Cooling Hardware With Microfans 197
Jeriten writes "NewsScientist puplished this story about how your chips could be cooled down without that huge and noisy fan. Answer is multiple fans sized smaller than head of a pin and growed directly to a surface of a chips." Now if they could just make hard drives silent, we finally could hear ourselves think
in a room with 3-4 computers. I tell ya, the noise generated by a few
PCs doesn't seem like much until you turn off the tunes.
Big fans (Score:2)
OK, (Score:1)
What about (Score:2)
Re:What about (Score:1)
--
Pedantry (Score:1)
Why not collect that heat? (Score:3)
A related idea would be to make some sort of heat exchanger that'd simultaneously cool the cpu and warm your room. With lots of machines, that could save a lot in heating costs.
I don't know, just an idea. Anyone know if something like this has ever been done before?
How Valuable (Score:1)
Not only Maxtors (Score:1)
Re:Pedantry (Score:1)
> Jeriten writes "NewsScientist
Ah yes. I remember the News Scientist. Is that the dude who reads the news in a white coat?
> puplished this story
Puplished? Oh so it's one of those nasty scientists that does tests on dogs. Nasty man.
> about how your chips could be cooled down
Personally I find the best way to cool chips is to leave them out of the oven for a while after you've fried them.
> Answer is multiple fans
Is this a definite article? [poor grammatical joke]
> sized smaller than head of a pin and growed
Like Topsy? She just growed and growed and growed.
> directly to a surface of a chips.
A chips?
[In silly 60s detective film voice]: "Excuse me Mr. Chips we have a situation here."
I like my loud computer. (Score:1)
two on the power supply,
two 80mm fans venting the hard drives,
the CPU fan, the fan on the GeForce 2 GTS,
the 80mm fan on the front, the 3-fan 5.25" bay vent, and
two Antec Cyclone slotfans (one below my sound card, one below my video card).
I also have two 3.7" fans that are just waiting to be put in (one from an old 386, one from an ancient Bernoulli drive). If I did happen to find places for those two, then my fan total would go up to 14.
No, I'm not overclocking this system; I hate overclocking. I just hate heat and dust buildup inside the case.
Re:OK, (Score:3)
There are two fanless computers already (Score:2)
What am I talking about? The iMac and the Cube, both by Apple. Both are completely convection cooled, and the only sound you hear is the clicking of the hard drive (and the sound of the other poor bozos getting fragged in Q3A
OK, so I work for Apple, but c'mon folks, do the design right and you don't need a fan!
--Paul
Problem: Processors are too hot. (Score:2)
While I understand that electronics will (for the foreseeable future) generate heat, it seems to me that it is just as important to find ways to make them run cooler as it is to find ways to cool them. Apple's latest iMac line (I think) was convection cooled - but the monitor, processor, and hard drive give off enough heat to make the entire machine very warm. It definitely a step towards a quieter computer, but an iMac won't suffice for your average Slashdot reader.
As for microfans - they're not really even microelectronics. They belong in the region of "mesoscale," which means macroscopic but small. I saw pictures of Apple's latest G4 (which rivals the Pentium in terms of energy consumption) in which Apple had encased the entire processor card in plastic to dampen fan noise.
Anyway, just some thoughts.
For those interested in Science (Score:1)
The link. [wired.com]
New Scientist is NOT a reliable source (Score:1)
Great (Score:1)
As for noisy hard-drives this probably won't be a problem as you'd probably not want data to be stored away from devices that you physically interact with and accessible from several of them. Each home could have a server built in in the loft with fast networking around the home. All data is stored on the server, then lower performance systems could be installed next to your tv/hifi etc.
Good idea (Score:3)
/me looks around the room and counts.
Okay, i have some 6 computers sorrounding me in this real small room. I live in Australia. The Temperature outside is over 40C. (110F). Its hot.
Luckily, we have the A/C on, so its real cool in here. But the computers make FAR too much noise with all the darn fans.
I'm not even going to begin counting how many fans my main workstation (Dual PIII-500 512mb ram) has. Well. Okay. Cpu fans, 2. Case fans (extra added by me), power supply fans, and even fans on my CD-R.
