Watercooling Drifting Mainstream

pacc writes "With Prescott said to dissipate 103 W and the dual Apple G5 playing in the same league, air cooling seems less than sensible. Nikkei Electronics has an article about watercoolers getting standardized by Hitachi. A technology pioneered by a NEC desktop last May."

Comments

Go to the junkyard instead

[corebreech]

Watercools his system using a radiator from a '55 Lincoln. You gotta love it.

Not a bad looking box, either (though I usually end up looking at my monitor more than I do my computer case.)

It seems to me that with all the concern over cyber-pollution these days (discarded monitors and other computer components) maybe it's time to take a greener approach and harvest whatever relics we can from the last great love affair with speed and power: the automobile.

The trend is towards customized boxes we build ourselves anyways, right? So go to the local junkyard and shop American for a change.

Re:Go to the junkyard instead

[sharkey]

My pappy said, "Son, you're gonna drive me to drinkin'
If you don't stop moddin' that Hot-Rod Lincoln!"

Nope, just doesn't have the same ring to it.

Re:Go to the junkyard instead

[gmhowell]

To join in with the peanut gallery: it's not a radiator, it's a heater core. OTOH, it's larger than the radiator on many motorcycles, is constructed the same way, and does a similar job.

The guy did some great work, but the English wheel to make a simple curve was big time overkill. English wheels are used to make compound curves, usually.

As far as the 'last' great love affair with speed and power being the automobile, America's love for speedy and powerful autos is as strong as it ever was. Fast computers are barely a glint in the eye for the average person. Hell, even most geeks who really make it buy rather nice cars (ask John Romero). And the lowly Dodge Neon is quicker than most average Dodge musclecars of the late 60's, with superior economy and handling. Only seriously high end race only cars back in the day would stand a chance at hanging with something as relatively mundane as a Subaru WRX. Maybe a Yenko or other tuner car could beat them, but then you have to let me mention tuner Corvette's and Mercury sedans. Trust me, *this* is the golden age of the American auto, despite the prevalence of SUVs and trucks (which are quicker, safer, more fuel efficient, more powerful, and more durable than their brethern 'back in the day'.)

While I'm wound up, let me tell you why emacs rulez, and vi is teh suxx0r...

This can be a good thing, if...

[Falconpro10k]

this could be great if people knew how to service them properly, in my own mind, watercooling is more effective than aircooling in many applications (cars, computers etc) but CARE must be exercised. What was once a hardware hacker's toy is now becoming mainstream, this is a VERY good thing.. .

Air vs Water?

[Davak]

Has there ever been a head to head with air-cooling vs water-cooling?

Water better be damn good to risk my system to the exposure of fluids.

Davak

Re:Air vs Water?

[Aadain2001]

Yes. It's much like comparing a 486 to a P4 3GHz. The water cooling is much quieter, runs much cooler, and allows for creater control over temperature. If done right, there is no risk of frying your system, unless you cut the hoses or something, but even that can be protected from. Plus, if you have any kind of case window and lighting, you can injust florecent(sic) coloring into the water system and use clear tubes, giving a very cool light effect.

Watercoolers

[daeley]

I think watercooler computers are a bad idea. I have enough trouble getting interrupted in my cubicle without a crowd of people wanting to stand around my computer talking about yesterday's episode of "American Idol 4: The Revenge."

Wow.

[lifebouy]

Seems like forever ago when I first saw a water cooled system. But I never thought I would see the day it went mainstream.
OTOH, the media gave it a push lately, so what you are witnessing is probably a shortlived fad. Not that it isn't cool. (No pun intended.)

Re:Wow.

[El]

Seems like forever ago when I first saw a water cooled system.

Yeah, remember back when Gene Amdahl introduced the innovation of an air cooled computer back in the '70s? Up until then, they had always been water cooled... this ain't new technology, folks!

Comparison?

[insecuritiez]

I overclock. I run a decent cooling fan. I have never seen solid comparison results between water cooling and just high-performance fans. If I (and the public) were to see dramatic improvements published in say THG or some other more mainstream publications perhaps water-cooling will gain even more ground. But as it is I have never really seen anything that has jumped out at me and said "go water". If it is so good and is gaining more ground then why haven't I seen more about it? Slashdot educate me!

Re:Comparison?

