Skylake CPU's "extra" susceptible to damage??

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Hugo Drax

Diamond Member
Nov 20, 2011
5,646
47
91
Why I just ordered the Intel ts15a solution, they can be had for 26 bucks.. I just do not have the CPUs. I like the quick and easy push pin solution and it is a Intel heatsink, I do not over clock anyways.

I have one installed on a 3770k and they work well.
 

ehume

Golden Member
Nov 6, 2009
1,427
50
91
Says who?

Most heatsinks are incapable of warping the IHS, since they just attach with push pins or spring loaded screws.

Where is this spec about curved heat sinks and warped IHSs?
I review heatsinks. I get to see how they mount. Some are convex. Some are very convex. Some use screws with positive stops, and depend upon springs to put 50lbf on the CPU, and no more than that. However, some heatsinks have no screw stops. They depend on the frame to apply pressure. You can easily overdo it and bend the pins in your socket.

I have taken to taking pictures of the convexities, and commenting on the mounts. I believe users should know if a mount is dangerous before they buy.

BTW -- almost everyone has convexity on their mounts. Further, studies done around 2006 showed the more pressure you apply, the better the heat transfer. Later Skinnee Labs found that out. I have read a bitter comment by an OEM rep who commented that some manufacturers will apply excess pressure to make their heatsinks cool better. Caveat emptor.

BTW2 -- none of this is sanctioned by Intel, and no standards body dictates convexities or excess pressure.
 

LTC8K6

Lifer
Mar 10, 2004
28,523
1,565
126
I review heatsinks. I get to see how they mount. Some are convex. Some are very convex. Some use screws with positive stops, and depend upon springs to put 50lbf on the CPU, and no more than that. However, some heatsinks have no screw stops. They depend on the frame to apply pressure. You can easily overdo it and bend the pins in your socket.

I have taken to taking pictures of the convexities, and commenting on the mounts. I believe users should know if a mount is dangerous before they buy.

BTW -- almost everyone has convexity on their mounts. Further, studies done around 2006 showed the more pressure you apply, the better the heat transfer. Later Skinnee Labs found that out. I have read a bitter comment by an OEM rep who commented that some manufacturers will apply excess pressure to make their heatsinks cool better. Caveat emptor.

BTW2 -- none of this is sanctioned by Intel, and no standards body dictates convexities or excess pressure.
Thanks, but I already know all of that. :)
 

aigomorla

Cases and Cooling Mod PC Gaming Mod Elite Member
Super Moderator
Sep 28, 2005
17,398
579
126
Says who?

Most heatsinks are incapable of warping the IHS, since they just attach with push pins or spring loaded screws.

Where is this spec about curved heat sinks and warped IHSs?
There is no spec listed because they arent going to tell you that the heatsink is not flat.

This problem was discovered back during nikhsub's testing days on XS systems.

http://www.xtremesystems.org/forums/showthread.php?142998-The-anatomy-of-a-bow

This was a design flaw which actually showed improvement. The more mounting pressure one applied the better the contact was with the CPU + IHS.

My tests later on showed he was correct hence why i have that picture of the extreme mounting pressure and the results from it.

It was then shown on the Thermalright heat sinks.
(this was why they were the king, because they were probably the first to impliment the bow on heat sinks here)

http://www.xtremesystems.org/forums/showthread.php?147593-Does-this-heatsink-looked-BOWED

It was after that where most manufacturers started using a bow and firm retention hardware so they could get that extra couple of C performance.

Thanks, but I already know all of that. :)
Then why did you ask me to show documentation?

BTW -- almost everyone has convexity on their mounts. Further, studies done around 2006 showed the more pressure you apply, the better the heat transfer. Later Skinnee Labs found that out. I have read a bitter comment by an OEM rep who commented that some manufacturers will apply excess pressure to make their heatsinks cool better. Caveat emptor.
actually it was scott ... aka (nikhsub) who discoverd this.
He was the first to bring it to attention, and then i went into depth on it with the difference in mounting pressure.
This all predates skinnee which i am personal friends with. :)
 
Last edited:

LTC8K6

Lifer
Mar 10, 2004
28,523
1,565
126
Then why did you ask me to show documentation?
Because I was pretty sure you would not be able to. :)

I am right, so far. No specifications about curved sinks and warping the IHS.

You have linked to a couple of 8 year old threads where they do indeed talk about bad heatsinks, and they do indeed mention that the bad heatsinks might work better, but they also talk about lapping them to get them flatter. :)

BTW, I'm never going to use a visibly warped heatsink, and I am never going to try to warp an IHS to match such a heatsink. :D

Certainly nothing is perfectly flat, and I don't expect that.
 

aigomorla

Cases and Cooling Mod PC Gaming Mod Elite Member
Super Moderator
Sep 28, 2005
17,398
579
126
BTW, I'm never going to use a visibly warped heatsink, and I am never going to try to warp an IHS to match such a heatsink. :D

Certainly nothing is perfectly flat, and I don't expect that.
actually a grid line test can detect imperfections...

and yea.... 7 years ago.. man that was ages....
it was a age where we lapped everything to a mirror shine!
 

ASK THE COMMUNITY