i7-2700k + Noctua NH-D14

[KTZeen]

Without seeing your cpu and heatsink, it's hard to say for sure. You could take a razor blade or a straight ruler and check the contact surfaces of your heatsink and your cpu for concavity/convexity.

Here is the pic of my heatsink and cpu with the thermal paste and the heatsink without the paste after i cleaned it.

I do notice there are certain areas of the cpu and heatsink that the paste isn't covering in the inside area.

http://imageshack.us/photo/my-images/825/heatsink.jpg/

 

[muddocktor]

Yeah, 1 mount didn't get a very good spread for sure. As for the marks; I've seen that happen before with new processors and I think that it's because the IHS still has sharp edges to it from manufacturing on it. After your've mounted the heatsink several times I find that the edges on the IHS don't leave marks on the base any more. My 930 processor in my test system used to do this when it was newer, but now it doesn't leave any marks on the bases any more. Of course, it has seen quite a few remounts of heatsinks too.

 

[KTZeen]

Thanks for the info muddocktor, is there a reason the mount didnt get a very good spread? I applied the dot of paste right on the center but it appears it didnt spread evenly.
Here is another result of my OC to 4.4ghz at 1.272v. Ran prime95 for roughly 52 mins.
http://imageshack.us/photo/my-images/607/oc44ghztemp1272v.jpg/

 

[ehume]

Those marks look just like the ones on my D14 and my Megahalems.

Looks like you need to make a little bigger bleb. It takes practice. But viscosity and bleb size control is why I use Gelid GC Extreme.

 

[TsunamiJuan]

I use bobnovas thermal paste method 3 very thin strips crossing each other in the middle kinda like an asterix. It yeilds a very even spread. Throw some dots inbetween the outer edges on GPU's to fill in the spread a little, but cpu's dont generally need the dots. I post a pic next time i am mounting pots.

 

[KTZeen]

There has to be something wrong...... I mean right now I am running 1.27v at 4.4GHz ran prime95 for 1.5 hours and the temps are 61-66-66-65.

I practically double my temps +2c....

 

[KTZeen]

Thanks everyone for ur inputs and help. From this I have learned alot and now my skills in mounting heatsinks went to the intermediate level ahhahaa.

I have tried reversing the heatsink pulling air from the back of the case and swapped the push fan to the 2000 rpm CM R4 but the temps actually increased for reasons I know why.

I have tried single line, double line, dot method and I can say the dot method gave me the best results.

I am now OC'ed at 4.4ghz at 1.27v . Although my bus speed is 99.8 instead of 100 so my top clock speed is 4390.5mhz instead of 4400. My temps are now 57-61-62-60 (realtemps) 56-60-61-59(hwmonitor) on prime 95.


Thanks again everyone!

 

[muddocktor]

KTZeen, there is a setting in your mobo bios to stop that variation in bus speed. The option is called "spread spectrum" and turn that off and it will run at the correct clock speed. That is something the FCC foisted off a long time ago on the computer industry to keep from having RF radio interference, IIRC and it's not needed, especially when overclocking.

 

[KTZeen]

KTZeen, there is a setting in your mobo bios to stop that variation in bus speed. The option is called "spread spectrum" and turn that off and it will run at the correct clock speed. That is something the FCC foisted off a long time ago on the computer industry to keep from having RF radio interference, IIRC and it's not needed, especially when overclocking.

Thanks alot! Didn't know about it, now its running at stated OC speed.

 

[KTZeen]

If anyone is interested I am currently trying to achieve the lowest vcore possible on a stable 4.6Ghz core speed. Here's a link of prime95 around 35 minutes.
I am using 1.29v as vcore but on load it got boosted up to 1.296v.
Temp were 61-66c on load for realtemp.

http://imageshack.us/photo/my-images/18/oc46ghztemp129v.jpg/

My case is fully closed, according to other forum inquiries of the i7, because of hyper threading, the temps will be a little higher than i5-2500k.

 

[KTZeen]

Here is another update, this time at 4.8ghz at 1.32v for vcore. I am unsure how long I am suppose to run prime95 for to know if the system is stable. Yet again ran for around 30 minutes.

http://imageshack.us/photo/my-images/27/oc48ghztemp132v.jpg/

Can anyone give me more info on that please.

 

[PolRoger]

Seems like a pretty decent chip... 48x with ~1.33v (load) is better than plenty of others that I've seen posted on the forums... if your chip continues to scale with vcore it might be able to do 50X with ~1.4v or less.

 

[KTZeen]

Just an update, I tested 1.31v at 4.8ghz and prime 95 gave me bsod.
Went back to 1.32v and crashed after 45 min bsod
I wont be going any higher since I dont want my temps over 73c. Its just me... I know the chip can handle much higher.
I guess 4.6GHz is the highest for me.

 

[KTZeen]

One quick question. The top of the nh-d14 heatsink where the heatpipe sticks out, are they suppose to be warm or hot when running prime 95?

I touched them but they were cold.

 

[KTZeen]

Here is yet another test, 4.6ghz at 1.275v vcore. Unsure why the vcore is low but temp is high. Compared with i5-2500k they get a bit lower temps than me at 1.35v vcore with the same speed. I guess it all comes down to if u got a good chip or not.

http://imageshack.us/photo/my-images/846/oc46ghztemp1275v.jpg/

Ran prime95 for over 1 hour and seems stable with temps from 60-66c.

 

[TsunamiJuan]

One quick question. The top of the nh-d14 heatsink where the heatpipe sticks out, are they suppose to be warm or hot when running prime 95?

I touched them but they were cold.

The tops of the heat pipes usually do not get very hot, which is optimal. This means the substance inside is rising to the top as a gas, releasing its heat and returning to a liquid and dropping back to the base of the heatpipe to repeat the process. This is what makes heat pipes so efficient and thermal exchange.

If you looked at a heat pipe with a finely calibrated IR camera you would be able to see the temperatures in the heat pipe drop the further up the pipes go.