[Sevn]

http://www.water-cooling.com/
Water cooling obliterates air cooling. I used to run a thermaltake slk800 with a 120mm fan wired to it with an adapter. It pushed 80cfm and kept my seriously overclocked athlon tbred 1700+ running at 38c at idle, and never over 48c at full load. With water, I haven't seen 40c yet under extreme load. I idle at 33c. And I have a crappy thrown together setup right now. There are guys that have never seen 27c.

Re:Comparison?

[Skater]

But...so what? I mean, at some point, it doesn't matter how cool it is; the only thing to worry about is that it doesn't overheat. At some point it becomes bragging rights rather than actually useful.

My processor runs at 35c pretty much constantly no matter what load I put on it. Room variations sometimes make that tick slightly upward. Removing the case cover the other day dropped it 2c.

I guess if you're going to overclock it, then you'd want it cooler. But if not, then all you really need to accomplish is keeping it below "lockup/meltdown" level.

--RJ

Re:Comparison?

[MBCook]

Lots of sites do many MANY reviews. Overclockers.com, Hardocp.com, and even THG have done stories on watercooling. I've been following the "scene" for quite a while now, as the noise from my PCs drives me nots. There are a few thing I can comment on:

• Watercooling is MUCH more efficent than the average stock heatsink. You can beat a cheap watercooling system with a REALLY GOOD heatsink, but...
• Watercooling is much QUIETER. In a normal heatsink, you are cooling a small area with a small fan (on the order of 60x60mm for a good heatsink/fan, but you can use an 80x80mm fan). But with the radiator that cools in a (standard) watercooling setup, you can fit at least one 80mm fan, or even 2. And since the air is designed to pass through it and over it (instead of onto it and off the sides) it's quieter. You can either run your system cold at a decent noise level, or go near silent and get fine temperatures.
• You can cool the water many ways. While most of the time you run it though a radiator, I have seem setups on the 'net that use a bong (Water is sprayed in a tube of air as a mist, it loses it's heat as it falls through the air), groud cooling (one guy buried a welding tank DEEP in his yard. He pumps water in and out, and the earth cools it for him), watercooling (you could make a little heat exchanger that runs cold water from your water pipes next to the water from your PC to cool it down), etc. You have OPTIONS.
• The biggest problem I've seen is usually the cost. This is mostly due to the fact that a LARGE number of watercoolers are overclockers, and they are willing to PAY big cash for a great waterblock and such. So the majority of waterblocks you find cost $50 or more. So if you cool your CPU, Graphics card, and chipser, you could easily spend $150 on the blocks alone if you wanted to. Most watercooling kits (that cool the CPU and graphic card) seem to be around $300. This is due both to the aformentioned situation, and low volume of sales (relative to other options, like a new heatsink).
• Customisation! You think putting a cold cathode in your PC is cool? How 'bout putting an adative in your watercooling water that under blacklights or ultraviolet lights glows a bright color. It looks REALLY cool. Check the forums mentioned below to find some pics of this.

Learn more, it is facinating. Look around the old articles on HardOCP and Overclockers.com and you can find out a ton. Just search google! Also, if you look at like the HardOCP forums under cooling, you can find tons of pics of people's Watercooled PCs.

Impact on Productivity

[wildsurf]

Finally, an excuse to hang around the water cooler all day...

Not sure this is a good idea

[JFMulder]

To have mass public acceptance, it has to be pretty cheap to buy. And by being cheap to buy, it may also be cheap material, or sub-par, so it may have more chances to leak. I've been burned (more exactly my CPU) twice by a cheap CPU fan and an AMD CPU fan. The last thing I need now if being "flooded" by a cheap watercooler. Especially since a burned CPU breaks the CPU and the motherboard most of the time, but water spilled on a running motherboard, that's gonna do a LOT of damage.

Prescott will actually dissipate around 130W

[Alereon]

The 103W figure for the Prescott 3.6Ghz is actually the Thermal Design Power. This is the amount of power the processor is expected to use during "normal" operation. A P4-C 3.0Ghz with HyperThreading has a TDP of about 80W, with an actual maximum power usage of 104W. Assuming a similar scale, a Prescott 3.6Ghz can be expected to dissipate around 130W. It's this maximum figure that really matters, since I don't think most people want their processor to throttle during gaming or whenever they are driving their CPU hard.