And thats not my only Dual CPU box
Computers definitely have a problem with heat, and shoving ever more fans into cases is not the solution. New tech such as this, is.
When the power goes out, its almost surprising at the silence around you
D.
Re:I like my loud computer. (Score:1)
I think a real solution to that would be some kind of filter. I have one cast that has a filter over the fan in the front of the case and that works very well, I think that would be better then more fans.
--
Re:There are two fanless computers already (Score:1)
Why? because they overheat and shut off. There was also reported problems with the cubes doing this.
--
one meg rpm? (Score:2)
Re:I like my loud computer. (Score:1)
---
Re:Why not collect that heat? (Score:1)
Re:What about (Score:1)
OC possibilities....*grins* (Score:1)
Cries of "I have 12 fans on my computer" from all the OC sites will rapidly be replaced by "I have 300,000 fans on my computer!" However, things like Sub-zero alcohol cooled PC's [tech-junkie.com] will never change.
47.5% Slashdot Pure(52.5% Corrupt)
fan failure? (Score:1)
Re:I like my loud computer. (Score:1)
Still, the computer itself has a commanding stance: it's a full tower standing on my desk, and you always know when it's on (whirrrrrrrr!!!!). As you can probably tell by now, I hated the G4 cube; it was too damn quiet (of course, I'm leaving out its other downfalls).
Good Feng Shui (Score:2)
I've lately heard a lot of good things about how quiet MacIntoshes are. Since I've been making this the year of my Microsoft purge, I have become self-sufficient in GNU/Linux with the help of the local users group. But lately I've become aware just how much noise fills a room with a single tower... or even with no tower and only a 3com Superstack 24port hub, running its fan. Such a harmful shar can be maddening when compounded.
There has got to be a more efficient way to "recycle" wasted energy in a system... particularly in notepads. If a modern CPU generates 25watts of energy into raw heat, and a fan is required to cool it, reducing battery life, there's got to be a way to use that more efficiently. If the CPU heat can not only run its own fans, but maybe also backlight a display or something else useful, then the waste of heat and noise are replaced with greater efficiency.
Apple anyone? (Score:3)
Re:Why not collect that heat? (Score:1)
Re:I like my loud computer. (Score:1)
Most likely not, I live in Massachusetts, 20 miles west of Boston (as the train rolls, not as the crow flies).
Re:Why not collect that heat? (Score:1)
Sorry to dampen our enthousiams, but the first or second law of thermodynamics prevent this. In substance it says that you cannot do it, because you will create heat in the process anyway.
Or is that a new kind of thermodynamic trolling ?
It's wAIfer thin mOnsieur :) (Score:1)
Re:more fans - more power - more fans - more power (Score:1)
progress ? (Score:2)
Come on !
i just think we should take a deeper look to the low-consumption alternatives around, like this [transmeta.com], or this [arm.com].
What ? Vapourware. Nope. I own many machines running these processors and my brother just bought a transmeta laptop which he's in love with.
Don't believe the hype and aim your purchases towards a brighter future.
Intel's selling radiators, so is Nvidia.
--
can they.. (Score:1)
Of course this could lead to propagating smells, while cooling off the person wearing the special micro fan shirt...
E.
www.randomdrivel.com [randomdrivel.com] -- All that is NOT fit to link to
Re:I like my loud computer. (Score:1)
"growed"? (Score:1)
Fan Speed (Score:2)
Re:Why not collect that heat? (Score:1)
"Titanic was 3hr and 17min long. They could have lost 3hr and 17min from that."
Re:Pedantry (Score:1)
You can reset that default in your user prefs.
It is too bad that most of the slashdot readership caint ferm gud sentanses if there live dependsed onit. As usualy, the greater the size of the community, the lower the average IQ level. Oh, well [/useless bitching]
Mr. Chips rocked.
--
Re:What about (Score:1)
I can sense my brain exploding already.
Re:Why not collect that heat? (Score:2)
We already have such a "heat exchanger" -- it's the heatsink and fan already on your CPU. And yes, it does put out quite a bit of heat. The 7 computers in my bedroom keep it pretty toasty, way warmer than the rest of the house. The power source and hard drive contribute quite a bit to the amount of heat that comes out, but I think a significant amount of heat comes from the 9 CPUs.
