View Single Post
  #2   Report Post  
Posted to sci.electronics.repair
[email protected] jurb6006@gmail.com is offline
external usenet poster
 
Posts: 1,630
Default LG 50PJ550-UD No Picture

When I run into something like that I generally scope an output from the Y sustain, see if it is putting out the waveform. There is a typical type waveform on all of them because of the way it works. you are not looking at the actual input to the panel, you are looking at the output to the COFs on the ribbon cables.

If the waveform is ther eI go to the Z side. Generally all those connections are really jkust one. they use that many because of the capacitance and all causing the great peak curent requirement to drive these things. If you see a waveform there, then unplug the panel from the Z sus and see if the board (not the panel) still shows a waveform.

Do someting to invoke an on screen menu, ONE of these waveforms should change slightly. you are at the frame rate, usually near 60 Hz so you are looking at top to bottom. One or the other waveform almost always changes with that type of input.

This simply confirms output from the sustain boards. They are controlled by the logic board, which also can cause that symptom. Another thing to note is that the Z and Y waveforms must be different. Usually, like if the Z board is dead, there will be feedthrough from the panel makng it appear there is output from the Y or Z, whichever might be dead. This is why you unplug the Z side. It isolates it and the Z sustain is usually ALOT easier to disconnect in most sets.

If indeed there are good waveforms, at least on if them having a significant negative portion of the waveform, go to the logic board. this is almost the same thing as a Tcon in an LCD. you will note the LVDS cable to it. Put the scope on AC coupling and the probe on 1X. (this is one of the FEW times I recomend not staying in 10X) At the frame rate on the scope, probe the LVDS. Sooner or later you come across sometning that looks like a vertical blanking pulse. It kinda is, but is for a very different purpose in these things.

You are not going to see a very good waveform here because it is a differential thing to reject noise. That's what the D in LVDS means, NOT digital. While you might expect three pairs going into the logic board, lke for red, green and blue, usually there are more. you have to find the pairs and while watching the scope, AGAIN do something to invoke an on screen display. Somehwere you should see a change, like a blip on the waveform. That would be the OSD.

If you DO see that, suspect the logic board. If you do NOT see that then suspect the main.

I said SUSPECT. I would search the net for case histories to just see what is more likely. Everything is inderdependent sometimes and then you might have to just make an educated guess. Part of the problem is the skimpy info they give.

Good luck wit it. To be really sure, you would need all the factory info, which would include their tech training, and even THEN halfg the time you cannot be sure.

There is one thing though, darken the room and turn it on from cold. Most of them will put out a little flash of light from the screen. Not bright but noticable. If you don't see that, suspect a sustain board a bit more. It's not a really bright flash, you might only see it with the room dark but most of them do it.

I looked in one database but it had notning that fitt what you got. One power supply problem for that model. I used to have another database but I stopped subscribing to it this year. Just Googling the model should get you something from fixya or something. It's not gospel, but if you see a thousand of the same cure for that symptom it is likely. Unless you designed the thing you are playing the numbers. Also, don't discout the possibility of multiple things wrong. The panel can blow the buffers, the buffers can blow the Ysus, the Ysus can blow the PS, the...you know.