• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Holy 95°C, Coolerman!

A

alxlwson

You Know Where I Live
2FA
Joined
Aug 25, 2013
Messages
10,615
Built my pops a new box. 7700k with CM 212Evo cooler. Fans are both operational. It's had several heat/cool cycles. 3000MHz RAM. Used IC Diamond TIM. 35°c at idle, IMMEDIATE 95°C at the slightest load, then BSOD. No CPU OC, only XMP from RAM with no extra optimizations. Strix Z270E mobo. Pulled the 212, pics are attached. I have tried reseat twice, same TIM layout. Everything points to bad CPU. Two hour drive to MC, so wanted to get opinions before the drive.
 
20170407_183052.jpg
20170407_183041.jpg
 
I'd never trust a 212 with that chip, but I'd also never run that chip stock... seems like you got a bad one.
 
silent-circuit said:
I'd never trust a 212 with that chip, but I'd also never run that chip stock... seems like you got a bad one.
Click to expand...

Can you elaborate on the 212? I told him to get an AIO at minimum, but being a retired sailor and very familiar with old ass liquid cooling systems...
 
Something has got to be wrong here. Is the bios defaulting to the wrong voltage?
 
alxlwson said:
Can you elaborate on the 212? I told him to get an AIO at minimum, but being a retired sailor and very familiar with old ass liquid cooling systems...
Click to expand...

It's not a bad HSF, it's just a budget HSF... and a 7700K deserves more. Like something from Noctua/Thermalright/Prolimatech if you're going with air.

For a parallel... it's like putting no-name oil from the gas station in a brand new Ferrari when for another $30 you could have had Mobil 1 full synthetic.

45454asasa said:
Something has got to be wrong here. Is the bios defaulting to the wrong voltage?
Click to expand...

Possibly this? Are you on most current BIOS version?
 
Does the heatsink get warm at all?

Could the mounting be a problem?

If a stock Intel HSF can keep a 7700 in check, I'd think a CM212 could keep a stock 7700K in check...
 
^Contact TIM pattern could be when you first mount since there is always more pressure exerted on the thing during initial mounting.

Not sure about this as well, but just throwing out possibilities.
 
What BIOS version is installed?

Go into the BIOS and restore defaults (even your RAM), then go into Windows and check temps again.
 
Does the Heatsink feel warm ? ( aka does it pull the heat )
does the CPU feel warma (aka is the temp real or a sensor bug)
 
Come on. The 7700K is ~90W at stock, maybe 130W tops OCed.
If a Hyper 212+ can keep an old i7 920 in check, it can keep a puny 7700K humming.

More than anything, on a non-delidded chip, it's the f*cking C/W of the TIM between the die and IHS that is the thermal limitation rather than anything outside the IHS.
 
Yea, chip is defuctive, board is defuctive or BIOS is set wrong.

That contact is fine between chip and cooler..... unless I cant see the clear sticker that is supposed to be peeled off :geek:.
 
paste looks really thick on both. you sure its screwed down correctly, correct screw posts used? 212 has different screw posts for amd vs intel, intel ones are shorter.
 
Last edited:
pendragon1 said:
paste looks really thick on both. you sure its screwed down correctly, correct screws/posts used? 212 has different screw posts for amd vs intel, intel ones are shorter.
Click to expand...

This was my thought also. A proper mount would leave large areas of direct metal-to-heatspreader contact where little paste would appear, as any paste is meant to get squeezed away to the sides from the mounting pressure. The pictures suggest a complete, intact coating of TIM sitting in-between the processor and the heatsink, which would cripple heat conductivity. Even a lowly cooler should not exhibit the symptoms described (immediate pegging to 95 degrees on load + BSOD).
 
It almost sounds like the interface between the IHS and HSF is air-gapped somehow due to mismounting.
Haven't had that happen to me for awhile now, except for during a 4770K test run at stock when I botched the HSF job and it pegged Tjmax under Prime95.
 
That TIM distribution looks awful, and really thick for some reason. I can attest that the 212 EVO is a fine cooler, so you might wanna start with the TIM. I would grab something like Arctic Cooling MX-4, spread an even layer over the CPU and try again.
 
You used IC diamond and put a shedload of it on there, its going to run hot. Just put a dot of the stuff that came with the 212 EVO and be happy with your now 20C cooler temps.

If you wanna use the IC stuff you need to use the little spreader spatula or an old credit card and get it on really thin. Not great for a direct contact heat pipe cooler like the 212. You need something that flows more easily.
 
Last edited:
Thinner application of TIM and check voltage settings in BIOS. Would be very very rare to have a bad cpu.
 
Board has the 0802 bios, iirc. Just updated it when I put the box together a few days ago. I'll try a TIM reapplication. I've never once had a bad application, but I agree that it's not the best. First time with IC Diamond as well. I was out of my Gelid, so picked that up as it was the only premium goop that MC had.
 
alxlwson said:
Board has the 0802 bios, iirc. Just updated it when I put the box together a few days ago. I'll try a TIM reapplication. I've never once had a bad application, but I agree that it's not the best. First time with IC Diamond as well. I was out of my Gelid, so picked that up as it was the only premium goop that MC had.
Click to expand...

IC diamond is a pain, I have a tube for emergencies but I use the thinner stuff. If I use the IC I have one of those little clear spreaders and it works OK, but I prefer Noctua NH-T1 or some of the similar stuff that don't have a cure time. Give is a go with a thin spread and see what happen!
 