Enigma
Re:What about (Score:1)
Re:I like my loud computer. (Score:1)
Silent Hard Drives Here? (Score:1)
Re:Why not collect that heat? (Score:1)
Re:can they.. (Score:1)
Re:Good Feng Shui (Score:1)
or perhaps what a harmful shar is.
Certain computers have lost their fans already! (Score:1)
Has anyone ever measured the AMD Athlon's contribution to global warming through both its excessive voltage requirements and heat dissipation.
I've been waiting for someone to bring out a 5 1/4" Toasted Sandwitch drive module for my overclocked dream machine. I doubt it would even need a heating element...
Man, I've hardly had the heating on this winter.
overlooked... (Score:1)
Anyone have an idea about how much noise would be generated by a ton of these fans all going at 1 million revolutions per minute?? It seems to me that it has the potential to be even noiser than what we have now.
(OT) - Growed? (Score:1)
Come on... are we all five here? This is just ridiculous.
Re:Reversable logic (Score:2)
Hearing problems due to computers (Score:1)
Re:Why not collect that heat? (Score:4)
The laws of thermodynamics do not prevent the useful collection of waste heat. What they prevent is achieving 100% efficiency by doing so because your collection and transformation system will loose some heat, not create necessarily create heat
Re:There are two fanless computers already (Score:2)
Noise Good (Score:3)
Fan noise does have an upside, though. In my home office I find that the noise from my SparcUltra10, 2 Regular PC's, Rackmount PC, and 3 laptops drowns out the noise of my wife
If the room were silent, I would probably have to respond to her calls to come down and take out the trash or something.
What he didn't mention... (Score:1)
...becuase I need to compensate for the load that my Windows 2000 box puts on the CPU.
Re:(OT) - Growed? (Score:2)
Re:There are two fanless computers already (Score:2)
Ack! (Score:1)
Re:Why not collect that heat? (Score:1)
fans are great, but better chipmaking is the way (Score:1)
Re:What about (Score:1)
Duh!!!
Re:There are two fanless computers already (Score:2)
Re:I like my loud computer. (Score:1)
J:)
In other news... (Score:3)
University of Colorado researchers state they never had the intention of using these micro fans outside of the computer world. Apparently, they underestimated the creative will of some insane scientists.
Protestors outside the research facility held up signs saying, "It's the end of the world! Pigs are flying!"
-------------
Seriously though, I think they are "cool" and can't wait until I can buy a chip with them on it. Hopefully, they will sell sheets of these fans to the consumer market. Plus, these could have great influency on small computer designs in which the heat could be dissapated more quickly and efficiently.
Amigori
------------
Duck! No, that's a pig flying!
Re:What about (Score:1)
Re:OK, (Score:1)
I can just sit in a server room for hours sipping Jolt and tweaking the shitty vbscript pages and ActiveX modules the IT guys throw at me all day long. I don't get headaches, my ears are perfectly fine and I can still tell the difference between a CD burned from mp3 vs the real thing.
Re:Why not collect that heat? (Score:1)
Re:There are two fanless computers already (Score:2)
--
Re:Hearing problems due to computers (Score:1)
Another thing is noise-induced stress as a psychological problem. There is an "ecological" quality label in Germany that actually specifies a maximum noise level (IIRC around 45 dB(a)).
In a job several years ago, I was working on the noise measurements in QA, and boxes with cheap, lousy fans had problems meeting that limit.
What is Feng Shui? (Score:2)
It has many applications. On one level, it deals with furnishing, landscapes, and building to increase comfort. On another level, more important to the work my wife and I do, it is about removing offensive stimulii, or balancing the qi/chi/whatever-you-call-energy. This does not mean maximizing efficiency in a machine like manner, because having long straight hallways, or doors evenly opposite each other in halls, can "point" offensive "energy"(noise, flow, stress) at a person.
I've known architects to redirect long hallways, or split them up with fire doors, just to slow the flow of a place into more pleasing directions. Ironically, this seems to parallel electronics in a metaphorical way, balancing "resistances" where an inducer could do harm. Stressful positions, like having your back to the door or world all the time, can make a person paranoid ("Big brother is watching" or "I could be stabbed in the back!"), and such offensive stimulii are called shars , which I think means "poison dart." Feng Shui prescribes remedies, such as having a desk mirror to see who's behind you.
