I was building my rig and before putting everything together I just wanted to test the parts and I used the stock cooler to boot the MB with the CPU and just one RAM.
The problem is: the stock fan was not well connected (my mistake) and when I went to the BIOS the CPU temp was 99.5C!!!
I turn it off REALLY fast. Removed the stock crap and clean it up the TIM on the top (the original stock cooler TIM was hard to remove, like it had "cooked by heat" but the Tcase was looking shinny, no weird black spots after cleaning it).
So, I added my CM 212+ correctly and everything is fine now (35C on idle).
The computer looks fine and I already run the Prime95 a couple of times to see if any calculation fails, but it didn't. So, my questions are:
1) How can I be sure that this "short period of cooking time" really didn't damaged my CPU at all? How can I test if the CPU was damaged?
2) What is the max temp my cores can go? On Prime95 I am hitting 72C sometimes, but I am reducing the CM 212+ coolers speed to keep them quiet. I can increase them if this temp is not good, but I just hit that temp running Prime95 for a while, like 10-20min. The worst case was running Prime95 for a long time and the Core1 got 78C and the Package got 79C.
3) When I run Prime95 my Core1 is ALWAYS far hotter then the others, like a difference of 5C. This is common?
Running 99.5C for a short period of time will not do any real damage.
As for your new cooler, that is still hot for stock clocks. REALLY HOT. The max safe temp for that CPU is 72.6C. Make sure you have enough thermal paste or if you have too much, make sure you have the correct amount and that everything is mounted correctly. Do you have any fans in your case? You should be in the 50 something Celsius with that cooler at stock clocks.
One of my cores is 7C hotter than the rest and that just happens with CPUs. Nothing to worry about.
1) I have a ASRock Z68 Extreme4 (2nd gen) and I changed my FAN config on CPU to "Automatic" and "Level 1" because of the silence. I can easily increase to Level 2 or Level 3 because the noise is not bad, but Level 9 (full RPM) is too noisy;
2) I have PUSH/PULL on the CM212+ using another CM cooler (R4-BMBS-20PK-R0);
3) I used a good TIM, the Shin-Etsu G751 one;
4) I OC'ed to 42X (4x CPU bins). It is a 2600, not a 2600K, so it is limited to 4 CPU bins, so I just did it. When I run Prime95 it runs at 3.9GHz on all 4 cores;
5) On idle and on normal operations (even gaming) I am running good. 30-35C on idle and 50-60C on load;
6) I have a good active cooling chassi, the Storm Scout one;
7) I am using HWMonitor to get those temperatures, and I am reporting the CPU cores and package temperatures. If I read the CPUIN from the MB, it says 53C on the same moment I've got 79C on package.
I know that 78-79C on Package is bad. But should I worry if that occurs so rarely?
But my main concern is still how can I double check if I didn't damaged my CPU on the short cooking time
What is the CPU voltage? That is pretty high especially for two fans and a good case. I would make sure everything is down nice and tight and that there is not one side not tightened down or something.
Having your CPU hot for just a little while won't hurt it as long as you do not do it often! What are your temperatures in the BIOS?
How much thermal paste did you use? About the size of a pea is the correct amount. Also, there should be little trenches in the bottom of the Hyper 212 copper pad, put a little paste in those if you didn't. If you did all of that, I would make sure the HSF is on correctly.
I added a small drop on the middle of the CPU.
I know it was not "too much" because it is not leaking or visible on the sides.
Do you think I should remove the CM212+, clean everything and put it back again?
Might be too little then. Hyper 212+ is kind of a weird one with the heat pipes. Need to fill those gaps in. If you remove it, you should see a pretty thin spread of paste under all the heat pipes. You may want to look up guides on how to mount this cooler, as 72C is very hot and especially at that voltage (assuming it's correct).
You should start with totally stock and check how everything runs before overclocking.