Re:Prescott will actually dissipate around 130W

[wik]

Or noise! Nobody has mentioned that Prescott will actually vibrate and generate a 1.6kHz tone at 110dB. It eliminates the need for both (a) the PC speaker and (b) a premise burglar alarm.

Am I the only one...

[achurch]

who thinks it would be more sensible to find ways of reducing power dissipation rather than (or even alongside) better methods of cooling?

(Yes, I know the answer is that nobody actually needs these new CPUs, but you know Microsoft and Intel won't stand for that...)

Re:Am I the only one...

[Alereon]

Intel Prescott, the chip that's slated to set power dissipation records for mainstream CPUs, uses a new .09 micron process. Apparently Intel is seeing fewer benefits than they expected.

Re:Am I the only one...

[randyest]

And about the only way to do this without sacrificing clockrate is by going to a smaller fabrication process.

Sorry, that's commonly believed, but wrong. There are lots of ways to reduce power consumption. Reducing gate widths (0.25um -> 0.13um -> 90nm) is commonly touted as a good way to reduce power, but in most cases that's more marketing pitch than reality.

First, there are two types of chip power to worry about (1) leakage, which happens all the time, just by being on, and which used to be always much much lower than (2) the switching power, or maximum dissipation when as many transistors as possible can switch at once (which, BTW, can never be all of them, and it's really, really hard to find the stimulus that makes maximum power happen. So, esitmates like the ones in the article for peak power are often made assuming a somewhat-arbitrary switching factor that may be low or high).

As gate sizes shrink, the effective capacitance of the gate shrinks, and voltage can be lowered (to a point). Capacitance varies with gate area and inversely with distance between "plates" of the gate (C = k*A / d). Reducing the gate width (space between the plates) actually increases capacitance, and this itself would increase power. But, you're also able to reduce the gate area (though not as much, but in 2-dimensions, so shrinking gates is usually a reduction in C). Most importantly, you can decrease voltage, since power varies with the square of voltage, this has much more impact on power than reducing gate capacitance (size). When we went from 0.25um (3.3V)to 0.13um (1.5V), we got a nice fat 1.8V drop in voltage. But 0.13um is 1.5V too, or 1.3V at best, and I've never heard of a 90nm (0.09um) process under 1.1V. The V isn't dropping as fast any more because the noise margins are getting too small.

Since p(switching) = 1/2*F*C*V^2 (F = clock freqyency, C = capacitance, and V = max voltage, lowering C (and moreso V which we can reduce some, but not much below 1.0V so far) will lower power a bit. Linearly with C. But unless we can reduce V, reducing C much more won't help a lot because we have more total C's (transistor gates) on the die, because they are smaller we can fit more.

But now, at 0.13um, and more at 90nm, it's not the switching power, but the leakage (always there) power that's getting worrisome. It used to be 1/20th of switching power or less, but now the gates are so small current of the same order of magnitude (almost) of switching leaks all the time.

So, the more you shrink, the more you have constant power, which is harder to deal with since you can't throttle it, and it's always cranking out. Worse yet, the more you shrink, the more gates you can fit on one tiny little die (the feasible mfg'able die size stays around 17-18mm max regardless of gate size once the process matures a bit, but bigger dice have ridiculous failure rates and thus silly high prices). And the gates shrink in 2 dimensions (L and W), so you get a squaring increase of the toal gate count, and only a linear decrease with C. Shrinking gates to save power doesn't work.

So, if we can't keep shrinking to save power, how can we? Lot's of ways. There are dozens of EDA companies with power-minded RTL coding, synthesis, and even place and route tools ready to help you reduce your power if you have a few $100k/seat/year. Or, you could use a SSC (Spread-spectrum Clock, where each clock edge is off by a bit to reduce power, but it slows down the max clock rate a bit too, of course). You can also try to use beneficial clock skew to reduce power after timing closure, or gate the hell out of all the clocks and only enable what you need (a la mobile chips). Or switch to asynchronous, or self-clocked design (every thing has it's own clock, which sends a clock to the next thing, etc. -- it's HARD to design!). Anyway you look at it, it's a hard problem. And people who

Solve colling with a new case Mod.

