Interesting new type of low profile CPU cooler

DontPeek

Weaksauce
Joined
Oct 22, 2014
Messages
67
So 30-40% more efficient then what, the total pieces of crap stock Intel fans?

Still no way an air cooler that small is going to be able to cool anything more than a low power chip that never runs at full load for more than a couple seconds.

The whole air gap type cooler like that to be successful is going to have to have super high QA standards. even the slightest variation in air gap is going to either cause interference with the base plate or greatly reduced cooling capability.

What happens when the bearing(s) start to wear? You can say goodbye to cooling performance.
 
show me again when they have them cooling 100W units and doing so for 50k+ hours then what happens if that motor stops working suddenly and you go to just having the base plate. As it stands currently the current tech has enough thermal mass to cool without the fan for a few minutes or during light use in most cases with throttling.
 
I'm looking forward to the cooler being released/reviewed. Even with potential drawbacks, if it works anywhere near as projected, it can be a boon for SFF. What's the limitation of some of the slimmer cases like the ISK110 and especially Mini-Box M350? Cooler height! Heck, the normal stock coolers won't even fit in the M350, though the lower power ones will.

I'm willing to wait for reviews before condemning future hardware. :p
 
Prototype might be small, but its more of a proof of concept. I'm sure it could be scaled up for higher TDP requirements. Any step towards trying to make fans quieter is a good thing.
 
Also think about the use cases for thin mITX builds. Currently, you have to resort to the Intel HTS1155LP to really make use of the thin form factor but that increased the footprint of the whole motherboard assembly to small µATX-like dimensions. With this, you can truly have a feature complete build of about 180*180*30mms. I consider that to be quite an achievement.

So you'd really compare this type of cooler with the 1U blower type CPU coolers that currently exist, and I think from that perspective it is quite an upgrade.

Only time will show how susceptible those new coolers are to wear.
 
So 30-40% more efficient then what, the total pieces of crap stock Intel fans?

Still no way an air cooler that small is going to be able to cool anything more than a low power chip that never runs at full load for more than a couple seconds.

The whole air gap type cooler like that to be successful is going to have to have super high QA standards. even the slightest variation in air gap is going to either cause interference with the base plate or greatly reduced cooling capability.

What happens when the bearing(s) start to wear? You can say goodbye to cooling performance.

You guys are totally missing the point. It's a proof of concept. Is it going to cool as well as a coolmaster hyper 212? no, no one is claiming that. That's simply not what it's competing against. As the technology matures we could see these used in laptops, see more powerful CPUs inside NUCs, or even have an ITX computer case be slightly taller than the io shield.
 
show me again when they have them cooling 100W units and doing so for 50k+ hours then what happens if that motor stops working suddenly and you go to just having the base plate. As it stands currently the current tech has enough thermal mass to cool without the fan for a few minutes or during light use in most cases with throttling.

All modern consumer grade boards have protections against CPU overheating, so the amount of time a CPU will last if the fan dies is irrelevant.
 
All modern consumer grade boards have protections against CPU overheating, so the amount of time a CPU will last if the fan dies is irrelevant.

depends how you evaluate and handle down time... also i would like to be sure that if im playing a game with the chip at full throttle if the fan suddenly stops that it wont just die.
 
depends how you evaluate and handle down time... also i would like to be sure that if im playing a game with the chip at full throttle if the fan suddenly stops that it wont just die.

again you're missing the point. This isn't designed to cool a 6 core i7 extreme edition, it's a proof of concept that will allow more powerful CPUs to be put in smaller form factors or shrink down the size of already small low powered computers even more.
 
You guys are totally missing the point. It's a proof of concept. Is it going to cool as well as a coolmaster hyper 212? no, no one is claiming that. That's simply not what it's competing against. As the technology matures we could see these used in laptops, see more powerful CPUs inside NUCs, or even have an ITX computer case be slightly taller than the io shield.

For some reason when you said that the first picture that popped into my head was of a Hyper 212 spinning in a fast circle :)
 
again you're missing the point. This isn't designed to cool a 6 core i7 extreme edition, it's a proof of concept that will allow more powerful CPUs to be put in smaller form factors or shrink down the size of already small low powered computers even more.

