SFF.Network [SFF Network] Reports That High Pressure Heatsinks Can Damage Skylake

Intel's new Skylake processors are in the news again, and not for a good reason. According to PCGamesHardware.de, high pressure or mass coolers could damage the CPU package and/or socket. As you can see in the image above, the substrate of the processor package is significantly thinner, and being made of glass fibre (as are most PCBs), it can flex significantly more than the previous, thicker, packages. The pressure of a hefty cooler could push the CPU package into the socket and damage the pins!

Read more here.

 

[Phuncz]

Even though Intel CPUs don't seem to become any cheaper or losing ground to AMD, they still seem to have a focus on reducing production costs even more. So not only is the IHS adhesive an issue, now you also need to reinforce the package ? Maybe I should start an Intel lid upgrade service where I delid a CPU, apply a machined copper lid which strengthens the package and eliminates the issues with the adhesive.

 

[iFreilicht]

Wow, if this turns out to be true, I'd be very disappointed in Intel. You'd expect that they would want users to get the best performance out of their chips, but apparently, saving a dollar on the production is more important.

 

[EdZ]

Unless CPUs are failing at or below the clamping pressure proscribed by Intel in the mechanical specifications (222N, same as every consumer CPU socket since Socket 1156), this is likely due to heatsink manufacturers applying excessive force (or force not distributed evenly) and assuming the CPU can take it, rather than an issue with Skylake CPUs being unable to take the specified clamping load.

Or to put it another way: if you;re relying on the CPU package itself performing beyond specifications to prevent your heatsink from damaging the CPU and/or socket, then your have a serious issue with your heatsink clamping solution.

 

[jeshikat]

SPCR did some investigating two years ago into the subject of heatsinks with convex bases: http://www.silentpcreview.com/article1366-page2.html

Here's a quote from when they talked to a heatsink rep about the subject of convex heatsinks deforming the CPU IHS.

"It's true that most Thermalright heatsinks have convex bases, but this is extremely difficult to control in production, so they vary from almost flat to extremely convex. For obvious reasons, the issue is more likely to occur the more convex the base is and the higher the mounting pressure gets. Some coolers (from TR [Thermalright] and other vendors) feature mounting mechanisms that apply pressure well beyond sane levels (or at least allow to do so). This way, you're also more likely to run into said issue.

Emphasis added.

 

[iFreilicht]

Huh, who would've thought.
Was the specified clamping pressure reduced from Socket 1150 to 1151, does anybody know that?

 

[jeshikat]

From what I've been reading the specced clamping pressure is the same. The speculation is that older chips had more safety margin for heatsinks that exceeded the spec which is why it hasn't become an issue until Skylake with its thinner PCB.

 

[confusis]

Cryorig has released a statement. View it here:
https://smallformfactor.net/news/cryorig-responds-to-the-skylake-cooler-mounting-pressure-saga

 

[Phuncz]

A very interesting statement and it seems reasonable. It also goes to show how much thought goes into the design of just the mounting system.