[jellomizer]

Get a portable Freezer or refrigerator put the computer parts in it. Find a way to keep the humidity out. Put a couple of ports for for USB and monitor and your all set.
I feel that heat is becoming a major problem with making faster processors. You guys in college should quit your Computer Science and Engineering and go into thermal physics. That is where the future is in.

Re:Solve colling with a new case Mod.

[insecuritiez]

Wont work, here is why:

The freezer you are speaking of is a heat pump. It moves heat from the air inside it to the coolant which then dissipates heat via coils on the back. What advantage do the coils have on the back that the blades on a heat sink don't have? Surface area? Maybe, but that could be changed with better heat sink designs. Besides, the heat sink has a fan to move the air, the freezer doesn't have air moving across those coils. The freezer may sound like a good idea but when you get down to the thermodynamics of the beast it too is using air dissipate heat.

Re:Solve colling with a new case Mod.

[toddestan]

The heat coils have a lot more surface area, and are located outside of the fridge. Most fridges simply use convection to cool the coils, and thus no fans (I have seen some fridges that do use a fan though). The compressor generally makes much less noise than the computer's fans. Fridges contain a lot of insulation and are air-tight, and thus noise cannot get out, so chances are you could put a ton of fans inside the thing and not hear them. This is the theory.

But a standard little dorm fridge will not work. They are not very powerful units, only capable of moving as little as 100 BTU/hr of heat out. (Most window air conditioners can pull out 5000 BTU/hr or more). 100BTU/hr is about 100,000J/hr or about 30W (30J/s). Now this is fine for cooling your pop and salsa and holding it cool - but most computers pump out considerably more heat, the new Intel processor all by itself puts out 3 times as much heat! What will happen if you try this mod is that the poor fridge is not going to be capable of pulling out that much heat and the computer is going to roast itself in the insulated fridge. Not good.

A regular full sized fridge can pull out about 650BTU/hr, or about 190W of heat. This is closer to workable, but considering 350W+ power supplies are the norm now, even then I think the computer will slowly roast itself in the fridge anyway. And you also have the problem of wear on the compressor which will be running almost continously, and will probably burn out quickly. Not to mention the energy use of running a fridge like that.

I think the only solution for a super-cool rig will be to either get something like an aquarium chiller, that can take out something around 1600 BTU/hr and is already setup to chill liquids (perfect for water cooling I guess), or work with trying to convert a window AC unit as a computer cooler. You could also try full sized freezers, especially the top-loading ones. I don't know how powerful they are, but they are probably more powerful than your standard fridge.

Gives new meaning to "watercooler conversation"

[vudufixit]

Hmmm... what exactly do two hot CPUs talk about by the watercooler, anyway? How much better than DDR guy was than the Rambus man?

My dream setup

[Sevn]

The Ultimate Waterblock

Ultimate Pump

Ultimate Radiator

Two of these to cool the radiator at only 30db

Round it out with a Cool Reservoir and some tubing. Maybe toss in a GPU cooler. Plenty of pump to support it.

Not really

[MasterVidBoi]

Less then sensible? Maybe you just need a better air cooling design. Since the G5 was brought up in the post, it seems reasonable to mention that Apple is really pushing the idea that the G5's are quiet*:

http://www.apple.com/powermac/design.html

If a system is having trouble dissipating that kind of heat with air flow alone (or sounding like a jet engine), then you just have a poorly designed system. And maybe it's just me, but I have some qualms about putting water in a poorly designed system.

* of course, we haven't had independent reviews yet, so...

recycle the heat with an oven

[wadiwood]

Having installed a car radiator on the computer, perhaps the heat could also be used for lunch - although right now I'd be happy with a heater, or an evaporative cooler in summer.

The problem I think...

[corebreech]

...is that most watercooling solutions try to keep it all within the case. So you end up putting the radiator where one of the fans normally goes, and then mount a fan on the radiator to blow air over it to cool it down.

But which way does the fan blow? I think most people end up having it blow hot air out, which means you're not cooling the radiator as much as want to.

But if you have it blow the air in, then you're essentially pumping warm/hot air back into the case, which seems counterproductive.

I saw one comparison (and no, sorry, I don't have the link anymore) where there was almost no difference between the air and water, until they moved the radiator out of the box. Then the difference was very dramatic.