In a world where it's easy to go hard-of-hearing amidst computer equipment, Good Feng Shui should be considered in this design, as it is in any other field of design. Microfans could be far more harmonious, or quiet (yin), than conventional cooling methods.
PS: I've also made little microcontroller "pets" whose LEDs simulate breathing rhythms. It has a cool, soothing energy about it.
Sorry, but all moving parts prone to failure. (Score:1)
Fans are out in this environment. Fans gunk up with dust and die. And a trip to fix it costs thousands of dollars. Actually *anything* with moving parts is out. The oil in fan motors and even in hard drive motors can gel up when it gets that cold. So we use solid state flash+static ram drives, and 486/25s that need no fans. Cursoe may finally be the next CPU upgrade because it runs cool without help.
Cooling fan tip (Score:1)
Re:Problem: Processors are too hot. (Score:5)
Apple encased the whole G4 processor card in plastic to dampen fan noise? Not really. There IS no processor fan on the new G4 models. There's a huge honkin' heat sink on it (which sits next to the power supply and an external vent when the door's closed), but there is no direct cooling on the processor. So, no, I don't think you've seen a picture of the latest G4. If you had, you wouldn't've claimed they encased the processor in plastic. Wouldn't that defeat the purpose of cooling? Heat can't escape through plastic as easily as it can through air.
The hard drive in the latest iMacs don't make that much heat, actually. Apple uses three kinds of hard drives in their latest lines: Maxtor, Seagate, and Quantum. They all run rather cool, with the Quantum being the hottest of them all (this is all subjective, and I haven't scientifically measured this stuff). The Seagate drives are definitely the quietest, though.
The micro fans might be nice on paper, but how long do readers think it'll take for MAJOR chip vendors to implement them? The heatsink/fan combo has been with us for as long as I can remember, and considering how cheap and easy it is, I don't see it changing that much very soon. We need cooler processors, not better fans/heatsinks.
Path of least resistance, I guess.
Here's a Pentium III Datasheet. If anyone can find the wattage for the P3 in this marketing mess, I'd appreciate knowing it: http://www.intel.com/design/pentiumiii/datashts/2
Here's a PowerPC 7400 Datasheet: http://e-www.motorola.com/brdata/PDFDB/MICROPROCE
And here's a PowerPC 7450 datasheet: http://e-www.motorola.com/collateral/MPC7450FSR0.
Re:Fan Speed (Score:1)
Re:OK, (Score:2)
Microfans won't reduce the noise (Score:1)
I doubt those microfans really would reduce the noise generated by the fans. First of all, the efficiency of air cooling is determined mainly by the volume velocity of the airflow, and you'll still have to blow the same amount of air with the microfan technology -- with fans with no aerodynamic/acoustic design. Size of the fan does matter, though. Usually bigger is better (quieter). Blade noise is easier to handle with big, slowly rotating blades. Microfans seem to move into opposite direction, suggesting they might be even noisier than conventional fans.
There are ways to get rid of the noise, though. This site [swipnet.se] is a good starting point for screwing that silencer on your PC. Particularly, there was an interesting link to a Korean company [cnssystem.com], which is going to introduce a free-flow refridgeration system for computers. With that, you can throw away every (except the power source) fan from your computer. Hopefully it'll work as well as they tell.
Re:above should be moderated up. (Score:2)
Re:Why not collect that heat? (Score:1)
Re:Why not collect that heat? (Score:2)
1) You can't get anything done without working for it.
2) The most you can accomplish by working is to break even.
3) You can only break even at absolute zero.
-Mith
PS
(From the original story) "growed"? Is that a word?
--
Re:OK, (Score:1)
Kiwaiti
Re:Why not collect that heat? (Score:1)
Thermodynamics (first law) says that energy is neither created nor destroyed. So how can the universe 'lose' energy ?
According to the second law of Thermodynamics heat will flow only in one direction when there is no external work done on it (from hot to cold obviously). Applied to the universe this means that the universe tends to 'even out' temperature differences until an equilibrium is reached.
