Intel i5 3570k is reaching 103 celcius in Idle mode

[MoGun]

My Intel i5 3570k is reaching 103 celcius in Idle mode. Plz help me cant eveb play a single game.
purchased Cabin, Mother board, and Processor befor 3 years. Smps n graphics are new.

My Specs:- Operating System - Windows 10 Home 64-bit

CPU - Intel Core i5 3570K @ 3.40GH 101 °C Ivy Bridge 22nm
Ivy Bridge 22nm Technology

RAM - 8.00GB Single-Channel DDR3 @ 532MHz (7-7-7-19)

Motherboad - ASRock H61M-HG4 (CPUSocket) 36 °C


Graphics - Acer X193HQ (1366x768@60Hz)
4095MB NVIDIA GeForce GTX 970 (MSI) 33 °C


Storage - 931GB Seagate ST1000DM003-1CH162 (SATA) 37 °C

SMPS - CORSAIR 650



other detail about CPU


CPU
Intel Core i5 3570K
Cores 4
Threads 4
Name Intel Core i5 3570K
Code Name Ivy Bridge
Package Socket 1155 LGA
Technology 22nm
Specification Intel Core i5-3570K CPU @ 3.40GHz
Family 6
Extended Family 6
Model A
Extended Model 3A
Stepping 9
Revision E1/L1
Instructions MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, Intel 64, NX, VMX, AES, AVX
Virtualization Supported, Enabled
Hyperthreading Not supported
Fan Speed 2011 RPM
Bus Speed 99.8 MHz
Stock Core Speed 3400 MHz
Stock Bus Speed 100 MHz
Average Temperature 81 °C
Caches
L1 Data Cache Size 4 x 32 KBytes
L1 Instructions Cache Size 4 x 32 KBytes
L2 Unified Cache Size 4 x 256 KBytes
L3 Unified Cache Size 6144 KBytes
Cores
Core Speed Multiplier Bus Speed Temperature Threads
Core 0 1596.2 MHz x 16.0 99.8 MHz 79 °C APIC ID: 0
Core 1 1596.2 MHz x 16.0 99.8 MHz 80 °C APIC ID: 2
Core 2 1596.2 MHz x 16.0 99.8 MHz 81 °C APIC ID: 4
Core 3 3791.0 MHz x 38.0 99.8 MHz 82 °C APIC ID: 6

 

[Karadjgne]

Is the pc clean of dirt and dust?
Does the fan on the cpu cooler actually work?

 

[outlw6669]

If the CPU is really idling at 100°C+ you need to turn off the computer immediately!
The CPU cooler is likely no longer seated properly and will need to be reinstalled.

Take off the cooler, remove the old TIM fully from both it and the CPU heatspreader using isopropyl alcohol and a lint free cloth (a coffee filter will work fine) and reinstall the cooler using a new coating of TIM.
If you are using a stock intel cooler, ensure that the cheap plastic retention legs are not damaged and are able to properly engage; it would also be wise to consider an inexpensive aftermarket tower cooler with a more robust mounting solution.
After reinstalling the cooler, turn on your computer with the side panel off and ensure that the fan on the cooler is still working.
If it is not, do not use the computer until you are able to obtain and install a new cooler.

If you run into any further issues, just post back.

-outlw

 

[MoGun]

It is clean and It work good

 

[DSzymborski]

It is clean and It work good

Well, it's clearly not working well or your temperatures would be lower - a cooler works by basic physics, so if it's working, it *will* remove heat. You need to remove it, apply thermal paste, and verify everything's attached firmly and that the fan is moving.

 

[Bo Lee]

I think you have a bad sensor. Generally a PC will shut down way before that, typically between 75 and 80 degrees and a PC will shut off by design to prevent your CPU from melting down.

 

[synphul]

I would try checking with another program like realtemp. If temps fluctuate then it's probably not a sensor, usually sensors get stuck at a specific value and way out of bounds like 150c or something ridiculous on all cores. Try another cooler, even a budget cooler like a 212 evo or cryorig h7 would be solid for that cpu. The cpu won't thermal throttle at 75-80c, throttle temps is around 100-105c in which case the pc won't just shut down. What it will do is lower the cpu speed in an attempt to reduce temps back within safe operating ranges.

If using the stock cooler it may have gotten bumped and loosened the plastic retaining pins or one of the pins may have broken or gotten to the point it no longer holds the cooler down tight to the cpu. A new cooler should resolve that issue.

 

[Bo Lee]

I would try checking with another program like realtemp. If temps fluctuate then it's probably not a sensor, usually sensors get stuck at a specific value and way out of bounds like 150c or something ridiculous on all cores. Try another cooler, even a budget cooler like a 212 evo or cryorig h7 would be solid for that cpu. The cpu won't thermal throttle at 75-80c, throttle temps is around 100-105c in which case the pc won't just shut down. What it will do is lower the cpu speed in an attempt to reduce temps back within safe operating ranges.

If using the stock cooler it may have gotten bumped and loosened the plastic retaining pins or one of the pins may have broken or gotten to the point it no longer holds the cooler down tight to the cpu. A new cooler should resolve that issue.

