A1990 820-01814 had PPBUS_HS_CPU short (fixed) and 2 days later stopped working again...

Paul Daniels

Super Moderator
Machine came in originally dead, visually located C7253 had burned up ( no track damage ) and replaced by hand-soldering replacement on ( no hot air ). Worked fine for a couple of days but when I went to perform the packing-exit-check to dispatch it no longer booted.

Machine now will take 20V after what feels like a bit of a delay (seconds longer than usual) and has very short 250mA peak draws but mostly 50~70mA and loops every 10 seconds.

All 4 ports exhibit the same behaviour.

NAND 2V5/1V8/0V9 rails good likewise good resistances there.

Going through the startup rail sequences all seems fine until P5VG3S_EN / P3VG3S_EN are cycling every 10 seconds.

PMU_ONOFF_L is present but attempting to pull it low doesn't bring 3V3_S5 or 1V85_S5 up and not seeing anything on the scope either ( transient capture ).

My thoughts are that something was damaged from the original HS_CPU short, though I have replaced R5400 in the small possibility it went high.

Thought I'd ask here before I start looking at the CD3215C00Z's

Parting notes:
* Connecting to AC2 gives "Cannot use/support this thunderbolt device"
* Every 10~15 seconds I can see the USB-C meter flick back to 0V for an short moment ( 1 sample, about 100sps on the meter ) but it doesn't reset back to 5V
 

2informaticos

Administrator
Staff member
Compare diode mode readings on big coils with good board.

The drop on G3S voltages can be caused by a problem on Intel side power rails; S5 to S0.
 
Top