vapor chamber heatsinks - yay or nay?
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
vapor chamber heatsinks - yay or nay?
just noticed this over at ocforums.com, someone finally made a heatsink that used a full-fledged vapor chamber instead of heatpipes. it's not the first, i know intel sells one for their 1u servers and thermacore seems to sell them to someone (not that i've ever heard of them), i'm just wondering how optimal this vapochill model will be. rrp is 27 pounds sterling.
from what i've read, a well-designed (and oriented) vapor chamber can dissipate over 200 watts, though i can still see a use for complementary heatpipes to help get the heat from atop the chamber to the fins themselves.
is this the future of heatsinks? the chambers seem to do a terrific job of spreading heat from the die to the entire base area.
oh and one absolute gem of execu-drivel from the product blurb:
"It’s hard to imagine a strategic partnership with stronger synergies; both companies have very strong and highly specialized technology know-how. This, combined with our company’s experience in volume production and quality control and asetek’s extensive application expertise and industry network, is nothing less than ideal." Mr. Eberhard Günther, Managing Shareholders of LG ThermoTechnologies, the LG Industries subsidiary in operational dialogue with asetek, points out further synergies: "The asetek / LG industries partnership facilitates unique opportunities to fast track future R&D cycles; we are certain that this partnership is a significant step towards the centre of the industry for mainstream CPU cooling".
from what i've read, a well-designed (and oriented) vapor chamber can dissipate over 200 watts, though i can still see a use for complementary heatpipes to help get the heat from atop the chamber to the fins themselves.
is this the future of heatsinks? the chambers seem to do a terrific job of spreading heat from the die to the entire base area.
oh and one absolute gem of execu-drivel from the product blurb:
"It’s hard to imagine a strategic partnership with stronger synergies; both companies have very strong and highly specialized technology know-how. This, combined with our company’s experience in volume production and quality control and asetek’s extensive application expertise and industry network, is nothing less than ideal." Mr. Eberhard Günther, Managing Shareholders of LG ThermoTechnologies, the LG Industries subsidiary in operational dialogue with asetek, points out further synergies: "The asetek / LG industries partnership facilitates unique opportunities to fast track future R&D cycles; we are certain that this partnership is a significant step towards the centre of the industry for mainstream CPU cooling".
-
- Posts: 968
- Joined: Fri Jan 07, 2005 7:13 pm
- Location: Bristol, UK
- Contact:
Re: vapor chamber heatsinks - yay or nay?
*relief from Intel as their triple-core Prescott has been made possible*yeha wrote:from what i've read, a well-designed (and oriented) vapor chamber can dissipate over 200 watts...
-
- SPCR Reviewer
- Posts: 2696
- Joined: Thu Dec 11, 2003 9:53 pm
- Location: Scarsdale, NY
- Contact:
Re: vapor chamber heatsinks - yay or nay?
Depends...if they ran full triple core Prescott at full speed and full voltage we'd be talking close to double that (at 130watts/core, three cores comes close to 400 watts...).StarfishChris wrote:*relief from Intel as their triple-core Prescott has been made possible*yeha wrote:from what i've read, a well-designed (and oriented) vapor chamber can dissipate over 200 watts...
hrm the link still works on my end, but here's the ocforums thread just in case.
definitely looks like it'd be cheaper to build than a heatpipe sink, perhaps you could extrude the fins and chamber in one piece, then seal the chamber afterwards to eliminate the chamber->fins bond. and as we know, extruded chunks of metal (even with a bit of work done afterwards) cost a lot less than multi-fin pressed ones.
definitely looks like it'd be cheaper to build than a heatpipe sink, perhaps you could extrude the fins and chamber in one piece, then seal the chamber afterwards to eliminate the chamber->fins bond. and as we know, extruded chunks of metal (even with a bit of work done afterwards) cost a lot less than multi-fin pressed ones.
