Repair of 404652HHC8LV1.8 T-CON board

[Yanleb]

I just want to share the fix of my Sony KDL-46XBR4. It had an intermittent milky white image, typical of a bad chroma chip. The T-CON board model is 404652HHC8LV1.8 from Samsung.

The chroma chip on the T-CON board is an ISL24005. This is a programmable one (through an I2C interface), but it has 4 input references (VREFL_L, VREFL_H, VREFU_L and VREFU_H).

Thanks to the designers of this T-CON board, all the tests points for VGMA1-VGMA18 are on the BACK side of the board. I had to measure the VGMA voltages on the front side at the vias going to the surface mount connectors.

As expected, the VGMA voltages were off while the problem happened. I traced down the problem to the 4 input reference voltages. They were easily measured at the top of the 4 capacitors (CG4-CG7) just below the chroma chip (ICG1). The 4 reference voltages were produced by a series of resistors on the bottom side of the board. Here are the measured voltages through the series resistors :
GND
RG24 (390 ohm) : 0.19 volt good, 0.01 volt bad (measured at CG7)
RG51 (0 ohm)
RG23 (10k ohm) : 6.14 volt good, 15.18 volt bad (measured at CG6)
RG 49 (0 ohm)
RG22 (2.7k ohm) : 10.23 volt good, 15.24 volt bad (measured at CG5)
RG50 (0 ohm)
RG21 (10k ohm) : 15.12 volt good, 15.31 volt bad (measured at CG4)
RG20 (390 ohm)
16v input

It turned out that the 10k ohm resistor (marked 01C in EIA-96 notation) at position RG23 was intermittently open. Later on I also found the 2.7k resistor at position RG22 with an intermittent problem. Replaced the two resistors and the TV set works like a new, with vivid colors.

The fix would also apply to the similar 404652FIX2HC6LV1.2 T-CON board also from Samsung, and to the following TV sets:
KDL-40WL135
KDL-40XBR4
KDL-40XBR5
XBR-46HX909
KDL-46W5000
KDL-46WL135
KDL-46XBR4
KDL-46XBR5
KDL-52XBR4

If you are aware of any other models not in my list please post them.

Thanks.

[ramondo]

Re: Repair of 404652HHC8LV1.8 T-CON board

Thank You for an informative repair solution with steps to follow.

[Ranmamez]

Re: Repair of 404652HHC8LV1.8 T-CON board

Hi.
I have a Sony kdl-52x3500 with the same tcon, and the tv-set has a weird issue: the right half of the image is brighter than the left side (see picture).
I measured the same points on the tcon and these are the voltages:

• Cg7: 0,252v
• Cg6: 6,73v
• Cg5: 8,47v
• Cg4: 14,97v

Where can I measure the 16v input?
Btw, I tried to measure the resistances on the back of the tcon board and the 10kΩ measured kinda 6,7kΩ but onboard, not desoldered, and the 390Ω more or less 384~Ω, still not desoldered: should I try them off board?
I almost forgot: where did you get the correct values for this tcon? Do you have a schematic diagram?
Thanks.

[mlemon]

Re: Repair of 404652HHC8LV1.8 T-CON board

Nice work and thanks for sharing the details.

[Yanleb]

Re: Repair of 404652HHC8LV1.8 T-CON board

Hi Ranmamez,

Your 16v input is good since the other voltages are present. Your problem is definitely not the same as mine. There's only one chroma chip and a problem about it would affect the entire screen.

The answer for your second question about the correct values is that I have measured them myself. The problem was intermittent, so I had the chance of measuring them when the display was good, and compare them when the problem occurred.

The answer to your third question about the schematic diagram is that I don't have it. In fact, there is a very limited amount of T-CON board diagrams available. Often I make my way by looking at the spec sheet of the various chips present on the board. Often you will find a typical application diagram for that chip. And sometimes it helps since the developer of that board used the same design as the one in the chip's documentation.

In a few cases, you can find documentation about the LCD panel itself. Often the T-CON board comes with the LCD panel from the manufacturer. So you can find some information about the inputs and outputs of the T-CON board. AUO is an example of a manufacturer that sometimes provide this information.

