I have a MacBook Pro 15" late 2011, it has a dead GPU and was sent to a repair center for this issue multiple times, the last time the original owner of this machine decided to not fix it anymore. The repair center returned the machine, but they desoldered the GPU and VRAM, it just physically isn't there anymore. The machine did originally work, we'll get to it.
What it does now: does not turn on in SMC bypass mode, the fans just twitch and I can hear a ticking sound from the board, almost like there was a short circuit. But in normal mode it does start, fans spin for about 10 seconds, power light glows, no chime, then fans stop spinning for about 5 seconds, then they spin up for 10 seconds and this cycle just repeats. If I start it without RAM, fans stay spinning indefinitely and it does beep as it should. Both CPU and PCH are getting warm. I checked for shorts, couldn't find any. There is no visible corrosion on the board. I measured voltages, all measurements were done with battery disconncted, here are the results:
PPBUS_G3H 12.6V
PPVIN_S5_HS_COMPUTING_ISNS 12.6V
PPVIN_S5_HS_GPU_ISNS 12.6V
PPVIN_S5_HS_OTHER_ISNS 12.6V
PPDCIN_G3H 16.5V when fan spins 16.2 when fan doesn't spin, battery disconnected
PP3V42_G3H 3.48V
PPVRTC_G3H 3.35V
PP5V_S5 5V
PP5V_SUS 5V
PP5V_S3 5V
PP5V_S0 5V
PP3V3_S4 3.35V
PP3V3_SUS 3.35V
PP3V3_S5 3.35V
PP3V3_S3 3.35V
PP3V3_S0 3.35V
PP3V3_T29 3.35V
PP1V05_T29 1.03V
PP15V_T29 0.5V not shorted (84kOhm)
PP1V8_S0 1.8V
PP1V8_S0_CPU_VCCPLL_R 1.8V
PP1V5_S3 1.5V
PP1V5_S3RS0_CPUDDR 1.5V
PP1V5_S0 1.5V
PPVTTDDR_S3 0.75V
PP0V75_S0_DDRVTT 0.75V
PP1V2_S0 1.2V
PP1V05_SUS 1.05V
PP1V05_S0 1.05V
PP3V3_S0GPU 3.35V
PP1V8_GPUIFPX 1.8V
PP1V8_S0GPU_ISNS 1.8V
PP1V5_GPU_REG 1.5V
PP1V5_S0GPU_ISNS 1.5V
PP1V0_S0GPU 1V
PP1V0_S0GPU_ISNS 1V
PPVCORE_GPU 1.1V
PPVCORE_S0_CPU 1.15V
PPVCORE_S0_AXG 0V no short, should just mean it's not getting far enough in POST
PP1V5_S3_CPU_VCCDQ 1.5V
PP1V05_S0_CPU_VCCPQE 1.05V
PPVCCSA_S0_REG 0.8V
PPVP_FW 12.65V
PP3V3_FW_FWPHY 3.35V
PP1V0_FW_FWPHY 1V
PPBUS_SW_BKL 0V no short, probably just doesn't get far enough to initialize backlight
PP3V3_ENET 3.35V
PP1V2_ENET 1.2V
ALL_SYS_PWRGOOD 3.3V
SMC_RESET_L 3.47V
Some more info: This machine used to belong to an ex colleague who left it at work, so I decided to take it. I installed macOS on a different machine, deleted radeon kexts and enabled screen sharing. I swapped the drive into it, let it boot up with a blank screen, remoted into it and wrote NVRAM parameters to enable IGP. It did work just fine after that. Then I decided to upgrade RAM, so I did and when I powered it on afterwards, it started doing the power cycle loop thing. As I already said, a repair canter physically removed both the GPU and VRAM. They didn't clean up, pads still had solder balls and a ton of flux. So I cleaned the pads with some solder wick and used alcohol to get rid of the old flux. No change, still power cycling. So I left it do its power cycle thing, after about 10 minutes it suddenly chimed and started right up. So I shut it down, upgraded RAM, turned it on and after about 10 minutes of power cycling, it started. But the next day it wouldn't start even after two hours. I tried desoldering R8911 to disable GPU power completely, that changed nothing. So I ran a bodge wire to reenable GPU power, again, no change. The machine did originally start even in SMC bypass mode btw, in the brief time it worked. It started even in normal mode. I'm not a board repair technician, I'm just trying to fix this thing for fun, but I'm already at my wits' end. Do you have any idea what could be wrong with it? All voltages seem to be there, the machine should live, but it doesn't. Thank you very much for any help.
