(347) 552-2258 [email protected]

820-3332 no gpu vcore and nothing on screen

ytesfay80

ytesfay80

Member
This board had a lot of water damage in the area of the cpu/gpu. The fans would spin and the heatsink would get slightly warm. While the liquid damage was still there I measured ppcpu_s0vcore and ppvcore_gpu and I got 0.83v on cpuvcore and I believe I had around 0.9v on gpuvcore(I am not 100% sure about gpu vcore).
I removed the taped cover that is above the heatsink area and there was bad liquid damage under there. I replaced Q7510/Q7550/Q8930/R7469 since they looked bad. I put flux and heated the rest of the areas that looked damaged then UC'd the board.

The fans still spin now but now I dont feel the heatsink getting warm. I have around 0.82v on ppvcore_s0_cpu but 0v on ppvcore_gpu and pp3v3_s0gpu also has 0v
Pin 1 of U9000 is 0v even with display cable plugged in.

I'm not sure if any of those components I replaced caused this.
pp3v3_s0 present
ppbus_g3h - stable 12.6v
PPVIN S5 HS COMPUTING ISNS - 12.6v

Here were the areas I marked that had liquid damage before cleaning them.

c8928 and c8923
q7510/q8930
j5815
u5860
j4410
c9820
u7541
c7576 and caps next to

I did these measurements with no external components plugged in and with a genuine 85w charger.
 
Last edited:
ytesfay80

ytesfay80

Member
U9100 had no signs of liquid damage near it. All the probe points around it were all gold.

I still fluxed and heated U9100 but I still have no gpu vcore or anything on L8360.
 
D

dukefawks

Administrator
PP3V3_S0GPU is the first GPU rail that should come on. Check if it pulses maybe once after power on. Did you check for shorts on any GPU rails?
 
ytesfay80

ytesfay80

Member
I checked for PP3V3_S0GPU on L8604. As soon as the charger was plugged in it would show around 1.4v and quickly go down towards 0v.

In diode mode on Pin 1 of L8604 I got 0.46v so that doesn't look like a short.

Could any of this been caused when I replaced Q7510/Q7550/Q8930? It's difficult to see if it's correctly in place since it has a cover on it. I also had to use a lot of hot air to get them off so hopefully I didn't kill the cpu/gpu.
 
Last edited:
ytesfay80

ytesfay80

Member
I checked GPU Vcore on C8962/C8961 in diode mode and I got 0.04v.
PP1V05_GPU_FB (Pin 2 of R8305) in diode mode shows 0.48v
 
ytesfay80

ytesfay80

Member
Okay I replaced Q8930, Q7550 and Q7510 one by one and turned it on.
There was still no Gpu vcore after replacing the first two mosfets but after I replaced the last one(Q7510) the fans will spin for a couple of seconds then turn off and it keeps doing the cycle over and over.
 
ytesfay80

ytesfay80

Member
Gpu vcore only pulsed to about 0.1v

In diode mode cpu vcore is reading 0.002v and gpu vcore is 0.034
Is that a short on cpu vcore?
 
Last edited:
D

dukefawks

Administrator
Nah, that is probably ok. I really have no ideas for this one, but to check all your work and inspect for corroded traces/resistors.
 
ytesfay80

ytesfay80

Member
Okay so I replaced some caps that looked bad. First was C7121. Still didn't make a change.
C8923/C8928 looked really bad so I removed them and there was no pad on the Pin 1 side. I plugged in the charger again and measured for PPVIN_S5_HS_GPU_ISNS on Q8930 and It showed 12.6v.
I added solder and scraped both sides and I could see some of the black coating of the board coming off of the ground pads. I had to run a wire from C8923/C8928 to Pin 7 of Q8930.

I plugged the charger in and now the fans weren't spinning at all and PPVIN_S5_HS_GPU_ISNS was 1.4v and in diode mode it was 0.035v to ground
I'm not sure if the black enamel coating that scraped from the board is causing the short.

I injected around 2v into Pin 7 of Q8930(PPVIN_S5_HS_GPU_ISNS) and the only thing getting warm was the U8000 Nvidia chip.
 
You must log in or register to reply here.
Top