Buildapc

3790 readers

20 users here now

founded 1 year ago

MODERATORS

Lemmenmo@lemmy.world

Core Distance to TjMAX randomly spikes to 100°C in OCCT (lemmy.ml)

submitted 7 months ago by GolfNovemberUniform@lemmy.ml to c/buildapc@lemmy.world

8 comments fedilink hide all child comments

Recently I tried to clean my PC with an I5 11400F CPU. I removed the GPU and wanted to remove the CPU cooler too but couldn't do it. I tried though and I think I might have damaged the motherboard with a screwdriver in a few places during the process (don't ask how). Can it be the reason for the spikes? Like could I damage some sensors and now they sometimes fail and show 0°C (which is what 100°C to TjMAX means)? There are no other visible issues with the PC though so I guess it can be a just a bug. Btw I can't add screenshots because of some weird Jerboa bug

top 8 comments

sorted by: hot top controversial new old

[–] Just_Pizza_Crust@lemmy.world 7 points 7 months ago (2 children)

Oh boy. I'm not 100% sure about Intel CPUs, but typically the sensors are inside of the metal casing with the silicon. You might've damaged the frame of the cooler so it's not making contact with the CPU and causing the CPU to spike in temp. I'd recommend reaching out to someone who has experience with building PCs and see if they can help you get the CPU cooler off properly, replace the thermal paste, and hopefully get it remounted properly.

Don't stress your CPU for now because it will cause long-term damage that could fry the CPU/mobo.

[–] kernelle@0d.gs 4 points 7 months ago (1 children)

I agree with this, you probably dislodged or loosened the cooler. Which needs firm pressure to work properly, causing one of your cores to spike in temperature.

[–] GolfNovemberUniform@lemmy.ml 1 points 7 months ago* (last edited 7 months ago) (1 children)

Distance to TjMAX is a value opposite to temperature. If it's 100°C, it means the core's temperature is 0°C (at least for most of the Intel CPUs). The core temperature doesn't spike to 0 though...

[–] kernelle@0d.gs 1 points 7 months ago (1 children)

So what you're suggesting is that it's not the actual temperature but rather a sensor issue? Could be, the problem is this:

remove the CPU cooler too but couldn't do it

CPU coolers have one job, and there is only one way to properly attach them. Any tempering with that system will have big consequences. If redoing that doesn't fix the issue, and you notice no other problems at all, might as well be a sensor issue.

[–] GolfNovemberUniform@lemmy.ml 2 points 7 months ago (1 children)

Well then I guess I don't have a PC anymore because I can't get it repaired any time soon. Thank you for help though

[–] kernelle@0d.gs 1 points 7 months ago (1 children)

You sound like you'd be able to figure it out, I'm sure if you look up your cooler on the web you'll find video and instructions. If you have cooling paste available I'd say go for it.

[–] GolfNovemberUniform@lemmy.ml 1 points 7 months ago* (last edited 7 months ago)

The problem is that the cooler is absolutely massive (because cheap materials = bad heat conductivity = more volume required) and the screws are unaccessible unless I remove the motherboard and its radiators (and I'm not even sure in that). I can try but I know I may not be able to reconnect the cables (the case is not cable management friendly at all). And also I have a physical disability so I can't do anything like that myself anyways. Oh and the cooler is some kind of noname ODM manufacturer I think so finding instructions for it is hard or even impossible

[–] GolfNovemberUniform@lemmy.ml 1 points 7 months ago

The temperatures don't spike and are completely normal though, even in stress tests. Maybe OCCT calculates the value and it fails because of freezes caused by stress tests? I don't even see the 100°C on the live sheets. It just appears in the max column smh and stays there all session (as it should)