-
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?
- Thread starter c627627
- Start date
- Joined
- Oct 30, 2009
Well, yes and no. Scythe tells you not to "overtighten" your hold-down screws. But what may be overtightening for a lightly-built person is barely tight for a weight-lifter. This issue is why we have screw stops.
How is it user error when there is nothing to go by tightening the heatsink screws on skylake. How do you measure and adjust care?
Also this was not a problem on previous Intel generations.Common sense prevails.How is it user error when there is nothing to go by tightening the heatsink screws on skylake. How do you measure and adjust care?
It was not a problem before because the substrate on other platforms was thicker.
Now, how is it the cooler manufacturers problem that Intel cheepo'ed out on their end?
- Joined
- Sep 13, 2015
Part of this falls back on the heatsink manufacturers exceeding the intel specs on allowed pressure. They may have gotten away with it before due to the thicker pcb but now it is causing more problems as the pcb cannot take the extra stress. Take scythe for example, I think they realized this in their mounting system because they are now shipping with an extra spacer specifically for the 1151 socket.
I disagree. Unless their cooler was specifically designed for Skylake, they are just providing spacers so people can still use their older products on their Skylake builds. I consider that exceptional CS on the cooler manufacturers part.
I mean, no cooler manufacturer out there knew that there was going to be a bending problem until well after release. It's just now coming to light. Who's fault is that? When did Intel inform these manufacturers about the specific mounting pressure?
Answer: They didn't.
They are now, because it's costing them in RMA's and they need to place the blame elsewhere.
Last edited:
- Joined
- Feb 4, 2009
- Location
- cypress, tx
They are admitting they were exceeding Intel's spec.
I highly doubt that.
That would make cooler manufacturers liable for every bent processor out there. That is a quick way to the poorhouse.
- Joined
- Feb 4, 2009
- Location
- cypress, tx
And them providing free kits to remove the issue is a very simplistic way of saying we messed up.
I don't doubt that at all.
As I asked before, did Intel even provide a spec before the problem showed up?
I doubt it.
- Joined
- Feb 4, 2009
- Location
- cypress, tx
PCB is rated to 50PSI
I understand that.
Again, what I'm asking is, was that 50 PSI rating provided to anyone before the bending problem was exposed?
- Joined
- Sep 13, 2015
I don't know Scotty as I cannot find a spec sheet for socket 1151 to verify the release date of the publishment. My argument would be that all of the prior sockets were rated at 50 lbs as well. Find socket 1150 specs here, page 24 has the mounting pressure specifications.
So even if the 1151 specs were not released prior, the mounting pressure should still not exceed 50 lbs. based off of prior specifications.
The following is purely my speculation so please take it with a grain of salt.
My supposition is that socket 1150 and prior, the cpu pcb had a listed spec of 50 lbs mounting pressure but due to the thicker pcb the actual product was able to withstand pressure at a higher value than 50 lbs so coolers that exceeded the listed spec of 50 lbs did not tend to cause problems.
So along comes socket 1151 with a 50 lb rating and a thinner pcb and lough and behold the pcb really is only suitable for 50 lbs of pressure as the spec states. So all of the prior coolers that exceeded this specification and got away with it are now having problems because they are exceeding this specification. Hence the add on spacers to bring the cooler back to the listed specification.
So please do not misunderstand me. Yes the end user has a responsibility to use their own judgement on mounting pressure but the manufacturer should make this achievable without to much issue. I commend the manufacturers for coming through to supply the hardware to bring their products back in line with listed specifications but at the end of the day If it met listed specifications from the get go (even for older sockets) they would not need to enact this remedy.
By the way if anyone can find a socket 1151 application guide or spec sheet I would love a link.
So even if the 1151 specs were not released prior, the mounting pressure should still not exceed 50 lbs. based off of prior specifications.
The following is purely my speculation so please take it with a grain of salt.
My supposition is that socket 1150 and prior, the cpu pcb had a listed spec of 50 lbs mounting pressure but due to the thicker pcb the actual product was able to withstand pressure at a higher value than 50 lbs so coolers that exceeded the listed spec of 50 lbs did not tend to cause problems.
So along comes socket 1151 with a 50 lb rating and a thinner pcb and lough and behold the pcb really is only suitable for 50 lbs of pressure as the spec states. So all of the prior coolers that exceeded this specification and got away with it are now having problems because they are exceeding this specification. Hence the add on spacers to bring the cooler back to the listed specification.
So please do not misunderstand me. Yes the end user has a responsibility to use their own judgement on mounting pressure but the manufacturer should make this achievable without to much issue. I commend the manufacturers for coming through to supply the hardware to bring their products back in line with listed specifications but at the end of the day If it met listed specifications from the get go (even for older sockets) they would not need to enact this remedy.
By the way if anyone can find a socket 1151 application guide or spec sheet I would love a link.
The spec came out with the cpu I'd have to imagine (just like always in the past)..we have no reason to think it wasn't published or withheld, do we?
- Joined
- Feb 4, 2009
- Location
- cypress, tx
Yes Intel has always been very specific with coolers and goes as far to include dead zones coolers cannot enter.
- Joined
- Oct 30, 2009
Yes. The rating is not 50 psi but 50 lbs total. Intel published this years ago. I can't lay my hands on it now, but I have seen it in Intel specs that go back to pre-Lynnfield days.
The search term is "static compressive load." One iten you can find is 4th generation, from 2013, here:
http://www.intel.com/content/dam/ww.../guides/4th-gen-core-lga1150-socket-guide.pdf
And a guy from a heatsink OEM wrote me this in 2013, "It's not uncommon that heatsink manufacturers exceed the Intel specs by a huge amount trying to achieve better performance. The maximum recommended static compressive load on LGA115x is 50lbf. For LGA2011, it is 90lbf."
Last edited by a moderator:
