Weird Temps on 3770k using H100
Tags:
Last response: in Overclocking
Hi guys,
Today i just ran into a big problem. My idle temps are resonnable (25-30 degrees on normal room temp) but my load temps are really high. I ran intel burn test (set to very high) with my h100 set to high and i went as far as 75 degrees ... I don't think this is normal ...
I'm using a push configuration and when i'm on load, my motherboard automaticly put the multiplier of the processor to 41 (4100MHz).
Thanks for the help
- The Brownie
Today i just ran into a big problem. My idle temps are resonnable (25-30 degrees on normal room temp) but my load temps are really high. I ran intel burn test (set to very high) with my h100 set to high and i went as far as 75 degrees ... I don't think this is normal ...
I'm using a push configuration and when i'm on load, my motherboard automaticly put the multiplier of the processor to 41 (4100MHz).
Thanks for the help
- The Brownie
More about : weird temps 3770k h100
Case : Corsair Carbide 500r (Stock fans on it, none added)
Motherboard : Asus P8Z77-V LX
Processor : Intel Core i7 3770K
Memory : G.Skill Ripjaws 2x4GB 1600MHz
CPU Cooler : Corsair H100 (top mounting in push configuration)
Video card : Asus GeForce GTX 660Ti
SSD : Intel 320 120GB
Hard drive : Seagate 5400rpm SataII (Don't know the exact model)
Power supply : Corsair TX750M
It shouldn't jump as high as 75 degrees with that cooler it really piss me off .... Bye bye overclocking
- The Brownie
Motherboard : Asus P8Z77-V LX
Processor : Intel Core i7 3770K
Memory : G.Skill Ripjaws 2x4GB 1600MHz
CPU Cooler : Corsair H100 (top mounting in push configuration)
Video card : Asus GeForce GTX 660Ti
SSD : Intel 320 120GB
Hard drive : Seagate 5400rpm SataII (Don't know the exact model)
Power supply : Corsair TX750M
It shouldn't jump as high as 75 degrees with that cooler it really piss me off .... Bye bye overclocking
- The Brownie
Related ressources
- Intel I7 3770K With H100 - Forum
- I7 3770K & H100 Temperatures - Forum
- I7 3770k oveclocking weird result - Forum
- OverClocked I7 3770k Temps - Forum
- Weird Heat Issue with Intel 3770k @4.3GHz - Forum
My Asus board does an auto-OC so i can't control the voltage ...
I redid the test checking in CPU-Z for the core voltage and it's switching between 2.32 and 2.40 when at 4100MHz, still on High. This time i ran Prime95 instead and the hottest core was at 76, sometimes 77. (Core 1 : 70 Core 2 : 76 Core 3 : 76 Core 4 : 75)
- The Brownie
Edit : Now the core voltage is stable at 2.232
I redid the test checking in CPU-Z for the core voltage and it's switching between 2.32 and 2.40 when at 4100MHz, still on High. This time i ran Prime95 instead and the hottest core was at 76, sometimes 77. (Core 1 : 70 Core 2 : 76 Core 3 : 76 Core 4 : 75)
- The Brownie
Edit : Now the core voltage is stable at 2.232
COLGeek said:
Only if you put too much thermal compound. The idea is to use as little as possible.I used the rice grain method (or pea method if you want). But i think i know my problem here. I used some Manhattan Silver Paste which uses 50% Siliconne, 20% Carbon and 30% Metal Oxide....
I have some Noctua NT-H1 laying around so i will try that.
I will buy some Arctic Silver during the week i think...
- The Brownie
Best solution
A Bad Day said:
Part of the reason why you have high temp is the 2.00+ voltage from that auto-OC, which is quite high. Either get a better cooling rig, or do the OCing yourself (with hopefully lesser voltage).So what voltage would you consider using on 4000MHz using 3770k and H100 ?
wdmfiber said:
Have you considered replacing the cheap thermal paste under the IHS? A young fella has an interesting video on youtube showing how he did it. Just search "Ivy Bridge IHS", it's 21:38 min. Even if you don't plan on doing it, ...still good to know.I will search for that tomorrow but what does the video treats exactly ?
- The Brownie
It's a modification to make Ivy Bridge CPU's like your 3770K run much cooler; way beyond anything you can do with different pastes between the CPU & H100.
Intel usually solders the heat speader to the chip. However on Ivy they cheaped out and used thermal paste. And to make matters worse they use a very cheap thermal paste. An i7-3770K makes less heat than say an i7-2700K (77 watts vs. 95 watts). But the H100(or any cooler) can't pull that heat away from the CPU after a certain point, as the heat transfer is just so poor. You could go as far to say the cheap tim(thermal interface material) is almost an insulator.
