820-3476 Boots fully, no display no backlight

realmacmods

Member
Working on a 820-3476. It will boot fully (sometimes it says aloud "to choose english as your language press the return key), but no display via internal screen or thunderbolt display.

Doesn't appear to have liquid damage anywhere. not getting enables on u8330. if i remove u8330 and hot wire the enables the backlight will turn on, and the screen will get power at the connector. HPD 2.5v at r8301. i2c at connector appears to all be there. nothing measurable on the EDP termination resistors, but the resistors themselves are 1M to ground

Pram reset successfully, but no change, same for SMC reset.

known good screen, put a known good logic board in the chassis and all worked.

looks like the enables come directly from u0500 (cpu), so i suspect something isn't happening to tell the CPU to take those high, but i cant figure out what could possibly be missing. Especially given that it boots fully.
 

Alan.L

Member
you might get to see an external display if you disconnect the internal display cable first.

check if you get any voltage on L8300 for a few seconds when it boots up. it should have least 5v even if you don't have a screen connected,
 

realmacmods

Member
With internal screen disconnected, no external display. Also, SMC lid is working correctly as well. L8300 will indeed go high (5V) briefly on initial startup.

swapped U8340 from a donor board to no change. Was hoping for a HPD translation issue, no luck.
 
Last edited:

2informaticos

Administrator
Staff member
J8300 pin 9 goes to 2.5V when you have 5V at pins 28-30?
LCD must be connected.
If not, display is not recognized and 5V dissapears.
 

realmacmods

Member
DP_INT_HPD goes to 3.3V from about 1/2 second before fan spin, and remains at 3.3V indefinitely. Being attached to logic level shifter, this must be a binary signal. Dont have a scope here at the moment, but can grab one tomorrow. Is there any documentation of what the HPD handshake is supposed to look like? If im even on the right track with the HPD signal, please feel free to ignore me if you have further diagnosis steps

EDIT: DP_INT_HPD indeed drops to 0 as the 2.5v drops to zero. Not sure what happened on that first measurement, but now it consistently drops down
 
Last edited:

2informaticos

Administrator
Staff member
If HPD drops, LCD communication is lost.
No image, nor backlight possible then...
Did you try known good LCD assembly?
 

2informaticos

Administrator
Staff member
If you don't get external video, looks like Intel graphics is dead...
Check around CPU for any knocked off, or moved component.
Try new BIOS, just in case.
 
Top