So it's like we need a mini-case for the radiator. A little clunky maybe, but I can live with that. Not everything belongs in the case anyways. One of the best improvements in my computing experience was when I got the Soundblaster Audigy External, not necessarily because the Audigy is a great soundcard (which it is) but because all the noise generated by the EM within the case is no longer audible.

The soothing sounds of silence

[IncohereD]

People who aren't hardcore about there overclock are starting to get annoyed by all that racket. I know a friend of mine who upgraded his CPU just to get a quieter fan with the upgrade.

Creative Labs is also keen on marketing 'whisper quiet' cases so people can actually hear those ridiculous SNR values they claim.

Power Supply

[Dylan2000]

This is by far the loudest component in my system. Everybody seems to think that the CPU fan is the biggest culprit but that's only because they have crappy CPU fans. It really doesn't cost much these days to get a 'silent' CPU fan which can be undervolted to be truly silent and still cool these nuclear reactor CPUs that modern PCs have.

After the CPU fan the hard drive is the loudest, but since the Seagate Barracuda IV - the best hard drive ever in the world, which is freakily quiet - hard drive makers have been using the fluid bearing system and I guess most new hard drives are now as quiet as the Barracuda IV.

That leaves case fans, which can be silent and graphics card fans, which apparently are getting quieter too (no fan on my Geforce2MX, so I wouldn't know).

So why didn't the article address the only component which can't be quietened cheaply and easily? 'Silent' PSUs cost a fortune (>$50 for something that most people expect to just be built in to their $30 case) and are far and away the biggest obstacle standing between sanity and tinnitus. I know they must be coming because manufacturers aren't dummies, but they have to realise by now that they are more of a priority that CPU fans, don't they?

Re:Power Supply

[Dylan2000]

If by slowest you mean of the 7200rpm drives then you are correct, but it's much faster and quieter than a 5400rpm drive. The thing that makes it the best hard drive in the world is not the speed or loudness that it has compared to drives from 2003 - of course they are better, it's refined tech - but that it was such a landmark product. At the time the Barracuda IV came out you could have shown me a 2.5 inch 4200rpm IDE drive with 200 gigabytes and a transfer rate better than a 10,000rpm SCSI RAID array and I wouldn't have been as shocked and impressed as I was when I heard the Barracuda IV.

Seriously, an 80 gigabyte 7200rpm drive, not slow, just not the absolute fastest, and i had to actually hold it up to my ear to hear it. The WD and IBM drives that it replaced could be heard in the next room!

For me it was the equivalent of seeing a 10 gigahertz x86 CPU shipping tomorrow, it was just so far ahead of everything which was around at the time.

It's still my main HD, even though it's only got a handful of gigs left on it and it's fast enough for my needs. Word processing? No, actually I'm doing multitrack audio recording on it and so far I haven't reached its transfer limit. Of course it has one, but it's new software and I haven't pushed it that hard yet, but right now I have a song with about 15 stereo audio tracks and the drive has no problem with that. A 5400rpm drive couldn't do that and I doubt that the fastest modern HD could give me more than a couple of extra tracks so the tradeoff is definitely not there for me.

Don't take the 'best ever in the world' bit too seriously, it's just an expression; one could say it about the Xerox Star or the Amiga. Everything will be obsoleted at some stage. Really it's just a mark of respect for a piece of tech.

Re:Peltier effect?

[MBCook]

Overclockers use Peltiers often. The problem is, that while a Peltier gets one side VERY cold, the other side gets VERY hot. Cooling this side requires you to either attach it to a waterblock in a watercooling system (which is what you were trying to fix in the first place) or put a big fan on it with a heatsink (which is LOUD and innefficent). The fact is, a Peliter would only make things WORSE as far as "cooling things quietly" goes.

Decible Scale

[trippinonbsd]

In the article it mentions:
The water cooling technology can significantly reduce the noise level. Equipped with a microprocessor whose heat dissipation is measured to reach 75W, the NEC desktop PC can suppress noise level to 33dB (A) owing to a water cooling module inside it. As its level was measured to be 43dB (A) with an air cooling system, the noise actually has gone down to one-tenth.
Do they not realize that the decible is measured on a logrithmic scale?

Re:Decible Scale

[gricholson75]

Do they not realize that the decible is measured on a logrithmic scale?

