Our Lapped CPU Heatsink Test Platform

Cooling Processors quietly

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Re: Our Lapped CPU Heatsink Test Platform

Post by MikeC » Tue Dec 03, 2013 7:43 am

kevith wrote:Yeah, kind of like that. If two surfaces has to fit each other real tight, that´s the way to go.

Regarding practicality, it probably isn´t... :-) Both parts would have to be fixated - unless you have a very steady hand - as you polish, which requires some thinking and tinkering, and when you´re finished, these two parts will be forever "married until death do them part".

But if you actually succeeded, I think it might give an extra degree or two. The ideal scenario is the two surfaces fitting so snugly together, that they don´t need no TIM. That´s impossible in the real world, where we pay for things with real money, though, but a polish like this might get us a little closer.
A simpler way to achieve the same -- or even better -- cooling performance would be to delid the CPU and use a flat or lapped base heatsink. You'd need to make sure the heatsink mounting is done in an even, consistent way to avoid crushing edges or corners of the exposed die, but that die surface is mirror flat. Most of the good through-the-board bolt mounting systems would be fine for this. Of course, this is easiest done with CPUs where the IHS is not soldered but TIM-ed, tho it's possible to do with the former if you heat it up nicely with a hair dryer or heat gun and are willing to clean up the solder gunk underneath. (There's probably standard solutions for solder cleanup.)

But having said all that, unless you're oc'ing to the bleeding edge & need every cooling advantage, I really don't see the need for users to go to such trouble. The best heatsinks around (either convex or flat base) are perfectly capable of cooling the hottest CPUs, especially if super low noise is not a high priority.

kevith
Posts: 11
Joined: Fri Oct 21, 2011 3:06 am

Re: Our Lapped CPU Heatsink Test Platform

Post by kevith » Thu Dec 05, 2013 12:49 am

A simpler way to achieve the same -- or even better -- cooling performance would be to delid the CPU and use a flat or lapped base heatsink. You'd need to make sure the heatsink mounting is done in an even, consistent way to avoid crushing edges or corners of the exposed die, but that die surface is mirror flat. Most of the good through-the-board bolt mounting systems would be fine for this. Of course, this is easiest done with CPUs where the IHS is not soldered but TIM-ed, tho it's possible to do with the former if you heat it up nicely with a hair dryer or heat gun and are willing to clean up the solder gunk underneath. (There's probably standard solutions for solder cleanup.)
Wow, that´s hardcore :) The cooler directly on the die, yeah! I have seen videos of guys delidding a soldered CPU, it´s not something I would do, but done with care it works fine. You can buy solder-cleanup-flatwire for the purpose, put a strip over the solder, heat it with the iron, and it sucks up a portion of solder. Cut the dirty end off the wire and repeat til it´s all gone.
But having said all that, unless you're oc'ing to the bleeding edge & need every cooling advantage, I really don't see the need for users to go to such trouble. The best heatsinks around (either convex or flat base) are perfectly capable of cooling the hottest CPUs, especially if super low noise is not a high priority.
I couldn´t agree more, I´m much more in to quiet computing than anything else, please accept my compliment you on your dedication and great site.

Das_Saunamies
*Lifetime Patron*
Posts: 2000
Joined: Tue May 15, 2007 1:39 am
Location: Finland

Re: Our Lapped CPU Heatsink Test Platform

Post by Das_Saunamies » Thu Dec 05, 2013 1:40 am

We could install coolers on our Thunderbirds without IHS or incident, we can do it again no sweat. 8)

(just have to be veeeeery veeeery careful, and suddenly dosage is going to matter a lot again)

quest_for_silence
Posts: 5275
Joined: Wed Jun 13, 2007 10:12 am
Location: ITALY

Re: Our Lapped CPU Heatsink Test Platform

Post by quest_for_silence » Wed Nov 04, 2015 12:32 am

I'd like to update this old but useful thread/article with a very recent video showing how the alignment of a Noctua cooler base actually is by using a digital indicator:

https://www.youtube.com/watch?feature=p ... T7cX1fOcbk

Comparatively it's also interesting to me to point out what Noctua marketing department currently says about that base here:

"Why is the cooler’s contact surface slightly convex?
As the Integrated Heat Spreaders (IHS) of today’s CPUs are slightly concave, the cooler’s contact surface has been deliberately designed to be slightly convex in order to ensure optimal contact. This way, more contact pressure will be applied at the centre of the IHS directly above the DIE, which results in better heat transfer and improved overall performance.
"

As an added very personal thought, I think SPCR could revise how they check the coolers' base shape, and broadly speaking I guess single or dual axis sight inspection might be sometimes, somewhat misleading.

Das_Saunamies
*Lifetime Patron*
Posts: 2000
Joined: Tue May 15, 2007 1:39 am
Location: Finland

Re: Our Lapped CPU Heatsink Test Platform

Post by Das_Saunamies » Wed Nov 04, 2015 1:32 am

Well... them's some numbers alright. The video does not explain itself, so laymen like myself are left wondering if it's alright to move the object by hand and what and how the device is measuring. It seems consistent enough with itself in different areas of the base.

xan_user
*Lifetime Patron*
Posts: 2269
Joined: Sun May 21, 2006 9:09 am
Location: Northern California.

Re: Our Lapped CPU Heatsink Test Platform

Post by xan_user » Wed Nov 04, 2015 1:41 am

quest_for_silence wrote: As an added very personal thought, I think SPCR could revise how they check the coolers' base shape, and broadly speaking I guess single or dual axis sight inspection might be sometimes, somewhat misleading.
just what mike and crew wants, more work. :roll:

quest_for_silence
Posts: 5275
Joined: Wed Jun 13, 2007 10:12 am
Location: ITALY

Re: Our Lapped CPU Heatsink Test Platform

Post by quest_for_silence » Wed Nov 04, 2015 3:32 am

Das_Saunamies wrote:how the device is measuring.

I guess you have to look for by yourself: that's an introductory text on Wikipedia about those instruments.

Das_Saunamies wrote:It seems consistent enough with itself in different areas of the base.

To me it is clearly far from being (regularly) convex (differently from what Noctua marketing says), and within the accuracy of a quick check we can see that a great portion of the base left side is recorded about 0.2mm irregularly lower than the rest of the base. I'd say it's "bent" (that cooler is perfectly functional, as far as I know it's a new, retail specimen randomly taken from the warehouse shelves to perform that quick alignment check).

Das_Saunamies
*Lifetime Patron*
Posts: 2000
Joined: Tue May 15, 2007 1:39 am
Location: Finland

Re: Our Lapped CPU Heatsink Test Platform

Post by Das_Saunamies » Wed Nov 04, 2015 3:34 am

As I was commenting on the device, what I meant is that the instrument and the figure it is showing are consistent, not the object being measured. :wink:

Post Reply