Depends on the MB and BIOS. My previous Asus would shut down around 85 degrees. From what I have read about his processor, 105 degrees wouldn't be throttling. It would be melting, LOL. That CPU typically stays 70 and below under a load.

 

[Karadjgne]

Intel cpu's will start considering throttling over @70°C, will generally start throttling @85°ish, and are in full throttle by 100°ish, depending on actual cpu will generally shut down if throttle isn't working @100°+.

Of course that's not including cooling considerations, entirely possible that with aios or improper airflow and high cpu voltages, by the time the cpu reaches 85°+, the VRM's/Northbridge are well overheated and are throttling back/shutting the mobo down

 

[Bo Lee]

Intel cpu's will start considering throttling over @70°C, will generally start throttling @85°ish, and are in full throttle by 100°ish, depending on actual cpu will generally shut down if throttle isn't working @100°+.

Of course that's not including cooling considerations, entirely possible that with aios or improper airflow and high cpu voltages, by the time the cpu reaches 85°+, the VRM's/Northbridge are well overheated and are throttling back/shutting the mobo down

Interesting. The Intel site made it seem that the high 80s was the max safe temp for that unit. But yes, with that particular Asus board, it also had horrible VRM cooling characteristics.

 

[Karadjgne]

http://www.tomshardware.com/forum/id-1800828/intel-temperature-guide.html

This'll explain in further detail. For most users, the Intel assumed safe range is 70°C. Above that it's touchy as each individual cpu is different, different silicon batches, voltages etc. So you could start throttling at 71 or 92, but if throttle is going to happen, it'll be somewhere above 70. By the time you get close to 100,its almost guaranteed you'll be in full throttle mode, and if the cpu cools enough, speeds will rise, and this fluctuation will keep happening until the cpu has had enough, and decides to just shutdown. But there's no exact temp when this happens.

 

[mdd1963]

if your cpu is at 100C, your heatsink will be too hot to touch...

Is it?

What temps is HardwareMonitor showing?
http://www.cpuid.com/softwares/hwmonitor.html
(Are we sure it's thermal compound is still there, and not solidified over the years? Get a new tube of Arctic Silver 5 or equivalent for $5 or so...)

 

[synphul]

My 4th gen i5 doesn't throttle at 85c. Might have a look here. 3570k throttle temp is 105c, not 'meltdown'. These sound like assumptions rather than facts, specs or figures.

http://www.tomshardware.com/forum/id-1800828/intel-temperature-guide.html

If a cpu is at 80c or 100c it doesn't mean the temp of the heatsink matches it. Those core temps are internal, the hottest temps. The ihs would be a few degrees cooler and by the time the thermals are absorbed through the base of the cooler and cooled by the active fan blowing air over the fins the heatsink won't be 100c.

It's always possible to have a board that acts out of the ordinary or for a board to react to overheating vrm. That's a different issue than the cpu overheating and it's quite possible for a cpu running at stock to overheat at the cpu itself while the vrm are fine. Vrm would be more likely to overheat during stresses like overclocking where higher than normal voltage is applied to the cpu (and coincidentally going through the vrm).

Safe ranges in the abstract or more to do with safely operating the cpu for longevity since excessively high temps over long periods of time can cause a cpu to die sooner rather than later. Ill effects like electromigration result from things like ecessive current and excessive heat.

Here's the results of puget systems testings and a snippet from their conclusions.
"The results of our testing can pretty much be summarized with the following three points:

Modern Intel CPUs run at full speed (including the full Turbo Boost allowed based on the number of cores and workload) all the way up to 100 °C
Even after the CPU hits 100 °C, the performance is not greatly affected until the CPU spends about 20% of the time > 99 °C
While stock cooling only causes around a 2.5% drop in performance, even a budget after market cooler will dramatically improve CPU temperatures"

https://www.pugetsystems.com/labs/articles/Impact-of-Temperature-on-Intel-CPU-Performance-606/

There's a lot of misinformation running around this thread about the pc shutting down unexpectedly on cpu's which have throttle capabilities, cpu's reaching throttle at between 70-80c etc.

Unless the thermal paste is completely dried out or missing, highly doubtful that new or 'special' thermal paste will solve a major overheating problem. Most thermal pastes cool within 1-3c of each other. If a reapplication of thermal paste improves things then it's likely because the old paste was gone, the cooler was knocked loose or not attached properly or some other issue that was corrected with the re-installation of the cooler rather than what paste someone upgraded to.

 

[USAFRet]

I had a 3570k running, unnoticed, at 95-100+ for ~2 weeks.
The HTPC, playing movies and music. Noticed no performance burps.

Why was it that hot? Misinstalled stock cooler. Simply wasn't making full contact.
Reinstall, and today, 3 months later, (right now watching a movie) it is at ~42C.

How did I eventually notice? The fan noise seemed a little bit louder.
It didn't melt, it didn't explode.

 

[Karadjgne]

My 3570k started throttling in the lower 90's, shutdown at 104 peak. Of course that was fully stock and ran occt w/linpack and disabled the auto shutoff. Incidentally, that was also the only time I attempted that, was more for curiosity. Swapped cooler out, added some OC, dropped the vcore dramatically and now only sees over 50 if I get lazy on cleaning.