Samsung QN75Q6FNA backlight no display, red LED blink 2x, fault in panel

[howardc64]

• Prior owner tried a replacement main board. No difference.
• Samsung 2x red blinking standby LED = panel fault
• Jumpered FB_TRDY_1 (pin 1 CN1300 left. main tcon->left side buffer boards) and FB_TRDY_3 (pin 95 CN1301 right. main tcon->right side buffer boards) This allows testing individual T-Con (on main board) connection to panel on this Samsung. Found fault on left side (viewing from rear of TV)
• Each side has 2 buffer boards, disconnected left side flex between 2 buffer boards. Fault is on the left most board.
• Tape masking eventually isolated to CN1300 pin 15 (CKV4_GOA) Display runs fine but has lines on the left side every 5? lines of pixels (10x magnifying loupe) lines runs across 2/3 of the screen towards the right eventually fading away
• CKV4_GOA (Gate On Array) directly connected to left most buffer board to panel flex. These are row gate lines and unfortunately gate is in the glass
• Service Manual ( link ) pg 5-9 5-10 are TCon to buffer board flex connection (CN1300 CN1301) signal lists. Typical Samsung USIT (Samsung's Unified Standard Interface for TV, basically high speed signaling technology for 4K pixel panels) signal listing

I think no way to get rid of the lines? Cutting CKV4 line on buffer board shouldn't change anything? Checked CK4_GOA and seems unconnected to any buffer board components and not shorted to ground. Maybe I'll check if can find what signal its shorted to by probing around remainder of the buffer board to panel flex connections.

Also found some debris under the glass but panel appears to have never been disassembled. I’ve seen dust get inside 2012+ 27" iMac screens between LCD and diffuser. I also hear bugs getting in between LCD and diffuser. Maybe happened here as well?

[nubinator]

check my thread here, i had those exact lines and solved by shorting the clock lines together after masking
https://www.badcaps.net/forum/troubl...samsung-ru7100

[howardc64]

check my thread here, i had those exact lines and solved by shorting the clock lines together after masking
https://www.badcaps.net/forum/troubl...samsung-ru7100

Thanks. Interesting trick.

I tried also blocking the CKVB4 but no difference so removed it.

Was it necessarily to block VGH VGL?

Finally I wonder if can block and short less CKV* lines to get good image. Will study more.

[nubinator]

Thanks. Interesting trick.

I tried also blocking the CKVB4 but no difference so removed it.

Was it necessarily to block VGH VGL?

Finally I wonder if can block and short less CKV* lines to get good image. Will study more.

I tried many combination of masking and nothing made any real difference until shorting the clock lines. the color fade was much better too so didn't feel the need to do anything else. I don't know if vgl,vgh was necessary i only masked those not shorted.
in your pic of buffer board those 4 diode packs are the clock lines just short the top most pins all together that's what I did, looks like you have 12 clock lines for 75" mine was 8 for 65"

[howardc64]

I tried many combination of masking and nothing made any real difference until shorting the clock lines. the color fade was much better too so didn't feel the need to do anything else. I don't know if vgl,vgh was necessary i only masked those not shorted.
in your pic of buffer board those 4 diode packs are the clock lines just short the top most pins all together that's what I did, looks like you have 12 clock lines for 75" mine was 8 for 65"

Thanks. Before anyone suggest. Just traced the CKV4_GOA signal to the TVS (Transient Voltage Supressor diode array, marking is E5 4(side ways) in SOT363 package, could'n't locate data sheet) and removed it. Fault is unfortunately in the COF/panel. Studying more on CKV*/CKVB* before trying the mask all CKV* lines and short their ESD diodes. Buffer board shows 12 CKV lines (CKV[1-6] CKVB[1-6]) but service manual T-Con connector (CN1300) list only 8 (CKV[1-4] CKVB[1-4])

[lotas]

TVS -"E5" = ESD5V0S5US

[lotas]

E5 is PESD24VS5UD (24v)

[howardc64]

TVS -"E5" = ESD5V0S5US

Thanks! I actually did find these data sheets but couldn't understand the buffer board circuit. pin 4,5,6 are all connected to ground on every TVS (4x on left and right buffer board) I guess only 3/5 channels are used on each TVS. Thanks again for confirming the TVS spec.

