MP 1,1-5,1 Mac pro 5.1 ramdom shutdown

[Daniel Cardozo]

hi Guys,

i need help with my Mac Pro as I already did a lot of research on net and nothing helped...

i have a Mac Pro 5.1
12 core 2.93
32gb of ram
Gtx 970 (now running a gtx 120 for testing purposes)
2 ssd for OS (mac and windows)

i most usually for 3dmax and autocad on windows bootcamp

it starts 4 months ago with randomly shutdowns, and become more frequently.
On windows, gave me BSOD with WHEA error code. First I reinstalled windows and hold for 3 weeks without shutdown. After started again, I swapped the gpu and tried MAC OS, but it shutted down too randomly. So I run hardwares test and everything was fine.

the shutdown become more frequently at the point that it stay on sometimes only for 10 min and restart....

I already swapped all memories, cleaned everything inside, reset SMC, check Northbridge plastic rivet, applied new thermal paste on all 3 cpu And install a fan control to check the temperatures and control fan speed. It seems to hold it for some more time but after a while it starts again... the temperatures seems to be ok as I saw in a lot of posts, but what get my attention, was that the heatsink B and memories (1-4) were extremely hot to touch but the sensor on fan control not shows nothing extreme....

I brought it to Apple authorized (as on my city don’t have apple store) and they said it was the logic board (easiest answer) without doing any test, just because the higher temperature to touch and justifying that they don’t have any part of this mac to swap for test purposes...

so my last call, was running ASD, and it gave me only one error (attached).

can you guys help me on this? Is the psu or cpu board the problem?

 

[tsialex]

hi Guys,

i need help with my Mac Pro as I already did a lot of research on net and nothing helped...

i have a Mac Pro 5.1
12 core 2.93
32gb of ram
Gtx 970 (now running a gtx 120 for testing purposes)
2 ssd for OS (mac and windows)

i most usually for 3dmax and autocad on windows bootcamp

it starts 4 months ago with randomly shutdowns, and become more frequently.
On windows, gave me BSOD with WHEA error code. First I reinstalled windows and hold for 3 weeks without shutdown. After started again, I swapped the gpu and tried MAC OS, but it shutted down too randomly. So I run hardwares test and everything was fine.

the shutdown become more frequently at the point that it stay on sometimes only for 10 min and restart....

I already swapped all memories, cleaned everything inside, reset SMC, check Northbridge plastic rivet, applied new thermal paste on all 3 cpu And install a fan control to check the temperatures and control fan speed. It seems to hold it for some more time but after a while it starts again... the temperatures seems to be ok as I saw in a lot of posts, but what get my attention, was that the heatsink B and memories (1-4) were extremely hot to touch but the sensor on fan control not shows nothing extreme....

I brought it to Apple authorized (as on my city don’t have apple store) and they said it was the logic board (easiest answer) without doing any test, just because the higher temperature to touch and justifying that they don’t have any part of this mac to swap for test purposes...

so my last call, was running ASD, and it gave me only one error (attached).
View attachment 885184View attachment 885184

can you guys help me on this? Is the psu or cpu board the problem?

Ip0C is a 12V rail power supply sensor, I'd start replacing it for a known good power supply.

Some of the ACBel power supplies fail a capacitor overtime, check if your PS is ACBel or Delta.

 

[Daniel Cardozo]

Ip0C is a 12V rail power supply sensor, I'd start replacing it for a known good power supply.

Some of the ACBel power supplies fail a capacitor overtime, check if your PS is ACBel or Delta.

So, Ip0C sensor is inside the power supply? Do you think that if I change the PSU, will resolve that? Or if it is ACBel, could I change only the capacitor?

 

[tsialex]

So, Ip0C sensor is inside the power supply? Do you think that if I change the PSU, will resolve that? Or if it is ACBel, could I change only the capacitor?

Sorry for not being clearer, if I remember correctly, power rail sensors are on underside of the backplane, not in the PSU.

Since Ip0C sensor is showing a fault with the rail, the first step is to test your Mac Pro with a known working PSU.

If your capacitor is just bulging, replacing it usually solve the problem, but if it's blown more components are damaged and you'll have to fully inspect the circuit.

 

[zedex]

@Daniel Cardozo..

I know your pain my friend. I had this exact same issue and I persisted with 'prayer' for far too long. I won't rehash my experience because you described it in adequate detail (..and it left me with PTSD).

My heartfelt advice - REPLACE the parts or shell before you log over 600 hours on diagnosis and repair for an issue that never self heals.

Like you I could not determine with adequate confidence whether the PSU or the mainboard/logicboard was at fault. After 3 months of intermittent, nearly impossible to predict shutdowns, I was lucky enough to find an affordable "shell" cMP 5,1 with low lifetime use and I am happy again.

Suffice to say - I am irritated that I didn't act a lot sooner.

RANDOM THOUGHTS RELATING TO THIS VEXING ISSUE..

I re-thermaled my CPU and GPU on 2 to 3 separate occasions before booting up with irrationally high hopes and delusional conviction that I'd fixed the problem.

I took a DEWALT blower vac to the internals (after removing PSU) in order to remove more dust from a cMP than any owner has ever managed to remove before.

I bought an external PCIe expansion chassis with dedicated PSU to stop the GPU from using the cMP 5,1 as a power source.

I tried 5 different GPUs to find a magic permutation that wouldn't trigger the dreaded instant shutdown.

Take it from me - you can't fix this problem because you can't isolate the 'cause', 'source' or 'original trigger'. Was it the VEGA 56 (8-pin + 8-pin) installation? Did my RX 480 fry my PCIe slot? Is it caused by software? Kext conflicts? iCloud sign-in? Fans management tools? Imperfect NVMe compatibility?

I still can't say categorically whether the PSU is faulty or the motherboard or both.

My PTSD is gradually subsiding as I slowly gain confidence in the reasonable expectation that the power on my 'new used' cMP 5,1 won't cut out without warning.

Hope this post prevents you from burying any more time in to this well of psychological destruction.


PS - How do I do that 'hardware test' that produced a report that helped Alex with a preliminary diagnosis?

 

[tsialex]

PS - How do I do that 'hardware test' that produced a report that helped Alex with a preliminary diagnosis?

It's an Apple internal tool for Mac Pro mid-2010 and mid-2012 diagnostics, ASD 3S149, that you can't legally download or have it, but you can find it in Russian sites or torrents.

 

[zedex]

Once again - grateful for your quick and helpful response Alex. Thank you