diodes are only protections, you can remove them, but to close the repair it is fundamental to replace them, so the next surge peak will not destroy panel, by your photos i see also 200 ohm and 51 ohm resistors..
Tested without any of the diode packs. The same issue remains.
Still 35ohm across a bunch of 0402 decoupling caps - NOTE! 35ohm across a bunch of caps on BOTH buffer boards even ribbons not connected! So buffer boards should not share supplylines at this point.
I was thinking about it because in recent Samsung the symptoms are similar in both cases, but if backlight is faulty it is not the panel disconnection that stops maiin board to sense BL fault, so i'm starting to think at tcon fault, if a semi short is present at a panel supply rail output, the removing of the load can prevent the pmic to feel the overload, or in other analisys since the tcon "software" is on the panel (flash) the tcon can't communicate with cpu the fault state (in truth are same chip but it doesn't matter)... but this is also difficult since you changed main board..
I could power the mainboard from external psu +13V but there is probably a BL OK / ERROR on PSU that I need to trick also
There is very few connections between PSU and MB.
Also in this set there is no separate T-CON, the panel is driven directly from MB.
I guess the signal "FAIL COUNT" is the one to check / trick ?
Should it be high or low when fault is detected ?
Summary:
1. When both panel ribbons are connected, this happens when plugging into AC:
- BL starts dimmed. After two seconds BL is full brightness for half a second, then
shuts off completely. After two seconds the operation goes through an endless loop
Also - standby led blinks in sequence of TWO blinks.
There is no start-up melody playing in audio.
2. When either of the panel ribbons are disconnected. BL behave correct and fades up
from dim to bright, also playing start-up melody in audio.
Standby led blinks in sequence of FOUR blinks.
None of the above tests give any picture information.
Fail count like common status signals in recent Samsung must be low to be inactive so pull it low..
Pulling it constantly low does not change anything...
So - when either of the two panel ribbons are disconnected the set starts-up correct with BL dimming up from dark to bright and plays start-up melody. So is there a "protection" line going through both the ribbons that does not reach MB is either of ribbons are disconnected?
I try to figure out which of the buffer boards is giving a problem...
Wait.. one user wrote the list of the psu tensions in a similar Samsung, since Fail count was 1,57v, it makes me think about a digital serial signal, can you probe it by oscilloscope?
No more time to mess with BL signals inverse engineering, open the pane and test the leds, i guess you can test each strip with a very low current about 20mA to see if only some leds doesn't light up and the rest only a bit, so the unlit ones must be changed..
i don't know that model, usually there is a metal frame around the screen with screws, better if tv is been sit on a low level, before the bezel must be detached the panel boards by its stands/locks, with or without screws, after the bezel, must lift millimeters the border of panel opposite to the boards and lifted slowly perpendicular and then removed to a safe place, now screws on the border to remove the black plastic foils holder bezel, then remove foils and keep them in order in other place, now remove the plastic pins, then remove the white foil..
Thanks. I appreciate it.
Does the PSU only monitor the led strips current in order to diagnose ?
IF that is the case I could simply use dummy loads to simulate a proper backlight and check if still boot loops ?
you have to watch the table serigraph on the PSB, for example 180v 2Amax you have to put 180/2=90ohm/360w load, but i'm not sure about this because this is the max output but tv has probably a % of the maximum setting.. you are safe if the controller only stops at the over maximum or under minimum..
Made some measurements on the led supply on PSU. There are 4 outputs and obviously they are 2 and 2 in series.
They all measure the same in terms of voltage and currents.
So does not indicate a problem with the leds.
I would assume there would be a unbalance if one of the led strips were bad, right ?
Comment