BECAUSE decible is a logirithmic scale, 43db is 10 times as loud as 33db. Hence, 33db is one-tenth as loud as 43db.

Re:Decible Scale

[russotto]

They got it right. 43 dB to 33dB is a drop to 1/10 of the former sound pressure level. Unfortunately, it doesn't mean it sounds 1/10 as loud. The dBA is roughly (VERY roughly) linear with loudness at ~40dB.

Water cooling dangers

[maelstrom]

Every friend of mine who has entered into the water cooling realm has burned out at least one CPU before getting the system stable enough to work properly. Have fun, but be safe :)

I have an idea

[prichardson]

As much as we all like have our big huge CPUs and VPUs I think perhaps it's time to rethink the "speed at all costs" mentality of processor design. A lot of companies don't even try to optimize code anymore using the argument that processors are fast enough to handle it. Then processor companies use the fact that fast processors are needed to run this clunky software (I know this is simplistic and there is also a big numbers war between processor and video card companies). I think instead of basicly brute forcing more cooling we need to design components that are more efficient (produce less heat) and design computers that can dissipate heat well (kudos to apple for thermal zones, 9 low speed and quiet fans that are controlled by a thermometer). Also, more efficient code all around is a good thing for everyone.

THG and Water

[CompuGlobalHyperMega]

Tom's Hardware Guide - search "water"

Re:Some thoughts on water

[aXis100]

The benefit of water comes from several aspects: 1) High thermal capacity - as you said, acts like an energy buffer. 2) Higher thermal conductivity than air - allows heat energy to be transferred faster. 3) Allows radiator (YES! you need a way of dissipating heat) to be located remotely from the CPU. This means you can have a much larger radiator, with far more surface area and airflow than would be possible with a CPU mounted heatsink. Remember, water is just a transport mechanism - ultimately the heat has to escape to the air. If you build the radiator large enough, the temps will be lower than you could practicalally achieve with standard air cooling.

G5s don't dissipate anywhere near that much

[RalphBNumbers]

Since when is 43 watts @ 1.8ghz, (I don't think they ever released the 2Ghz G5's power dissipation number, did they?) in the same league as 103watts?

While it puts out a bit more heat than the G3s and G4s mac users are used to, the G5 is still nowhere near as bad as prescott.
The prescott puts out more than doubble the heat.

Tips to remember for water cooling

[thriemus]

3 Tips for successful water cooling...

1: Never fill the water cooling system reservoir with boiling water from the kettle.

2: Coffee... as much as we all like it coffee _does not_ serve as an efficient coolant. (Tastes great though)

3: Dont run your water pump when there is no water passing through it. (that one is actually a serious one...)

Heatsink and watercooling roundup

[Alereon]

Overclockers.com maintains a nice database of the relative performance of various air and watercooling systems on a variety of platforms: The Heatsink and Watercooling Roundup.

Easy-Bake processing

[downix]

This is utterly rediculous. Think for a moment, water cooling, ok, sure, but what happens *NEXT*. Intel keeps pushing CPU's, AMD follows suit, they get so hot that water cooling doesn't do it anymore... what will they do next, liquid nitrogen?!?

Come on, ya won the speed war, now turn down the oven, PLEASE!

What about immersion (but not Freon)

[garyebickford]

The early Cray supercomputers (as well as the CDC6600) had Freon cooling systems. I recall pictures of an early prototype of (IIRC) the Cray II. It was one module of the new system immersed in an aquarium filled with Freon.

The high frequency EMF of the system caused some interesting color effects in the Freon, combined with the thermal gradients to make an interesting 'light show'.

Of course, we can't use Freon these days but what about other insulating oils (such as are used in transformers) & refrigerants? I haven't kept up - can modern chips handle being immersed in oil or in (for example) carbon tetrachloride? (yes, also a controlled, environmentally hazardous material)

Re:What about immersion (but not Freon)

[aXis100]

A few keen hobbyists/overclockers have done it - often using mineral oil.

In reality, it's not practical, or necessary. In fact, it's just plain messy. Most components work fine with air cooling. It's just a few hot spots (CPU, GPU, HDD etc) that can benefit.

Hot Water

[RetroGeek]

Two or three computers in the house, and pretty soon I will not need a hot water heater......

