820-00928 No Display

[tcampbell34]

Hello everyone! I have a 820-00928 board that is not displaying image to the internal display. This board had some minor liquid damage near U3200 and was originally only giving me 5v on the charger. I replaced U3200 with a new CD3215 chip and the board now gives me 20v and boots to the OS, but only displays image to an external display. I have read most of the posts here about the A1707's with no display and have already tried to replace U9850 3 times and the screen is still black. No image or backlight shows up on the screen but all the screen voltages seem to be present. External display is working fine using usb-c to HDMI adapter so not sure where to look next. The LCD connector on the board looks perfect also. Any help would be appreciated. Here are some voltage measurements I took while the board was connected to my know good test screen only.

PP5V_S0SW_LCD - 5.1
PP3V3_S0SW_LCD - 3.3
EDP_PANEL_PWR_EN - 3.3
PANEL_FET_EN_DLY - 3.4
DFR_DISP_PWR_EN - 3.4
BUF_EDP_PANEL_PWR_EN - 3.4
BUF_SMC_REST_L - 3.4
DFR_DISP_SMC_RST_L - 3.4
SMC_RESET_L - 3.4
DP_INT_HPD - 3.3

 

[2informaticos]

First of all, welcome to the forum!

Are you missing image and backlight at same time?
LCD panel looks to be recognized.
Check if there is image at least.
Post exact voltage for PPVOUT_S0_LCDBKLT.
Did you try known good full LCD assembly?
ALso check all SMC_LID signals.

 

[tcampbell34]

Thanks for the welcome! I am missing both backlight and image at the same time and I have tested this board using a full assembly that I know works with another board, and it still shows nothing on internal screen. I have tried to see if there is any image on the screen using a flashlight and there is nothing.

PPVOUT_S0_LCDBKLT - 0
SMC_LID - 3.4
SMC_LID_LEFT - 3.4
SMC_LID_RIGHT - 3.4

 

[2informaticos]

U9800 could be damaged too.
This one may have firmware, I'm not sure however.

 

[tcampbell34]

Ok I will try that and report back. I am waiting on a stencil so I can reball that chip. I only have a 820-00281 donor board at the moment which is a 2016 A1707 so I'm not sure if the U9800 on that donor will work on the 820-00928. Should I try it anyway?

 

[tcampbell34]

I replaced U9800 on this board with one from a 820-00281 donor but it did not change anything. Board still boots fine to external display but wont show image or backlight on a known good LCD assembly. Do you know what else could cause this other than U9800 and U9850?

 

[2informaticos]

CPU should have video output damaged then.
I don't see other reason.
Apart of possible bad solders on U9850.
Inspect for knocked off, or moved components in that area.

 

[tcampbell34]

I inspected the area and did find 1 resistor that was soldered on sideways. It was RA821, which is the resistor for EG_BKLT_EN. I checked the value of that resistor and the one next to it, which is for EG_LCD_PWR_EN and they both measure 8.3k when the schematic shows they should be 10k. I replaced them both with new 10k resistors but they still show as 8.3k when measured on the board. I'm not sure if this means anything or not. Is there something else I can measure to confirm that the issue is with the CPU video output?

 

[2informaticos]

For high ohm resistors is normal to find lower value measured onboard.
That resistor shouldn't affect CPU.

Do you still get external video?
When you plug external monitor, I expect U9850 switch to GPU output too.
Are you sure no image on internal LCD then?
Maybe missing only backlight...

 

[tcampbell34]

Ok thanks for letting me know that. I did check for image on the screen using a flashlight while the external monitor was connected and there is no image. The screen is completely black but external video is fine. EG_LCD_PWR_EN and EDP_PANEL_PWR_EN are both 3.3v but EG_BKLT_EN and EDP_BKLT_EN are both 0v. If the issue was with the CPU video output, shouldn't I still be getting the backlight enable voltage?

 

[2informaticos]

You may have now problem with U9800/50...

 

[tcampbell34]

I replaced U9800 again on this board but still got the same issue. I did notice while measuring that PP3V3_S0_GPU has 3.29v, but in diode mode, it shows as .031. I measured this on a working board and it measured 0.439. In resistance mode, it measures 60 ohm on this board and 65K ohms on a good board. Does this mean the GPU is bad?

 

[2informaticos]

Is possible, but there are other chips connected too.
You can try removing them (one by one) and checking the change in diode mode...

 

[tcampbell34]

I removed UA950, QA800, and UA801 and checked the diode mode value of PP3V3_S0_GPU and it was still 0.031. All the other chips on this line are not actually populated on the board so the only thing left is the capacitors, resistors and the GPU. Should I start removing caps and resistors or do you think its the GPU?

 

[2informaticos]

Strange to have bad GPU if you get external video.

Check on external monitor if GPU, or iGPU is the active graphics.
On older models external monitor is always driven by GPU.
I didn't check any new model to see if is still the same.

 

[tcampbell34]

Yea it does appear to be using the AMD GPU when I look on the external display. Idk where to go from here. This board is really starting to get on my nerves.

 

[2informaticos]

In such case dedicated GPU should pass through U9850 too.
Post LCD_MUX_SEL/EN signals value.
Check for knocked off, or moved resistors around U9800/50; especially pull-ups/downs ones.

 

[tcampbell34]

With the internal screen connected only, LCD_MUX_SEL/EN are both 3.3v. When I connect the external screen, LCD_MUX_SEL goes to 0v but the enable is still 3.3v. I checked the whole board again for knocked off or moved components but I did not find anything. I did notice almost all the capacitors around U6400 and U6450 are burnt on the PPBUS_G3H side, but this is for audio so I don't think it has anything to do with the screen. This area is close to the backlight driver though.

I checked the resistance of all the resistors on page 89 of the schematic and they all measure correctly except for the below resistors which should all be 100k. I know high ohm resistors can measure lower on the board but I did compare to a good board and they are all 100k on the good board.

R9873 - 54.6K
R9875 - 67.1K
R9837 - 55.1K
R9846 - 55.7K
R9848 - 54.7K
R9852 - 55.6K
R9853 - 31.3K

 

[2informaticos]

If LCD_MUX_SEL switch to 0V when connect external monitor, means U9850 should allow DP_INT_EG signals to pass.
Low chance to have both DP_INT_IG/EG video outputs bad.
More probable one problem with U9850, or some component around it.
If not, connection problem with the LCD assembly.

Can you check the input/output signals at U9850 with an o-scope?

Mentioned resistors are connected with U9800 and point to a problem with it.
If no video signal (on o-scope) at U9850 inputs, then U9800 is suspect.

 

[tcampbell34]

I am not able to check the inputs of U9850 because they only go to a ball under the CPU or GPU, but I checked all the outputs at the LCD connector with a test screen connected, and they are all 0v except for EDP_AUXCH_C_N, which is 0.98v with no wave on the o-scope. I did replace U9800 from a 820-00281 donor since I do not have a 820-0928 so that could be the issue.