Especially http://www.asetek.com/main/page.asp?sideid=628 looks interesting from a silent computing pov.
And it doesn't hurt either that asetek is a Danish company which should make their products cheaper for me personally :>
And it doesn't hurt either that asetek is a Danish company which should make their products cheaper for me personally :>
-
- Friend of SPCR
- Posts: 502
- Joined: Sat Oct 23, 2004 2:02 pm
-
- Patron of SPCR
- Posts: 2674
- Joined: Sun Jul 04, 2004 6:07 am
- Location: Houten, The Netherlands, Europe
That's probably why they mounted the rad slanted and not straight.
EDIT: On second thought, that would probably help with the flow back part, but not not with the contact between the fluid and the base. So unless there is some wick structure on the inside of the base, a tower setup is a no go (or al least less effective).
EDIT: On second thought, that would probably help with the flow back part, but not not with the contact between the fluid and the base. So unless there is some wick structure on the inside of the base, a tower setup is a no go (or al least less effective).
Last edited by Tibors on Wed Mar 09, 2005 7:00 am, edited 1 time in total.
I believe that is the reason for the whole heatsink to be at an angle, so the liquid gets back to the base an therefore the heatsink also works in tower casesultraboy wrote:I have that same question too since Asetek Website says it works on 'gravity'.Krispy wrote:Will this work ok with the motherboard in a vertical position?
My guess would be that the odd angled position is chosen to make it work when mounted vertically. But that's just guessing, really.Krispy wrote:Will this work ok with the motherboard in a vertical position?
EDIT: I'll remember to take the habit of reloading the thread before replying... this is not the first time I'm repeating what has been said 20min earlier
Last edited by JanW on Wed Mar 09, 2005 10:49 am, edited 1 time in total.
it looks like it'll still function somewhat horizontally, as the semisphere at the bottom will pool the working fluid to a degree allowing contact with the heat source. i can't see it working as well as the vertical orientation however. depends on the wick performance - most modern vapor chambers use a wick lining of the chamber, with capillary action drawing condensate back to the heat source.
Images of the Asetek Cooler from CeBit (via Techpowerup):
Installed on a mobo
http://www.techpowerup.com/reviews/Cebi ... micro1.jpg
Another installation shot
http://www.techpowerup.com/reviews/Cebi ... micro2.jpg
And Another
http://www.techpowerup.com/reviews/Cebi ... micro3.jpg
Availability starting May 2nd (according to Techpowerup).
Installed on a mobo
http://www.techpowerup.com/reviews/Cebi ... micro1.jpg
Another installation shot
http://www.techpowerup.com/reviews/Cebi ... micro2.jpg
And Another
http://www.techpowerup.com/reviews/Cebi ... micro3.jpg
Availability starting May 2nd (according to Techpowerup).
Last edited by halcyon on Sun Mar 13, 2005 11:37 am, edited 1 time in total.
http://www.hardware-test.dk/ht/test_show.asp?id=3697
more pictures
more pictures
-
- Posts: 419
- Joined: Sun Sep 19, 2004 1:05 pm
- Location: Palo Alto, CA
as probably 90% of folks have a tower type case,there would need to be a pretty clear alert if the mobo needed to lay horizontal. Still,I can envision a pretty neat "desktop",wood case,with an ambient air tube feeding outside air to this (passive) and the whole enchilada cooled by the Coolmax psu with the variable speed 140mm fan. As the A64 3000 Winchesters run pretty cool it seems doable. It might be prudent to have an Artic Cooling TC,80mm thermal control fan set up so it only kicks in on major loads,as a safety factor that would add little to net noise. Going passive on VGA+chipset,and suspending a Spinpoint,you could get pretty close to silent. There are heatpipe passives and effective ultra low rpm fan options that give good results but in theory this may be just a little quieter. Using an external mount passive PSU might allow a single 80mm low rpm to cool the system-floor mounted,ducted to the rear