The Carnot corollaries to the second law state that the amount of work obtained from a system between 2 thermal reservoirs is bound by a maxiumum efficiency. This efficiency is directly related to the temprature difference
All in all it means that if you have a certain amount of heat at a high (absolute) temperature the work that can be obtained from it is much higher than work obtained from the same amount of heat at lower temprature. After heat (witout losing it's energy) reaches a certain temprature, it becomes virtually useless because everything in it's surroundings is the same temprature
The heat from fans is at a relatively (to the surroundings) low temprature (low quality heat) and so the potential for work being done is very small due to the low theoretical maxiumum Carnot efficiency.
Re:Big fans (Score:2)
This site specializes in almost silent power supplies and processor fans. I've got several in my various systems, and I can personally vouch that they're all that they claim and more.
Disclaimer: I don't work for these people.
Fan noise?? (Score:1)
10 compaq proliant 3000 servers (ML530 now) sound like you stuck your head in a jet engine.
I see uses for silent hardware, like a library. But then you don't have to resort to exotic hardware to achive this. Just placing the computer in a closed case that is well ventilated or purchasing a premiuim enclosure in the first place. (Hard drives on rubber isolation mounts, and soft material grids on fan openings with rubber isolation mounts on the fans) make a huge difference. Try it. add rubber o rings around the mounting screws and then dont tighten them down so hard. makes a big difference. a little bit of engineering can achive huge changes.
and how do we get air to these fans? (Score:1)
So what are they going to do? Drill little holes in the casing? That'll do wonders for manufacturing costs, I'm sure.
Noise is one thing... (Score:2)
I could knit a sweater out of the dust I've cleaned out of fans over the past 3 years.
--
Re:Microfans won't reduce the noise (Score:2)
Re:There are two fanless computers already (Score:2)
But iMac bashing always gets good karma around here.
Re:Sorry, but all moving parts prone to failure. (Score:2)
If it needs a fan, then that's a sign of a BAD DESIGN.
Are you a designer?
I install radio repeater controllers on mountain tops for various communications companies (radio, TV, cellular relay, business 2-way, etc.)
Nope. Kindly keep your opinions stated as such and not as fact. Fans do NOT indicate bad design.
Now to get on with business: Fans in designs are NOT a bad thing if the design calls for them. Your communications equipment won't be spitting out too much heat since the control likely runs off of 24VDC bussed from a single (redundant) power supply which can probably endure a colder environment. Your actual transmitters will throw out heat in proportion to their strength.
Then you go on to say that the ambient is 80 below (degrees Farenheit) -- of course you won't need much to get rid of the heat! You've got a freaking 200 degree C temperature differential! If I'm not mistaken, you'll have heaters in the enclosures in order to keep the temp within component tolerances!
Now let's come to the Real World (in the sense of consumer equipment) -- People like to be in a 20 degree C ambient, so their equipment will be in there as well. Commercial components are spec'd to operate between 0 and 70 degrees C. So there's 50 degrees to play with there, not counting heat thrown out by the power supply, hard drives, motherboard chipset and expansion cards. Heat sinks can only distribute and radiate so much heat. While I'm not saying that 60W for a processor is an efficient design, it is by no means poor. Poor design is when you don't meet the design spec.
In the world of industrial equipment (I design for this environment) your components are rated -40 (I think, we don't run into the low end much) to 85 degrees C but you usually use commercial-rated components since your rarely in an enclosure that gets below zero. We too have equipment in the middle of nowhere where it's cold enough to freeze spit before it hits the ground and the location is unmanned. Know what though? It's the hot and/or high elevation remote locations which give us the most trouble. We literally have equipment in the amazon rainforest and in the Chillean mountain ranges (Andes?). Even when we have our power electronics bypassed we can't avoid generating 1 Watt / Phase / Amp (three phase equipment). This stuff has to go into NEMA 4 enclosures to keep the crap out. Fans are unacceptable here (watertight) so we need to oversize the enclosure in order to increase the air volume inside the enclosure and help get the heat radiated. In some cases we have to use industrial air conditioners (want to talk expensive? Try a NEMA 3R air conditioner!) in order to keep the heat down and that is in a design where we are already generating minimal heat! (you can't get much less minimal than a slab of copper!)
