Intel Skylake Processors Can Bend Under Pressure, Damage CPU and LGA Socket

Subject: Processors | December 4, 2015 - 11:35 PM |
Tagged: Skylake, Intel, heatsink, damage, cpu cooler, Core i7 6700K, Core i7 6600K, bend, 6th generation, 3rd party

Some Intel 6th-gen "Skylake" processors have been damaged by the heatsink mounts of 3rd-party CPU coolers according to a report that began with pcgameshardware.de and has since made its rounds throughout PC hardware media (including the sourced Ars Technica article).

View Full Size

The highly-referenced pcgameshardware.de image of a bent Skylake CPU

The problem is easy enough to explain, as Skylake has a notably thinner construction compared to earlier generations of Intel CPUs, and if enough pressure is exerted against these new processors the green substrate can bend, causing damage not only to the CPU but the pins in the LGA 1151 socket as well.

The only way to prevent the possibility of a bend is avoid overtightening the heatsink, but considering most compatible coolers on the market were designed for Haswell and earlier generations of Intel CPU this leaves users to guess what pressure might be adequate without potentially bending the CPU.

Intel has commented on the issue:

"The design specifications and guidelines for the 6th Gen Intel Core processor using the LGA 1151 socket are unchanged from previous generations and are available for partners and 3rd party manufacturers. Intel can’t comment on 3rdparty designs or their adherence to the recommended design specifications. For questions about a specific cooling product we must defer to the manufacturer."

It's worth noting that while Intel states that their "guidelines for the 6th Gen Intel Core processor using the LGA 1151 socket are unchanged from previous generations", it is specifically a change in substrate thickness that has caused the concerns. The problem is not limited to any specific brands, but certainly will be more of an issue for heatsink mounts that can exert a tremendous amount of pressure.

View Full Size

An LGA socket damaged from a bent Skylake CPU (credit: pcgameshardware)

From the Ars report:

"Noctua, EK Water Blocks, Scythe, Arctic, Thermaltake, and Thermalright, commenting to Games Hardware about the issue, suggested that damage from overly high mounting pressure is most likely to occur during shipping or relocation of a system. Some are recommending that the CPU cooler be removed altogether before a system is shipped."

Scythe has been the first vendor to offer a solution to the issue, releasing this statement on their support website:

"Japanese cooling expert Scythe announces a change of the mounting system for Skylake / Socket 1151 on several coolers of its portfolio. All coolers are compatible with Skylake sockets in general, but bear the possibility of damage to CPU and motherboard in some cases where the PC is exposed to strong shocks (e.g. during shipping or relocation).This problem particularly involves only coolers which will mounted with the H.P.M.S. mounting system. To prevent this, the mounting pressure has been reduced by an adjustment of the screw set. Of course, Scythe is going to ship a the new set of screws to every customer completely free of charge! To apply for the free screw set, please send your request via e-mail to support@scythe.com or use the contact form on our website."

View Full Size

The thickness of Skylake (left) compared to Haswell (right) (credit: pcgameshardware)

As owner of an Intel Skylake i5-6600K, which I have been testing with an assortment of CPU coolers for upcoming reviews, I can report that my processor appears to be free of any obvious damage. I am particularly careful about pressure when attaching a heatsink, but there have been a couple (including the above mentioned Scythe HPMS mounting system) that could easily have been tightened far beyond what was needed for a proper connection.

We will continue to monitor this situation and update as more vendors offer their response to the issue.

Source: Ars Technica

Video News


December 5, 2015 | 01:23 AM - Posted by Anonymous (not verified)

I remember LGA socket cap is supposed to protect cpu from being crashed by cooler.

December 5, 2015 | 02:36 AM - Posted by JohnGR

Shame to you AMD. You failed again AMD. You should go bankr........ What? We are talking about Intel?