I'm not saying I want that but I would like that safety but that said in small form factor we have heat pipes and regular fans I don't see this replacing current tech it has too many failure points. From what I can see the air gap this cooler is using is hard drive head thin if dust gets in there end the fan with so little thermal mass even an i3 or i5 would be in danger of either pooffing if it could not throttle low enough to fall within the thermal mass of the heat plate or turn off quick enough.
 
Interesting

My Silverstone GD04 has a 70mm height restriction as I have a Blu-ray drive fitted.. Currently using a Scythe Big Shuriken 2 but not impressed with the noise coming from it at idle.

Got a Zalman CNPS8900 Quiet sat here which I think will be quieter but not got round to fitting yet.

Cooling an i5-4690K (stock) in my HTPC/TV gaming rig.
 
It will be interesting to see how they overcome some of the hurdles you guys have mentioned, but I think they will take them into account even if there are (and I'm sure there will be) compromises compared to a traditional cooling setup. I do however think the idea is interesting and it does seem to make sense as a concept, although i'm no engineer. Having a heatsink move through the air does seem like it would be more efficient than using a fan to try and push air through a heat sink. That fan is working with a lot of resistance that this solution would not have to deal with. The main thing that i'm hoping for is that it is a quieter solution that can keep up with the more popular i5 and i7 intel chips. If it can be small, quiet, and adequately cool a 4690 or 4790 I think it could be a big win for small form factor PCs. Hopefully we get more information during CES, especially some actual numbers although i'm not super optimistic about that.
 
I'm not saying I want that but I would like that safety but that said in small form factor we have heat pipes and regular fans I don't see this replacing current tech it has too many failure points. From what I can see the air gap this cooler is using is hard drive head thin if dust gets in there end the fan with so little thermal mass even an i3 or i5 would be in danger of either pooffing if it could not throttle low enough to fall within the thermal mass of the heat plate or turn off quick enough.

Again...it's a proof of concept, no one is claiming it's ready for prime time. As technology gets smaller and smaller we need to come up with complimentary technology to cool it. This is one of the ideas. This concept could easily increase the cooling capabilities of laptops and NUCs if we used it in conjunction with heatpipes. We could easily see a NUC sized device with low powered variant of an i7 and mid range GPU
 
not enough surface area for high end oc'd desktop cpus

maybr useful in laptops
 
not enough surface area for high end oc'd desktop cpus

maybr useful in laptops

The one showed in the picture is TINY. Larger ones could be made and still be very small. For instance this is the Sandia cooler. Quite a bit larger but same concept.

Sandia-Cooler-6.jpg


Really though the size will depend on if these really are more efficient than regular heatsinks. If they are 30%-40% more efficient as they claim than they only need to be 60%-70% the size as well. You don't need as much surface area if that surface area is doing more.
 
Doesn't seem like it would really be as applicable to laptops and tablets - the last thing I'd want in a small form factor that's carried around is a relatively large and heavy spinning mass. I wonder how they efficiently transfer heat from the processor die/IHS to the rotating mass?
 
I wonder how they efficiently transfer heat from the processor die/IHS to the rotating mass?
I believe it's stated somewhere in the first news about the concept - The distance between surfaces is just enough for friction to break static/kinematic barrier but its small enough that heat energy transfer is still optimal.
 
HDDs rely on similar air-bearings, and those aren't sealed: they utilise vent-holes (to allow for ambient pressure change), recirculate their internal air, and filter the internal air through small mesh 'pillow' filters to prevent dust from impacting the drive heads.
If these coolers are assembled in a clean environment, with the air-bearing having asimilar semi-sealed setup (recirculating a small volume of filtered air) then I'm not particularly worried about dust ingress.

Their test data page directly compares one of their designs to the 1156/1155 reference cooler (73W TDP). Even assuming merely equal performance/volume, this is sufficient for most chips in an ITX rig.
 
Doesn't seem like it would really be as applicable to laptops and tablets - the last thing I'd want in a small form factor that's carried around is a relatively large and heavy spinning mass. I wonder how they efficiently transfer heat from the processor die/IHS to the rotating mass?

Basically, they make the gap between the rotating mass and the heat spreading plate underneath it very, very, very, very thin.
 
