TCL 55S425 few horizontal lines, potential failure on both side of panel/buffer board

[howardc64]

All left and right refers to looking from rear of TV (or looking from front with TV upside down)

• TV came with vertical bars and no image. Found faulty TVS on left buffer board. Also replaced T-Con (may have damaged the original while hot air removing components, board has slight warp) and have full image but few horizontal lines on the screen. Some of the lines appears to be 2 pixel tall while others are 1 pixel tall.
• Disconnecting left and right buffer board potentially reveals problems on both sides with different signals (unless this T-Con require jumper like Samsung FB_TRDY1&3) Fault on both sides makes masking T-Con signal debugging much more difficult. Unable to do standard masking steps. As an example, masking 8 clock lines on either left or right produces much worse image. Masking signal from one side could be removing the only remaining working side signal. This increase potential failure combination significantly.
• Checked many voltages and everything seems normal
• Checked buffer board caps for shorts. Found one on each buffer board @ 6 OHMs. Traced the 6 OHM measurement back on T-Con board itself. Not low resistance on buffer board or panel. Labled in the attached pics
• Replacement T-Con is from HiSense 55H6D with identical part number. Shopjimmy part ( link ) indicate T-Con works for both TCL 55S425 and HiSense 55H6D so I guess the 2 TVs use same panel
• Measured a couple of shorts (even at an inductor) on the T-Con but this is the orignal design (both old and new T-Con is same behavior, working T-Con measures 12v on the "shorted" inductor (when disconnected and not powered on))
• Tried to read Winbond EEPROM (W25X10CL link) image from original and replacement T-Con board. Unfortunatley flashrom app can not find EEPROM device (maybe board prevents it somehow? flashrom -L shows W25X10 is supported). To confirm my EEPROM reader is working correctly, read a Winbond 25X20CL from a GPU without problem.
• Checked remaining 3 TVS's (Each have 3 channels connected) Measures 2+ MOhm to ground on all channels except VSSQ channel (6K Ohm) on both left and right buffer board.

Perhaps have a panel short that affects both sides? Any suggestions on how to proceed? Maybe the few lines is the best result possible?

The lines also look suspicously like frame buffer errors. Some lines have color segments that moves with the UI objects. For example the white line below (above in the upside down TV) Roku Media Player moves with the icon below. Maybe its just picking up colors from nearby lines based on the internal panel failure.



BTW, fixed/improved 3 panels recently with TVS replacement. Will order a few to keep in stock


Disconnect Left/Right buffer board Independently



Left Buffer Board

Right Buffer Board

[lotas]

For diagnostics remove all TVS assemblies and check without them, they can leak under voltage...

[howardc64]

For diagnostics remove all TVS assemblies and check without them, they can leak under voltage...

Yes thought about similar. Did this (no change to the lines) and unfortunately seems to have developed another problem. Left 1/2 of the screen now has a white color overlay (with and without TVS) Not sure what caused it. This panel does have some strange behaviors. Have seen it go from no screen error (other than lines) to screen error back to no screen error without changing anything.

Trace the likely white overlay 1/2 screen to the base voltage of couple of transistors after discovering really hot resistor and COF (mp4 file attached) Transistor Q002 is putting out 14V across 100 ohm resistor to ground. The gate (base) and input emitter) and 1 output (collector) signals from these 2 transistors disappear into the flex and unfortunately no way to trace where they go (other than to the panel)

Datasheet for Q002 ( 2907 link ) and Q001 ( 2222A link )