Intel usually solders the heat speader to the chip. However on Ivy they cheaped out and used thermal paste. And to make matters worse they use a very cheap thermal paste. An i7-3770K makes less heat than say an i7-2700K (77 watts vs. 95 watts). But the H100(or any cooler) can't pull that heat away from the CPU after a certain point, as the heat transfer is just so poor. You could go as far to say the cheap tim(thermal interface material) is almost an insulator.
wdmfiber said:
It's a modification to make Ivy Bridge CPU's like your 3770K run much cooler; way beyond anything you can do with different pastes between the CPU & H100.Intel usually solders the heat speader to the chip. However on Ivy they cheaped out and used thermal paste. And to make matters worse they use a very cheap thermal paste. An i7-3770K makes less heat than say an i7-2700K (77 watts vs. 95 watts). But the H100(or any cooler) can't pull that heat away from the CPU after a certain point, as the heat transfer is just so poor. You could go as far to say the cheap tim(thermal interface material) is almost an insulator.
Yeah i already knew that but i didn't think it would be that hot and that i couldn't overclock it at all ... I was wondering between 60-70 degrees on high using my H100 but i got some crappy temps.
Rockdpm said:
Your temps are fine. Ivy is slightly hotter in most cases over Sandy bridge but thats due to the design of the chip its self.Yes i know it but if i compare to most of the 3770k using H100, mine is hotter than theirs
- The Brownie
In the video the guys CPU was getting uneven temps across all cores(during a stress test). Core 1 was approaching 100 degrees C, other cores were over 30 degrees cooler.
He pulled(cut) off the IHS, you could tell the factory application of thermal paste wasn't even. After cleaning the CPU, he evenly(line method) applied a quality paste and the CPU was fixxed. Although his CPU was so bad that it prob qualified for a warranty replacement; but it's now better for healthy overclock.
He pulled(cut) off the IHS, you could tell the factory application of thermal paste wasn't even. After cleaning the CPU, he evenly(line method) applied a quality paste and the CPU was fixxed. Although his CPU was so bad that it prob qualified for a warranty replacement; but it's now better for healthy overclock.
Ok so a little update for my temps here. I replaced my old thermal paste by some Noctua Nt-H1 and it seems to be okay now. I got 4.2GHz running with 1.2vcore and i'm under 70c !!! I guess that will do the job for now but if i see my temps going way too up i'm going to replace the thermal paster inside.
He pulled(cut) off the IHS, you could tell the factory application of thermal paste wasn't even. After cleaning the CPU, he evenly(line method) applied a quality paste and the CPU was fixxed. Although his CPU was so bad that it prob qualified for a warranty replacement; but it's now better for healthy overclock.
I just saw the video and as i said, if my temps get way too high i'm going to do it. He mentionned in the video that he couldn't use diamond paste because i will scratch the die. Is it going to do it with arctic silver ?
Thanks for the help guys .. now choosing the best answer :3
- The Brownie
wdmfiber said:
In the video the guys CPU was getting uneven temps across all cores(during a stress test). Core 1 was approaching 100 degrees C, other cores were over 30 degrees cooler.He pulled(cut) off the IHS, you could tell the factory application of thermal paste wasn't even. After cleaning the CPU, he evenly(line method) applied a quality paste and the CPU was fixxed. Although his CPU was so bad that it prob qualified for a warranty replacement; but it's now better for healthy overclock.
I just saw the video and as i said, if my temps get way too high i'm going to do it. He mentionned in the video that he couldn't use diamond paste because i will scratch the die. Is it going to do it with arctic silver ?
Thanks for the help guys .. now choosing the best answer :3
- The Brownie
Bromeh said:
My Asus board does an auto-OC so i can't control the voltage ...I redid the test checking in CPU-Z for the core voltage and it's switching between 2.32 and 2.40 when at 4100MHz, still on High. This time i ran Prime95 instead and the hottest core was at 76, sometimes 77. (Core 1 : 70 Core 2 : 76 Core 3 : 76 Core 4 : 75)
- The Brownie
Edit : Now the core voltage is stable at 2.232
Lower the VCORE to 1.2 for a 4.1 GHz OC, and do it in the BIOS
Related ressources:
- ForumI7 3770k Temps @ 4.6Ghz
- ForumI7 3770k OC'd temps
- ForumHigh temps when overclocking i7 3770k with corsair h80
- ForumH100 high temps ?
- ForumWeird Bus Speed and CPU not underclocking - 3770k
- ForumCan I use a corsair h80 or h100 ?
- ForumBest fan setup for corsair h100 , i7 3770k , corsair 550d case
- ForumArctic mx-4 VS H100 stock paste
- Forum2700k vs. 3770k and Noctua NH-D14 vs. Corsair H100
- Forum3770k vs 2700k vs 2600k
- ForumIs this heat normal? Gigabyte z77 ud5h I7 3770k 4.7 ghz 1.275v
- ForumCooler for OC'ing 3770k
- ForumI3770k & H100
- ForumI7 3770k 4.6GHz and 1.336V reaching 70* under max load
- ForumShould I connect all my case fans to a PWM slot?
- More resources
Read discussions in other Overclocking categories
!
2.232 is impossible ahah