There's a guide that lists common defects on T-CON boards. Google for "Collection of T-CON board repairing information by Kent Liew".

But there's a good news. Half of the display is ok. So you have a known good reference of the voltages going to the LCD panel on the good side. Take voltages measurements and compare both sides.

But there's also the possibility that the problem lies on the connections to the LCD panel's bottom board (or top board, depending on the T-CON board location). Re-seat all the ribbon connectors. If it doesn't help the LCD panel may have a tab bonding issue. Be careful and apply pressure on the flex cables going from the bottom/top board to the LCD panel. You may have to carefully remove the LCD panel metal frame to apply pressure to the flex cables going to the glass panel.

Good luck.

[Ranmamez]

Re: Repair of 404652HHC8LV1.8 T-CON board

Hi Yanleb, thank you very much for answering me.
I know my problem is completely different to yours, but I give it a shot, since I'm wandering around blind, not having info to what to do and/or what to check.
I see that you got your information and a way to fix your problem, while checking the presumably correct values on those test points.
Thanks also for the tip of what to search online, I just got it, I hope it would help me.
About the latest hints, I checked the VDD and the AVDD test points on the top boards (they're inside the frame as you stated but on top, luckily so I hadn't to stay bent ^^) and they are off of some tenth to hundredth of Volt between the two.
I tried to bend and move every flat cable but to no avail, the image didn't make a glitch, either in a way nor in another.
Anyway I don't think it's a panel issue, because if I disconnect the left flats, where the correct image is shown, the left side goes blank (to white or such) but the right side downs the brightness to normal, more or less, so I guess the issue is on the tcon.
Finally I was thinking to ask you a little bit more help: do you still have that tv set? maybe we can cross-check some voltages around? I found another point where the voltage differ some hundredth, i.e.15,22v vs 15,28v on some couple of (big) smd capacitors, on the left and on the right side on the top part of the tcon.
What do you think?
Thanks for reading.

[Davi.p]

Re: Repair of 404652HHC8LV1.8 T-CON board

Hi , i am James Bonz from PLCforum.. detach each flat cable from tcon to screen and test in volt every VGMA test point on the rear, then post them here with chrono order and name.. i suspect that 10k resister.
Suspected also something like a capacitor on the bad side's long board tied to vcom signal, search its test point on tcon, then trace it to both long boards and compare them on each..

[mlemon]

Re: Repair of 404652HHC8LV1.8 T-CON board

Curious if you can connect the left side (working side) to the right (not working)! You will need to maneuver the Tcon board and ground it but I think this will for sure tell if:
1. LCD is good or not &
2. Confirms if issue with the Tcon.
Good luck.

[Davi.p]

Re: Repair of 404652HHC8LV1.8 T-CON board

usually signals disposition on a side is different from the other, partially mirrored, so that action would destroy all..

[mlemon]

Re: Repair of 404652HHC8LV1.8 T-CON board

I see. Was thinking they are probably mirrored however, may have to flip the flat flex cables. I guess I wouldn't try that then. Thanks

[Davi.p]

Re: Repair of 404652HHC8LV1.8 T-CON board

no them aren't completely mirrored..

[Ranmamez]

Re: Repair of 404652HHC8LV1.8 T-CON board

Hi , i am James Bonz from PLCforum.. detach each flat cable from tcon to screen and test in volt every VGMA test point on the rear, then post them here with chrono order and name.. i suspect that 10k resister.
Suspected also something like a capacitor on the bad side's long board tied to vcom signal, search its test point on tcon, then trace it to both long boards and compare them on each..

Hi James, we meet here too ^^
Anyway, I measured the voltages on every point with the flat detached and they differ on a matter of hundredth of volt.
Instead, on that capacitors that I measured and that I pointed in the picture I posted on the other forum (cd141 cd142 cd144 and cd149) the voltage goes up of some hundredth to tenth of volt.
Anyway, Yanleb, can you help me confront measures, please?

[Davi.p]

Re: Repair of 404652HHC8LV1.8 T-CON board

maybe i wasn't clear, you better test the VGMA testpoints on the rear of the tcon, and post here the results..
but thinking on better i guess your reference volt measurements are different because you have a different panel, so i guess your problem is from another source..