• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Are Intel Skylake CPUs are bending under the pressure of some coolers?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

c627627

c(n*199780) Senior Member
Joined
Feb 18, 2002
Last edited:

cdawall

Member
Joined
Feb 4, 2009
Location
cypress, tx
That pic shows a CPU that was dropped. Look closely only two corners bent and they are physically broken. No cooler could do that. You would literally have to break the socket on the motherboard to get a cpu to bend that far.

I have had some home brew machines come into the shop with bent CPU (2 out of hundreds/thousands) and both of them were incorrectly mounted hyper 212's. The PCB was only visibly bent when you put a razor on top of it to check straightness.
 

cdawall

Member
Joined
Feb 4, 2009
Location
cypress, tx
If people didn't crank down the screws on a cooler like its going out of style they would never bend. Any damage done to any of the CPU's I have seen is user inflicted not design flaw.
 
OP
c627627

c627627

c(n*199780) Senior Member
Joined
Feb 18, 2002
Well good.
But still would you say that there is any reason to believe that the thinner wafer is a contributing factor?

In other words, are you saying all instances of this user error would also happen on thicker wafers?
 

Mr.Scott

Beamed Me Up!
Joined
Jun 9, 2013
Well good.
But still would you say that there is any reason to believe that the thinner wafer is a contributing factor?

In other words, are you saying all instances of this user error would also happen on thicker wafers?

The thinner PCB is the weak point yes, and user damage HAS happened before on on many other platforms with thicker 'wafers'.
Socket crush still happens on most Intel sockets, and platforms with bare die processors have had more than their share of cracked/chipped core complaints. All are usually user inflicted rather than cooler or CPU mis-design.
 

WhitehawkEQ

Premium Member
Joined
Dec 6, 2010
EDIT: Better source links

http://arstechnica.com/gadgets/2015...-bent-and-broken-by-some-third-party-coolers/
http://www.tomshardware.com/news/skylake-cpus-damaged-by-coolers,30690.html

Intel used a thinner wafer on Skylake CPUs -- older CPUs had a thicker wafer.
Intel claims the thinner size is still rated for the same 50 pound static load.

This news is almost a year old, if you've heard any developments since then, please post in this thread.

When you say wafer, what your referring to is the fiberglass PCB (Printed Circuit Board) that the silicon chip is mounted to. And the PCB can be bent.
A Wafer is a 12" disk used to make lots of chips on then the CPU's are cut out of the big disk.
 
OP
c627627

c627627

c(n*199780) Senior Member
Joined
Feb 18, 2002
Please post a better word to describe the thinner Skylake (left), versus Haswell (right):

Wafer.jpg
 
OP
c627627

c627627

c(n*199780) Senior Member
Joined
Feb 18, 2002
All right. The two articles linked use the word 'substrate' to describe it.
The original article, now removed, was a pcgamer article that use the incorrect word 'wafer'.

I'm glad we cleared that up.
 

EarthDog

Gulper Nozzle Co-Owner
Joined
Dec 15, 2008
Location
Buckeyes!
Old news...

But yeah it happens, and is happening to me now. I backed off a bit on the pressure of my waterblock (no stops, modified, not the blocks fault).
 

WhitehawkEQ

Premium Member
Joined
Dec 6, 2010
All right. The two articles linked use the word 'substrate' to describe it.
The original article, now removed, was a pcgamer article that use the incorrect word 'wafer'.

I'm glad we cleared that up.

Yes, substrate is another word used to describe what the chip is mounted on, I forgot that 1 :)
It's easy to get lazy and not use the right words for things :) I know this stuff because I'm a electronic and PC tech, some of my family worked at Moto and AMD :)
 

ehume

Member
Joined
Oct 30, 2009
This is why I am leery of heatsinks that do not use spring-mounted pressure screws. Based on a mid-2000's article posted on this site, I tried two different pressures when I reviewed the Scythe Fuma. To no one's surprise, the high pressure mount cooled better.

I had a conversation with a guy from Noctua. Seems they take some pride in not exceeding the Intel-specified mounting pressure. And now that Intel has thinned it's PCB, maybe virtue shall be rewarded. Note: of all the spring-mounted heatsinks, the NH-D15 cools the best.
 
Last edited:

Sgrinzo

New Member
Joined
Oct 14, 2016
Ok, it Holds 50 pounds statically, so no one should worry about an ultra 120 extreme if it does not move the pc?
 

EarthDog

Gulper Nozzle Co-Owner
Joined
Dec 15, 2008
Location
Buckeyes!
its not the weight of the item but the amount of force being used to push/hold the heatsink down on the CPU that is causing issues if one is not careful or using a cooler with 'stops' (where the screws only go so far). It really isn't a huge deal honestly.
 

Sgrinzo

New Member
Joined
Oct 14, 2016
if the heat sink is held in place by the screws with a spring, the force imparting should be in the standard of intel