Seriously. This looks bad. Considering that damaging the socket is enough reason to void the warranty of the motherboard, this is really really bad. And it is not uncommon people to force the cooler on the cpu as much as they can, thinking that, that last bit of pressure could drop the temperature a dozen degrees or something.

December 5, 2015 | 08:12 AM - Posted by Sihastru

You do realize that a tower cooler mounted on the socket is basically a lever. The same way you can pry a door open with a crowbar, a tower cooler can actually break not only the socket and the CPU installed in it, but the motherboard too. The mounting system can make this a very "efficient" (bad in this case) lever or an "inefficient" (good in this case) one.

The pressure/weight specifications are only valid for static environments ("static" is more restrictive then "stationary"). Once this requirement is overridden the "lever system" becomes active and damage to your components may occur.

The fact that this is not a widespread problem indicates that the reports we see here are in fact the result of improper use/handling. If it was simply a "constant pressure" problem then we'd have reports from everywhere, people using stock heatsinks, people using AIOs, people using water cooling and so on. But we don't. And not even massive tower coolers, like Noctua's NH-D14 or NH-D15, since they have far better engineered mounting systems.

We had a very similar problem before, systems with broken PCIe slots because they were manhandled. And yet you don't see people accusing the video card's or motherboard's manufacturer, but the the way the system was used/handled.

So, the blame is with the user in the first place (installing/handling), then with the manufacturer of the cooler (bad engineering). If the system was shipped, you can also blame it on that service, but good luck with that.

December 5, 2015 | 10:41 AM - Posted by Luthair

I'm not sure we really know the prevalence of the problem. If the processor is still functional while the CPU is seated most users wouldn't discover it had happened for months or years

December 5, 2015 | 12:41 PM - Posted by JohnGR

Skylake is available only 3+ months in the market and most people didn't consider upgrading to a Skylake seeing that they don't gain much compare to a Haswell while also having to change their memory. We could hear more in the future about this.

December 5, 2015 | 01:17 PM - Posted by Sihastru

Just to clarify, I'm not saying that Skylake CPUs aren't more prone to develop this issue then the older LGA115x processors. Because it is obvious that there is a sizeable difference in the thickness of the interface material. But if they are still in the specified limits for the LGA115x, Intel can't really be blamed.

It is more likely that the coolers' manufacturers that pushed the limits beyond the spec and the way the systems were handled are to blame here. And some coolers have a better design then others. Some are actually over-engineered and some, in order to cut costs, are under-engineered. Price is not actually a measure of the engineering that went into a product.

December 6, 2015 | 04:51 AM - Posted by AS118 (not verified)

Yeah, probably. A lot of these old coolers were made for thicker CPU's like Haswell, so we'll more than likely see it reported more in the future.

At any rate, this just makes me feel that it might be a good idea to make a CPU thicker than necessary just for ruggedness. Yes you can remove a cooler while transporting a case, but it's just more convenient not to have to.

And I also think it's good to support AMD to keep some heat (via competition) on Intel so they're not tempting to do more stuff like this, and take it lying down.

Hell, Intel's already so much of a near-monopoly, I encourage people to buy AMD anyways if their products suit their needs. I sure hope Zen is good, I don't want to ever see Intel be able to screw over PC gamers with monopoly prices and selection.

January 25, 2019 | 05:47 AM - Posted by Falkon Nightsdale (not verified)

Hello, yesterday, my PC died again.
And apparently for same reason like last time - bent pins.
First time it was running for over 2 years, now it was running just few months.

Thing is, I used stock Intel cooler, that I got with i3-6100 chip. It was my home PC, more an office type.

However, breakdown always occured after trying to play a game for a while, so I guess, that it has something to do with thermal change of mass.

December 5, 2015 | 03:10 AM - Posted by RabbitKing

This is scary. Just installed a CM Hyper 212+ Evo on an AMD FX6300 and new Asus mobo. Even after installing the cooler backplate and carefully tightening all the bolts and screws to fairly scary levels of pressure, the cooler was still loose enough to move around on the thermal compound.