What it does now: does not turn on in SMC bypass mode, the fans just twitch and I can hear a ticking sound from the board, almost like there was a short circuit. But in normal mode it does start, fans spin for about 10 seconds, power light glows, no chime, then fans stop spinning for about 5 seconds, then they spin up for 10 seconds and this cycle just repeats. If I start it without RAM, fans stay spinning indefinitely and it does beep as it should. Both CPU and PCH are getting warm. I checked for shorts, couldn't find any. There is no visible corrosion on the board. I measured voltages, all measurements were done with battery disconncted, here are the results:
PPBUS_G3H 12.6V
PPVIN_S5_HS_COMPUTING_ISNS 12.6V
PPVIN_S5_HS_GPU_ISNS 12.6V
PPVIN_S5_HS_OTHER_ISNS 12.6V
PPDCIN_G3H 16.5V when fan spins 16.2 when fan doesn't spin, battery disconnected
PP3V42_G3H 3.48V
PPVRTC_G3H 3.35V
PP5V_S5 5V
PP5V_SUS 5V
PP5V_S3 5V
PP5V_S0 5V
PP3V3_S4 3.35V
PP3V3_SUS 3.35V
PP3V3_S5 3.35V
PP3V3_S3 3.35V
PP3V3_S0 3.35V
PP3V3_T29 3.35V
PP1V05_T29 1.03V
PP15V_T29 0.5V not shorted (84kOhm)
PP1V8_S0 1.8V
PP1V8_S0_CPU_VCCPLL_R 1.8V
PP1V5_S3 1.5V
PP1V5_S3RS0_CPUDDR 1.5V
PP1V5_S0 1.5V
PPVTTDDR_S3 0.75V
PP0V75_S0_DDRVTT 0.75V
PP1V2_S0 1.2V
PP1V05_SUS 1.05V
PP1V05_S0 1.05V
PP3V3_S0GPU 3.35V
PP1V8_GPUIFPX 1.8V
PP1V8_S0GPU_ISNS 1.8V
PP1V5_GPU_REG 1.5V
PP1V5_S0GPU_ISNS 1.5V
PP1V0_S0GPU 1V
PP1V0_S0GPU_ISNS 1V
PPVCORE_GPU 1.1V
PPVCORE_S0_CPU 1.15V
PPVCORE_S0_AXG 0V no short, should just mean it's not getting far enough in POST
PP1V5_S3_CPU_VCCDQ 1.5V
PP1V05_S0_CPU_VCCPQE 1.05V
PPVCCSA_S0_REG 0.8V
PPVP_FW 12.65V
PP3V3_FW_FWPHY 3.35V
PP1V0_FW_FWPHY 1V
PPBUS_SW_BKL 0V no short, probably just doesn't get far enough to initialize backlight
PP3V3_ENET 3.35V
PP1V2_ENET 1.2V
ALL_SYS_PWRGOOD 3.3V
SMC_RESET_L 3.47V
Some more info: This machine used to belong to an ex colleague who left it at work, so I decided to take it. I installed macOS on a different machine, deleted radeon kexts and enabled screen sharing. I swapped the drive into it, let it boot up with a blank screen, remoted into it and wrote NVRAM parameters to enable IGP. It did work just fine after that. Then I decided to upgrade RAM, so I did and when I powered it on afterwards, it started doing the power cycle loop thing. As I already said, a repair canter physically removed both the GPU and VRAM. They didn't clean up, pads still had solder balls and a ton of flux. So I cleaned the pads with some solder wick and used alcohol to get rid of the old flux. No change, still power cycling. So I left it do its power cycle thing, after about 10 minutes it suddenly chimed and started right up. So I shut it down, upgraded RAM, turned it on and after about 10 minutes of power cycling, it started. But the next day it wouldn't start even after two hours. I tried desoldering R8911 to disable GPU power completely, that changed nothing. So I ran a bodge wire to reenable GPU power, again, no change. The machine did originally start even in SMC bypass mode btw, in the brief time it worked. It started even in normal mode. I'm not a board repair technician, I'm just trying to fix this thing for fun, but I'm already at my wits' end. Do you have any idea what could be wrong with it? All voltages seem to be there, the machine should live, but it doesn't. Thank you very much for any help.
Comment