I've got some IC, it's pretty finicky to do properly.
I had some cake together on a CPU, almost thought I broke it when I took it apart, it sits in a box somewhere now :)
 
pendragon1 said:
paste looks really thick on both. you sure its screwed down correctly, correct screw posts used? 212 has different screw posts for amd vs intel, intel ones are shorter.
Click to expand...
I agree looks like too much tim. But that wouldn't explain that high of a temp unless it wasn't secured properly.
 
jmilcher said:
I agree looks like too much tim. But that wouldn't explain that high of a temp unless it wasn't secured properly.
Click to expand...

IC diamond will do that, so will any of the other metal pastes if you get them too thick. They become an insulator. Temps act the same as if you used no TIM at all.
 
LigTasm said:
IC diamond will do that, so will any of the other metal pastes if you get them too thick. They become an insulator. Temps act the same as if you used no TIM at all.
Click to expand...

Then I'd just try some good old as5 or asc and call it a day.
 
jmilcher said:
Then I'd just try some good old as5 or asc and call it a day.
Click to expand...

Yeah, that was my suggestion to try a different TIM or a thin spread just in case, it only takes a few minutes and he won't waste his time going back to the store yet.
 
LigTasm said:
Yeah, that was my suggestion to try a different TIM or a thin spread just in case, it only takes a few minutes and he won't waste his time going back to the store yet.
Click to expand...

I just called him to let him know I was gonna come over and try some AS. He said he was almost to MC lol. He's getting an AIO. He said it did his 4930k well, so he decided to go get it.
 
alxlwson said:
I just called him to let him know I was gonna come over and try some AS. He said he was almost to MC lol. He's getting an AIO. He said it did his 4930k well, so he decided to go get it.
Click to expand...

Well I hope it works out in the end, let us know if it was a bad chip or not.
 
alxlwson said:
I just called him to let him know I was gonna come over and try some AS. He said he was almost to MC lol. He's getting an AIO. He said it did his 4930k well, so he decided to go get it.
Click to expand...

AS like Arctic Silver? The world has moved on, you should be using something nearly impossible to screw up like MX4.
 
Ocellaris said:
AS like Arctic Silver? The world has moved on, you should be using something nearly impossible to screw up like MX4.
Click to expand...

It's my go-to. Super easy to apply, and never once given me probs. I personally use Gelid GC Extreme normally though.
 
I hope he took the 7700K to get replaced as well. I can almost guarantee you that the cooler was not at fault nor was the TIM application. I am using a Coolmaster Hyper 212 Evo on my Ryzen 7 1700 at 3.8 GHz and 1.325v and the highest temp I saw was 68C with a full load of IBT AVX Very High.
 
ManofGod said:
I hope he took the 7700K to get replaced as well. I can almost guarantee you that the cooler was not at fault nor was the TIM application. I am using a Coolmaster Hyper 212 Evo on my Ryzen 7 1700 at 3.8 GHz and 1.325v and the highest temp I saw was 68C with a full load of IBT AVX Very High.
Click to expand...

Which is more likely... OP found a defective CPU, or the OP was using too much past or the wrong mounting hardware?
 
Ocellaris said:
Which is more likely... OP found a defective CPU, or the OP was using too much past or the wrong mounting hardware?
Click to expand...

In this case, defective cpu, based upon the tim pattern. It would not be the first out the box Intel CPU I have seen that is defective. I used to have a 6700K that had a defective memory controller and a customer who also had a defective memory controller on his 6700k as well. You may say, small sample size, I say, one person running into 2 of the same cpu's with the exact same problem is not small. I am not arguing this, if it is a defective cpu, one trip is easier than two.

Also, you cannot use the wrong mounting hardware on a 212 Evo and even too much paste would not cause an immediate jump to 95C under the slightest load.
 
ManofGod said:
In this case, defective cpu, based upon the tim pattern. It would not be the first out the box Intel CPU I have seen that is defective. I used to have a 6700K that had a defective memory controller and a customer who also had a defective memory controller on his 6700k as well. You may say, small sample size, I say, one person running into 2 of the same cpu's with the exact same problem is not small. I am not arguing this, if it is a defective cpu, one trip is easier than two.

Also, you cannot use the wrong mounting hardware on a 212 Evo and even too much paste would not cause an immediate jump to 95C under the slightest load.
Click to expand...

Using the wrong mounting hardware can absolutely fuck up temps. Also OP never confirmed what voltage his CPU was running at, some board increase CPU voltage when using XMP profiles (which is why I mentioned setting to defaults and checking temps)
 
Ocellaris said:
Using the wrong mounting hardware can absolutely fuck up temps. Also OP never confirmed what voltage his CPU was running at, some board increase CPU voltage when using XMP profiles (which is why I mentioned setting to defaults and checking temps)
Click to expand...

I said you cannot use the wrong mounting hardware with a Hyper 212 Evo otherwise, you could not mount it AT ALL! :) Also, his voltage would have to be at least 1.8v just to hit temps near instantly like that with just the slightest load. Resetting the bios would always be a good idea, though.

Oh well, not my CPU, if I am right or wrong really does not matter because I am not servicing the computer. OP, hope you get it resolved quickly.
 
Ah, come on...cake on some IC with your 212 and prove yourself correct :)
Aren't we on the [H]ard forums?
 
ChoGGi said:
Ah, come on...cake on some IC with your 212 and prove yourself correct :)
Aren't we on the [H]ard forums?
Click to expand...

Nope, not me, I have a perfectly good working machine that is on 24/7 at my work bench, which is where this 212 is being used. :) Besides, no real need to prove anything here, it is not my cpu nor am I servicing it. ;)
 
alxlwson said:
It's my go-to. Super easy to apply, and never once given me probs. I personally use Gelid GC Extreme normally though.
Click to expand...

The difference between various TIMs is about 10 C at most, and most are under 5 C difference. AS5 has the downside of requiring a curing time and being conductive in addition to being one of the more mediocre TIMs.
 
You must log in or register to reply here.
Back
Top