I have a very frustrating problem and I need your guys help to fix it.
I bought an imac 5k that I upgraded it's CPU, HDD, RAM.
This is my second one i have upgraded and the second one that keeps restarting randomly, sometimes once a few days, other times once a week or less.
I have put in an I7, EVO SSD and Apple Ram.
I have tried testing it with less and more ram and it still restarts our of the blue.
The only connection between the two imac upgrades is the Thermalpad Arctic( 1,5mm which has a 6 W/mk thermal conductivity, its the best I could find ) I have used to cool the GPU.
I am running multiple flash related software/streamers that are very heavy on the CPU and GPU, and some photoshoping.
It always does restart when in use, but that's because when I am on it/ rather say, my wife is on it, she uses it for work but i don't think it restarts during normal use.
The fans are loud but i did not notice any pattern in which when it does restart, the fans are louder than before. They stay pretty much over 2000 rpm, which is a huge difference than the I5 i had before, doing the same kind of work, but I read that this is normal considering the I7 is much hotter. Thermal paste used was Arctic Silver 5.
What makes it difficult is that it is very hard to monitor the iMac because I am not working on it and sitting there and waiting for the restart to happen is just nuts. I was thinking I could observe the temps but I am sure that is not a feasible way to see if temps are to blame.
I went to library>logs>but couldn't find any readable log about this kernel panic.
Is there a software that logs the temperatures so after it restarts I can compare them to see if heat is to blame?
I have gone through different set of rams, external USBs, OS versions, you name it, nothing seems to solve it.
How to go on and investigate this issue?
Thank you very much!
Dan
LE:
I managed to make it restart. I had the usual amount of work that i topped with 4 4k videos playing at the same time on youtube. Here is a remaining pictures from team viewer at the moment of the restart.
Do you think that's enough ? I read that the values can go over 100 C without any problems.
I am using iStats first one is CPU, GPU-Die, Ram, CPU Fan Rpm
Best Answer
Problem solved. I believe the 1.5mm thermal pads were too thick on the GPU rams and the cooler didn't make enough contacts on the GPU chip. I went to an Apple service provider and asked them nicely if they could sell me the thermal paste they use and they were very kind and gave me a bunch of 2CL syringes with paste manufactured by Shin Etsu Microsi - microsi.com, that is what they use on apple products and it did the job.
I put a small pea shape bit of paste in the middle of the GPU memory chips, they are 8 and I used 1CL for every chip and when you put the cooler ontop it spreads it evenly. You could use less and it won't leak when you push the cooler in but i wanted to be safe and fill the gap between the cooler and the GPU chips.
Be careful, this paste is thick, its a pain to spread it evenly. You have to get a credit card and put paste on the top edge and swipe it with a bit of force so you can apply it in one take because its a mess to spread it with your finger, its like it won't stick to the chip.
I now have 56 Degrees on the CPU and 55 degrees C on the GPU die, lightly used and never went over 75 degrees C no matter the stress of the system. Measured with iStat Pro.
I hope it will help someone even though nobody cared to give me a hand.
Peace!