What I want is a Beowulf cluster that uses water

[multiplexo]

cooling. Then I can pipe the hot water into my hot tub where all of my hot bitches are. Yeah, I know some hot bitches, what, are you saying that I'm full of shit because I have good slashdot karma? Hey, I'll bet that lots of guys here who post as often as I do know lots of hot bitches. Really, we do, Damnit! We do!

Apple TiBook...

[z-kungfu]

...is at least partially liquid cooled. Apple has been doing this since the G3 Powerbooks. So it only makes sense to use the technology on the G5's...

Re:Apple TiBook...

[aXis100]

For the benefit of other readers - Heat pipes are a completely different animal to the water cooling we're talkign about, though they have far greater potential.

Essentially, they're an evacuated pipe with some working fluid injected. This could be water, butane, ammonia or sodium (high temps). Because of the vacuumn, some of the liquid evaporates until equilibrium is reached.

So, we have a liquid/vapor environment. Add heat at one end and local equilibrium shifts, vaporising more liquid. Cool the other end, and local equilibrium goes the other way. The pressure diffence causes the vapor to travel at the speed of sound from one end to the other, whilst the liquid flows back the other way via gravity or wicking.

This leaves you with a device that is 1000 times more conductive than copper of the same dimensions. CPU one end, heatsink/radiator at the other, and there you go!

Please, please...

[rmdyer]

Manufacturers, please, please, start putting the processors on the back-sides of the motherboards!

The back side can be one huge heat sink, with large cooling fins, just like nice audio amp gear. If need be, the entire backplane can be one extruded piece of alloy. You can even include water cooling "safely" as no piping needs to enter the case at all. The back-side is the outside of the case!

What is so hard about this idea?

+2

Navy cooling method

[EricTheMad]

Personally, I like the Navy's method of liquid cooling. The circuit boards are coated with a thin layer of rubber. They're then plugged into their sockets that are located inside of a water filled trough. Not the most elegant of solutions, but it works.

Re:Peltier effect?

[dmeranda]

Actually with Peltiers your risk can be larger due to condensation. Peltiers can cool the chip down way below ambient temperature, so water can collect. This is why serious applications of Peltier coolers include rubber seals and other devices to manage the water problem.

A passive water cooling system won't lower the temperature below ambient, so condensation is not an issue..the water stays inside the tubing, not dripping from the bottom of your motherboard. (Now active water cooling is a different story).

IBM planned on refrigeration -peltier?-

[bob_calder]

Here is a snip from IBM- There are signigficant differences in the design cycle of this series. Here is the link to the full page BUT the snip contains reference to the thread.

http://www.research.ibm.com/journal/rd/435/katop is .html

The page is VERY long and this is near the bottom so I stole it outright for reading convience . . . really!

-quoth IBM-
the MCM technology provides us with the most dense configuration of chips in a two-dimensional arrangement, which facilitates refrigerated cooling. In fact, the GEMI MCM is the key contributor to a low-cost refrigeration apparatus that can enhance the performance of the system by more than 10%. In this refrigerated operation we have shown that the GEMI MCM can support 300-MHz synchronous interconnections of significant bandwidth. However, as was pointed out previously in this paper, this frequency limitation is affected by the delay of the electronic circuits associated with the interconnections. As CMOS technology advances, glass-ceramic MCMs with thin film should be able to support even higher interconnect frequencies
-unquoth IBM- ;-)

Zalman invents fanless computer (TNN 500A)

[SlashCrunchPop]

Zalman TNN 500A fanless computer

Now, is this something most people would need or use? In terms of noise most definitely.

Good Convection works great!

[evilviper]

First of all, I would like to say that I refuse to buy any computer system that uses any more power than current AMD processors. For one thing, the fact that I live in the desert causes serious problem. Electric bills going through the roof, if for no other reason than needing to keep the cooler on 18hours ever day. This will drive people from Intel to AMD, and if AMD keeps driving temps up, then to Apple, or IBM, who's PPC chip seems to be the coolest running modern processor around.

With that said, I'd still like to talk about cooling methods. I hate the current system as much as anybody, but it can work! Really! I swear!

First of all, I'd like to say that the ideal cooling system would be to make the top of the computer a huge heatsink, and conduct the heat from all internal components to it. That would remove the need for fans in residential computers all together.

