2015 MacBook Pro Retina 820-3787

[Dakay]

I'm stuck. Working on an 820-3787, 2015 MBP Retina. PPBUS_G3H does not have 12V, only about 2V. I discovered that mosfet Q9330 near the GPU was getting hot. I removed it and sure enough it restored my 12V. However, when I replaced it the short returned. I replaced all three mosfets in the area, Q9330, Q9331, and Q9361. No dice. So I resoldered and eventually replaced U9300. Still no luck. In fact now, even when I remove Q9330, there's a short on PPBUS_G3H. When I send power to it, it draws .2 amps, pin 1 on Q9330 sparks and gets hot, which is the UGATE from U9300, only showing about 1V on it.

U9300 isn't getting the right amount of voltage in, which would cause the gate to not open, but im not too sure why it's not getting the voltage. I think it might be a faulty GPU. Does anyone have any ideas? Someone else had worked on this board before me and I'd rather not waste a ton of time on something I can't fix.

[robertpritchard]

Re: 2015 MacBook Pro Retina 820-3787

isl6259 sends voltage to u9300 opening logic gate to its 5volt if no dice check for your short on gpu

[piernov]

Re: 2015 MacBook Pro Retina 820-3787

U9300 and Q9330 must be replaced at the same time.
Chances are GPU is damaged since it got 12V. You can at least measure exact resistance to ground on PPVCORE_S0_GFX_REG.

However I'm confused about what "pin 1 on Q9330 sparks and gets hot" and "U9300 isn’t getting the right amount of voltage in" mean.
If someone worked on this board already there must be signs of rework. Make sure GPU itself hasn't been messed with.