I'm no expert on transfer of thermal energy but I suspect better contact would probably be had by a connection that can't move around. Yet that's what this group of all-new hardware does. And I've had to resist the urge to take a power driver to it to really bolt it all down. The FX6300 is not hot chip and it won't be overclocked so this assembly should be fine. Temps seem within good ranges. And after burn-in, the movement is greatly reduced.

But I can see where someone with a hotter proc or wanting to maximize cooling might really torque it down. Tighter is always better. Right? Right? Guess not.

December 5, 2015 | 05:48 AM - Posted by bigboy678 (not verified)

actually that is just a design aspect of the evo. i have one as well and it wiggles despite the pressure on it. its just the way the mounting mechanism makes contact

December 5, 2015 | 03:25 AM - Posted by Anonymous (not verified)

If you had a bunch of time, a magnifying lens and fine tools, is there a way to fix bent LGA pins?

December 5, 2015 | 05:40 PM - Posted by Patrick Proctor (not verified)

Linustechtips has a video on how to do that.

December 5, 2015 | 03:25 AM - Posted by Anonymous (not verified)

If you had a bunch of time, a magnifying lens and fine tools, is there a way to fix bent LGA pins?

December 5, 2015 | 05:46 AM - Posted by bigboy678 (not verified)

most likely not. back in the hay day when the pins were on the cpu itself you could fix bent pins with a mechanical pencil. nowadays with their size and number youre likely to do more harm than good. sorry for the bad news

December 5, 2015 | 05:50 AM - Posted by JohnGR

If you try something like this, you also need patience. Usually people kill the thing they try to fix because they want to finish the job in 5 minutes, max.

December 5, 2015 | 08:44 AM - Posted by Alexander Ruiz (not verified)

So no more Mory sized Heatsinks?

December 5, 2015 | 01:22 PM - Posted by THEBULGARIANMIN

I guess I will be skipping this generation.

December 5, 2015 | 04:26 PM - Posted by Anonymous (not verified)

It looks like most of the higher end Skylake processors are out of stock, at least at newegg, so I don't know how easily you can buy one anyway.

December 5, 2015 | 04:24 PM - Posted by Anonymous (not verified)

Why did they make them so much thinner? These are not mobile parts. Making them thinner doesn't save you anything in a desktop case. Did they make them thinner just to save cost? If that is the case, then it is quite disappointing. I have been really annoyed recently with the so called ATX motherboards that are less than 12"x9.6". They leave an overhang which is not supported by screws or stand offs. In my opinion, if the board isn't the ATX standard size, then it isn't an ATX board.

December 5, 2015 | 05:46 PM - Posted by Anonymous (not verified)

I've seen it happen with a low profile server heatsink at work. The Tech did not tighten down the 4 mounting screws correctly, they fully tightened the first one and then in sequence the other three. This tilted the heatsink as the screws were done. It cracked two corners of the processor PCB, no damage to the socket though.

December 6, 2015 | 10:45 PM - Posted by -- (not verified)

holy hell has this tech never changed a fking tire!?!!!

"you're fired"

December 5, 2015 | 10:42 PM - Posted by Anonymous (not verified)

Sounds like they meed some very accurate machined metal spacer that does not allow the heat sink to come any closer to the processor die no matter how much the mounting bolts are torqued! Some from of very strong spacer cage that can transfer the stresses from the heat sink into the main-board and not allow any torsional forces to distort the CPU and its mounting package substrate or the socket. There should be designed into the motherboard standard a specialized set of extra mounting holes further from the CPU socket in addition to the ones near the CPU socket for mounting a standoff cage to allow for better heat sink stability even if it's just a set of holes that allows for some struts to pass through the motherboard and be connected directly to the case/hard-point mounts attached to the case/frame. Some of those heat sinks are huge and any torsional forces from movement are magnified, and even the static forces are great from the weight of these large heat sinks without the proper amount of attention paid to properly transferring the static and dynamic loads that these heat sinks place on the CPU/socket/package and the main boards physical interface to them.