Secondly, I have seen the light, and I now understand the benefits of convection.

With the above design, I would also sell a top that hooks on to the computer, and makes a covered space of 4" above the large system heat sink. What this would accomplish is to allow ducting (standard 4", dryer-sized) to be attached.

With my XP 2000+, here in the desert, things are so damn hot already that another 20 degrees from a computer pushes the temperature from incredible hot, to inhuman, and that just wasn't working out. What I did, was to duct the power supply fan through a duct leading outside, and venting through a one-way vent. With this system, not only is it 20 degrees cooler, but the system stays much much cooler, so much so that it doesn't need air conditioning anymore. It is now only drawing in cool air, and not it's own recirculated hot air, so things stay much cooler. It's much more tolerable for people as well. As an unanticipated benefit, the noise of the power supply fan is almost completely gone, because the noise is ducted outside as well.

The best thing about ducting hot air out, is that there is really no limit... If computer makers built air-tight, insulated cases, where you could control the air input and output, you could theoretically run millions of computers at incredibly hot temperatures, and not raise room temperature at all, because the heat is all going straight outside.

Additionally, fan noise is not a fact of life, but a byproduct of saving 2 cents on a fan. I replaced my power supply fan with a $10 one, which blows more than 3x as much air, but isn't as loud. On the CPU, I replaced the fan with a $5 unit that blows slightly more air, but is about 5x quieter. Finally, I replaced the junk on the heatsink with thermal grease, and that move ALONE dropped the CPU temperature by more than 10 degrees.

Now, why in the hell aren't computer manufacturers doing anything like this? Spending $1 more on fans would get them loads more customers, and spending a few cents on heatsink grease would get them a better reputation, higher maximum operating temperature, and less need for more powerful fans. Can anyone explain why the resort to expensive, complex, unreliable, crap like this, rather than just doing the current convection methods the right way?

Re:Doh!

[aXis100]

What sort of coolant? Do you mean automotive coolant?

In comparison to water, auto coolant has:
* Lower thermal capacity
* Higher electrical conductivity

That basically makes it worse performaing, at higher risk. The only benefit is corrosion inhibitors, which are only good if you are using an aluminum block/radiator.

Water is perfectly fine for PC's - especially in an all copper system.

Shuttle has had liquid cooling for years.

[Animats]

Shuttle PCs, those little breadbox units, have a sealed heat pipe with a cooler attached to the top of the CPU chip and a heat exchanger in front of the fan air outlet. This is probably one of the more useful liquid cooling systems around. The cooling unit is one rigid all-metal component; there are no flexible tubes or liquid connectors to leak. That's basically the way refrigerators are built. It works.

Pluggable liquid-cooled rack-mounted modules are not the way to go. Ask anyone who had a liquid-cooled IBM mainframe. It used to be said of the IBM 370/168 that it needed "six plumbers and a CE (customer engineer)."

Why water?

[Zog The Undeniable]

OK, it has a big specific heat capacity, but given that you can just increase the pump speed to compensate, why not use something non-conductive like paraffin? This wouldn't wreck your mobo when a joint popped.

Prescott is the beginning of the end

[Junks Jerzey]

A chip that dissipates 103 watts is well over the line. Geez, and people used to laugh at the DEC Alpha in 1993. Sure, you can get that heat away from the CPU...and right into the room with you. This moronic race of trading an 8% clockspeed increase for a 15% increase in power consumption is the beginning of the end of modern CPU design...especially as the keyword for the 21st century is *mobile*. The last thing we want--a few years down the road--is every insurance saleman and secretary havinga Prescott on his or her desk, simply because there aren't easily available alternatives

Here's to whoever breaks the trend. Transmeta looked pretty dumb and slow a few years ago, but now the Efficeon looks to outdo the Pentium M by a large margin. But what we need now is a revolution, and not just another giant pseudo-RISC chip that trades a teensy bit more performance for over the top heat problems.

What about hard drives?

[AltairMan]

I'm not an OCer but I wonder about this...what about the heat generated by hard drives? Hard drives generate a huge amount of heat. If you cooled the hard drive using a water cooling setup, how much could you OC the CPU without water cooling it since the ambient temp in the case would have been reduced by some large amount.

If you water cool both, you could probably run the computer with only the evaporator fan.