820-2915, USB supply (U4600) burned -> replaced now short on PP5V_S0, U9500 HOT!

george-a

Member
Hi guys,

this one had an exploded u4600 when it arrived (it smoked while the customer was testing it) and I replaced it with new one and the short seems to be gone.

now when I turn it on the fans start and stop after 1-2 secs OR can last upto 10sec - this is when I noticed that U9500 gets REALLY hot and by measuring around I got a diode mode reading on 0,018 on PP5V_S0. Of course I have no PP1V5_GPU_REG and I measure 0V on PPVCORE_S0_AXG, too.

I see PP5V_S0 is generated by Q7860, should I remove that and inject 5V with a power supply to see what gets hot? (there is also C7860 there which I may also have to remove just incase it shorts).

Should I replace U9500 first? Since it gets hot maybe thats the one shorted to ground?

Where would you start?

Please advise :)


BR,
george
 

dukefawks

Administrator
Surge on 5V bus, there may be lots of damage. Start with U9500 and go from there. Usually U4800 is the first one to die, check 5V_S3 for low resistance too.
 

george-a

Member
ok duke I will replace U9500 today and let you know..

diode mode to ground for PP5V_S3 is 0,381 when it was close to zero before u4600 replacement. Resistance between GND and PP5V_S3 is approx. 16.3Kohm.

u4800 is important for what? I see "IR Support" as title in the schematic.. or you mean it could also short?

george
 

george-a

Member
update: I removed U9500 which was getting very hot but measuring PP5V_S0 at C9501 gives 0.023 diode mode. This means the short is present elsewhere on this rail but was also somehow affecting U9500.

should I inject 5V now? You see me reluctant because I would really like to follow the correct procedure with this one without "sudden moves" ...
 

dukefawks

Administrator
Yeah inject 5V and see what heats up, but this board will have extensive damage. Probably 50% of chips on 5V_S0 bus are dead, starting with all the ISL chips.
 
Top