layer phenolic with lots of THT, so can be much easier to work on when replacing caps.  Cutting away parts you plan to throw away is always a helpful technique to assist with desoldering.

[WaveyDipole]

I haven't seen many high voltage caps fail either except perhaps very old wax paper types in vintage equipment.

The solder puzzles me. I might expect the leg soldered to the ground plane to be a little difficult to deal with due to the heat dissipation, but the other one usually comes out quite easily. The iron is more than capable, especially if I add a little solder to the joint to ensure good thermal contact. I applied as much heat as I dare but I didn't want to scorch or damage the PCB. The remnants of component legs are usually easy to remove as heating the protruding leftover piece is usually enough to loosen it and then pull out with tweezers, but not in this case. I couldn't even slightly melt or soften the solder.

I agree that the separate boards are usually single layer and much easier to work with.