December 5, 2015 | 10:44 PM - Posted by Anonymous (not verified)

Edit: meed
to: need

December 5, 2015 | 11:30 PM - Posted by Anonymous (not verified)

"The design specifications and guidelines for the 6th Gen Intel Core processor using the LGA 1151 socket are unchanged from previous generations and are available for partners and 3rd party manufacturers. Intel can’t comment on 3rdparty designs or their adherence to the recommended design specifications. For questions about a specific cooling product we must defer to the manufacturer."

piggybacking off what OP was saying, the most interesting bit of information thus far, is Intels comment on the issue. Im sure the legal team at ol' blue is regretting this statement. unless of course, therein lies more info, in this statement, of which we're currently unaware. the 'design specifications and guidelines' bit is important, in so far as it relates to the substrate. for they (intel), clearly state that the 'specifications' and 'guidelines' have not changed. so one would safely assume that the 'performance' of 3rd party HSFs would be similar, if not exactly the same. apparently this not the case. what has changed? the substrate is thinner, making it more prone to problems inherent to the situation. they (intel) take time to note the 'adherence' of 3rd party vendors to said specs/guidelines. unless, within those specs and guidelines, there is clear language (followed by facts/figures) as to the various aspects of the HSF-Socket-CPU relationship, that would exonerate them of any liability.... they're in trouble.

my guess is they'll be ok, as they most likely would not have issued a statement like this, had they not felt they were in the clear, so to speak. unless, of course, as i allude to earlier, someone fouled up. in which case, yea, the legal team at ol' blue is not only regretting this statement, but they're more of a red-ish hue at the moment as well, more than anything anything else :P

December 5, 2015 | 11:50 PM - Posted by Anonymous (not verified)

it would be interesting to know some other facts/figures, that we could use as a frame of reference. for example:

what is the number of cpus damaged by 3rdparty HSFs, regardless of cpu generation/platform, and how does that number compare to the number incidents reported w/skylake processors?

is this just 'coincidence', being mislabeled by victims of 'user-error'? moreover, is just because something is thinner, also mean it is invariably weaker?

December 6, 2015 | 12:14 AM - Posted by Master Chen (not verified)

Sandy-god here. I laugh at all the Scumfail-peasantry.

December 7, 2015 | 11:27 AM - Posted by Anonymous (not verified)

I think someone else is posing at Master Chen. I refuse to believe that the real Master Chen would post such a weak and pathetic trolling attempt. The real Master Chen is a professional, an MLB-level troll.

December 8, 2015 | 09:43 AM - Posted by Master Chen (not verified)

I hate baseball. Scram.

December 8, 2015 | 05:35 PM - Posted by Anonymous (not verified)

Strike two!

December 6, 2015 | 10:33 PM - Posted by -- (not verified)

really...intel has been making silly socket heat sink designs for a while.

Anyone remember the snap push pin heat sink that would bend your motherboard? awesome intel.

meanwhile AMD is sticking to the tried and true.

December 6, 2015 | 10:43 PM - Posted by Anonymous (not verified)

yea this is also why im not a fan of bolting a huge ass heat sink to my motherboard.... what is supporting that weight? a CIRCUIT BOARD

no thanks, i'll do water cooling before i bolt 5lbs of heat sink to my board.

December 7, 2015 | 03:54 PM - Posted by Mutation666 (not verified)

So thats why they dropped packaged heat sinks on K parts

December 12, 2015 | 05:25 AM - Posted by Anonymous (not verified)

See Thicker IS better !

January 20, 2016 | 11:30 AM - Posted by xnor

Any news on this?

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Lines and paragraphs break automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd> <blockquote><p><br>
  • Web page addresses and e-mail addresses turn into links automatically.

More information about formatting options

CAPTCHA
This question is for testing whether you are a human visitor and to prevent automated spam submissions.