Fans are out in this environment. Fans gunk up with dust and die. And a trip to fix it costs thousands of dollars. Actually *anything* with moving parts is out. The oil in fan motors and even in hard drive motors can gel up when it gets that cold. So we use solid state flash+static ram drives, and 486/25s that need no fans. Cursoe may finally be the next CPU upgrade because it runs cool without help.
Yes and you have the money to spend on lower power equipment and cold-temp tech. The industrial world has MUCH higher price margins than even the least competitive commercial sales environment. Please try to keep that in perspective before bashing commercial designers.
Fans are cheap, not perfect. Just because you have the advantage of a 200 degree temperature differential to improve your radiated heat transfer doesn't mean we all do. And just because you can afford to sell expensive technology (flash vs HDD) doesn't mean we all do.
Re:OK, (Score:2)
i also noticed a slight ringing in my ears that was later confirmed to be the beginnings of Tinnitis. i now wear earplugs when i DJ.
anyhow i now live in downtown Toronto a few blocks from a hospital, and about half a block from an
- j
Re:OK, (Score:2)
Growed??? (Score:2)
--
Big! Small! Meso! Meso! Meso! (Score:2)
Communication throughput is suboptimally parameterized when culturally marginalized nomenclature ("jargon") is specified in negative relation to nomenclature which is itself based on logical contradiction stemming from inconsistent taxonomy.
Putting it another way: the "micro" in "microelectronics" means "small" not "microscopic". So if "macroscopic" means anything in this context it means "big". So you just said "big but small". Hoist by your own jargon.
__________________
Re:one meg rpm? (Score:2)
well the surface area of your standard industrial size heat sinks is many times the surface are of your standard cpu, for example.
The impression I got was that these would be surface mounted on the cpu itself. (I could be drastically wrong, of course)
Applying these to the heat sink would not be so bad, but I am not sure of the cost benefit angle.
One use for the heat (Score:2)
Re:Apple anyone? (Score:2)
Re:Pedantry (Score:2)
Maybe people are just too busy trying to submit things quickly to check them over... and maybe I'm just too bothered by simple mistakes 8^)
--
Really quiet computers (Score:2)
I haven't tried this yet, but I want to make a silent computer with no hard disk at all -- it would boot from a network card.
With a 100Mbps full-duplex Ethernet connection, a decent network switch, and a server with a fast hard disk tucked away into a clost, I believe that a completely diskless workstation would be nice and fast. 100Mbps is about 10MBps, which is exactly the speed of a fast narrow SCSI bus; not that bad. Just put in 256MB of RAM so the system doesn't need to swap. (Last time I checked, you could get 256MB of RAM for well under $200!)
I'm typing this message on a computer I built, and by far the noisiest part of it is the CPU fan. (Anyone know of a really quiet Socket A cooling fan?) That's why I would love to buy one of those Transmeta Crusoe server-edition CPUs. With a big heat sink I wouldn't need a cooling fan.
I have hopes that IBM or HP will make one of their "legacy-free" managed PCs like this. Then all I would need to do is just buy one.
I have fond memories of typing on the Atari 520ST we used to have. No cooling fans, no hard drive... unless the floppy disk was whirring quietly, that thing was silent. Oh yes it was nice.
steveha
G4 in various context (Score:2)
Clock for Clock
$$ for $$
'Performance' for 'Performance
Clock for Clock, it would seem that both dissipate the same amount of power; 14W
That, however, doesn't tell us how much 'performance' the processor generates per Watt, as it were. A Gateway Select 1200 with similar options (but a faster processor) $2341 vs an Apple G4 667MHz tower for $2799.
So there is definitely a $450 delta between the two. The G4 gives off 14W, the Athlon at ~55W. If we want, we can do the math that 2x MHz and 3.7x energy dissapation.
As per performance, everyone thinks/knows that a G4 on Photoshop beats the pants off anything else on the market, supposedly, but we have that on a clock per clock, the G4 supposedly outperforms but has the same wattage, while at max MHz, the G4 *still* supposedly outperforms and uses much less watts.
Now, how about non-Photoshop? I dunno.
Geek dating! [bunnyhop.com]