[howardc64]

Unfortunately masking 8 CKV* lines and shorting their TVS connection didn't work for me

• masking CKV[1-4] and CKVB[1-4] and shorting all their connections on TVS didn't help. Picture got an extra line when 8 CKV* lines were masked
• TVS connection to CKV5 and CKVB5 has no T-Con connection on CN1300
• CKV6 connects to CN1300 pin 21 STVP1_GOA (but not pin 20 which is also STVP1_GOA)
• CKVB6 connects to CN1300 pin 22 ST_GOA
• Clearly CN1300 signal labling in service manual is wrong for 75" panels.
• Masking CKV4 and CKV4+CKVB4 shows identical line pattern on screen. So masking CKVB4 does nothing. This is the cleanest image.

Reluctant to short TVS connection for CKV5/6 CKVB5/6 since lacking T-Con signal clarity.

Maybe I'll just cut CKV4 line at the panel flex and see if thats different result than masking it at T-Con connector.

[howardc64]

UPDATE : Fixed enough to be useful. Don't know how long it will last or if faulty side panel color can be improved.

• All 6 pairs of CKV CKVB lines (12 total) masked
• All 6 pairs of CKV CKVB lines jumpers at TVS
• Faulty side has color fade and lines
• Service Manual T-Con connector labeling is likely wrong for 75" (probably also 85?) panel
• Making some guesses on clock line short failures

Masked 6 pairs of CKV lines

The Service Manual left panel T-Con connector CN1300 is likely wrong. Able to trace ~9V on all CKV* signals to determine the missing CKV*[5-6] pins (21-24) These were labeled ST* and DEMURA* lines and clearly wrong.
Masked all 6 pairs and get the following solid color test image


Left side looks bluish in the picture but general a lot less color in that region. Almost grayscale.
More prominant lines maybe where the panel fault is? Just a guess.
Lines would change colors dynamically becoming noticible.

Jumpered 6 pairs of CKV lines at TVS

This stopped the lines from changing colors and remain static. Much less noticible.
Faded color on the faulty side still present. See attached video. Might be the best possible for a large panel.



Larger Panel CKV line Masking

This is just a guess but the typical CKV line masking trick depends on the Panel's other side CKV to drive the lines all the way across the panel. Larger panel have longer distance to travel and probably more easily loose signal strength. Don't know if manufacturers will increase drive strength and line thickness for larger panels. I had great results in the past on a 43" LG 4K. This 75" panel have more problems on the masked side.

Result After CKV line masking maybe random

Fault is in the panel. Some circuit failed and may have damaged other nearby circuits. Result is surely random. All the youtube video showing success in various ways are likely just specific failures. Of course if mask more lines, then more circuit is removed.

Guess on Why Samsung (and other) panels have panel failures

For panel glass to include circuits. Glass need to have treatment to increase electron motility (how easy for electron to move on a material which then determines circuit speed.

• The 2 opposite ends for silicon electron motility are amorphous glass (windows, slowest) and single crystal silicon (silicon chip, fastest).
• There are various material treatments (usually by increasing heat and also different material coating) that increases eletron motility. LTPS (Low Temperature PolySilicon), HTPS (High Temp PolySilicon) are example methods to increase electron motility for making glass displays. HTPS is used for micro displays (eg 4x3cm FHD display)
• Faster = more manufacturing step and cost

I'd imagine panel manufacturers are constantly changing this area depending on circuit requirements and cost reduction. They probably chose a solution that unfortunately results in more circuit failures

Guess on why connecting masked CKV* signals reduce lines on screen

These lines when masked are floating when measured without power. Mine shorted CVK4 shows Open Loop when power is off. There are gates that probably need to be switched when the panel is running to encounter the short. Connecting them all together is probably just eliminating the floating lines? At least 1 CKV* signal probably need to not have a fault for this to work? Just a guess.

[nubinator]

Good work , looks like a decent result to me, I think when there is short in panel on clock lines this is the best you can expect.

[howardc64]

Good work , looks like a decent result to me, I think when there is short in panel on clock lines this is the best you can expect.

Yeah, thanks for the hint. The video you shared also shows masking LC1 LC2. Don't know what are equivalent Samsung signals. Regardless, the T-Con connector signal spec in this TV's service manual is incorrect so need some detective work to find them. Curious if that'd improve the few not so noticeable lines I have remaining.