HDDs rely on similar air-bearings, and those aren't sealed: they utilise vent-holes (to allow for ambient pressure change), recirculate their internal air, and filter the internal air through small mesh 'pillow' filters to prevent dust from impacting the drive heads.
If these coolers are assembled in a clean environment, with the air-bearing having asimilar semi-sealed setup (recirculating a small volume of filtered air) then I'm not particularly worried about dust ingress.

Their test data page directly compares one of their designs to the 1156/1155 reference cooler (73W TDP). Even assuming merely equal performance/volume, this is sufficient for most chips in an ITX rig.

Thanks for the link. According to them the thermal resistance is lower at about 20db than the stock cooler at 35db (the intel cooler's maximum speed according to their chart). Never seen anything measured in thermal resistance before so i'm not sure how that relates to actual cooling performance.

Also for those worried about the fan stopping without a heatsink it looks like they have a prototype with a stationary heatsink surrounding the impeller. Looks like it can also be set up with ducting which could be beneficial for exhausting hot air out of a small case.
 
i would like to be sure that if im playing a game with the chip at full throttle if the fan suddenly stops that it wont just die.

Unless BIOS was broken or settings intentionally altered to disable the built-in protections, your CPU will not "just die" from a catastrophic cooling failure at stock speeds/settings.

This was demonstrated over a decade ago with VIA C3 and Intel Pentium 4 that were able to stay running with no heatsink attached whatsoever. That's right! You can even find videos of fully operational computers where they intentionally remove the heatsink, and the system (albeit maybe with some throttling) keeps chugging along. Heck, I had a Pentium III that had the heatsink fall off, and it kept running for a couple minutes before it locked up. After letting it cool off and reattaching the heatsink, it started working again.

Of course the AMD Athlon chips of the day would immediately kill themselves... :eek:
 
I'm interested. I've got a few machines I can already see using these in, if they're capable of cooling while still remaining quieter than a stock cooler.
 
I'm more interested in that small cooler than in the large model that will apparently be released. I guess they'll try to find out how it can compete in the market and then go for other form factors.
 
It was cool to see it on a mini itx board. Seems like they were able to pack a lot of heatsink into a fairly small place. Can't wait to see some real numbers on some actual cpus.
 
It's definitely a highly interesting concept, and I hope my doubts concerning noise (and other issues) will be proven exaggerated.
 
show me again when they have them cooling 100W units and doing so for 50k+ hours then what happens if that motor stops working suddenly and you go to just having the base plate. As it stands currently the current tech has enough thermal mass to cool without the fan for a few minutes or during light use in most cases with throttling.
How is this any different from a plastic fan? What happens if that motor stops working suddenly and you go to just having the heatsink?

I'm sure that if the fan stops spinning, at least you'll still get heat dissipation to the movable metal fan versus a plastic fan on a heatsink with less metal (because designed to be cooled by fan so less metal needed).

I for one would be very interested in a heatsink for the Gigabyte Brix designed using this concept.
 
Looks like they're partnering with CoolerMaster to bring this technology to market: https://www.youtube.com/watch?v=ahgnVrULqUs

The CM version now has (stationary) heat pipes and fins in the circumference and looks gigantic. The rotating part of the cooler now has fins that appear to be cast as opposed to the machined channels in the previous prototypes. That looks really cheap and reminds me a bit of how Thermaltake created the Golden Orb from ArctiCooler's concept.
 
Last edited:
does not look very promising after reading some articles
for such spinning metal parts, the life of bearing and rotor will be in question
secondly, they first aimed at enterprise market, but after quite sometime no company wants to adapt their tech, which might hint the failure of their product
 
secondly, they first aimed at enterprise market, but after quite sometime no company wants to adapt their tech, which might hint the failure of their product

Might but it's not for sure. I'd say enterprise market will wait till the product is well tested by normal users and the technology is mature enough to introduce it in enterprise segment.
 
I was very interested to see how they mate the two parts, and indeed both video and link show that.
 
If they use a magnetic/barometric bearing then friction/start-stop wear isn't as big a factor...but that's a big if. Have to consider components that'd be in close proximity to the electromagnet, and whether it'd be able to start the potentially huge weight in the first place.

But because of the design, there'd be near-zero friction and it wouldn't have to spin very fast. I'd be interested to see one working, myself.
 
Back
Top