mercury cooling system

[thaltek]

before i even start MERCURY IS A HIGHLY TOXIC MATERIAL. i know that it is and respect it as should anyone whom handles it.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

with that said i have pondered the idea of taking a regular water cooler unit and converting it to handle liquid mercury as a main heat transfer medium.... this thought all surfaced when i was investigating how heat pipes worked about a year ago and discovered that some heatsinks that have heatpipes use a mercury vapor (when heated) to transfer the heat away from the processor more evenly.

with this info, my brain was adequately stimulated to do more research.

since mercury is a better heat conductor than water one could hypothetically take the transfer rate of a good water cooled system and also utilize the thermal conductivity of mercury and have a system that simply cools better.
----------------------------------------------------------------------------------------------
i would like to know how people here would approach this design while keeping the following in mind:
1. mercury eats aluminum! mercury is a highly erosive to aluminum.
2. density, mercury is much more dense than water.
3. being very toxic, how to keep contained if a leak occurred.
----------------------------------------------------------------------------------------------
MERCURY SHOULD ONLY BE USED AN A LABORATORY SETTING WITH THE PROPER CLEANUP EQUIPMENT ON HAND! this thread is purely for academic concept and really shouldn't be done unless you happen to be a trained person in the handling of this compound.

i will post some more details and pictograms tommrow.

 

[Yuritau]

I'm gonna take this opportunity to be that guy.

... of this ELEMENT.

Fixed.

When you first think about it (before mucking about with pesky facts), an Hg system would be pretty cool (in both senses of the word). But in practice, I don't know if you could get it to work safely, and I doubt it would be noticeably more effective than straight up water.

1. Your pump is going to have to work very hard to push mercury around. The harder it has to work, the warmer it gets. The warmer the pump runs, the faster it burns out. Also, extra heat in your loop is something you generally just want to avoid, obviously.

2. Since mercury actually has a SMALLER specific heat capacity than water, it will take LESS heat energy to increase the temperature of mercury. It will conduct heat away from your components faster than water, but it will also get hot much faster than water.

3. Everyone that's seen a thermometer is familiar with mercury's behavior of expanding when heated. If you neglected to account for this in a cooling loop, you'd get a leak within minutes of turning on your system.

4. In the event of a leak, get out your wallet and buy new hardware. Mercury conducts electricity quite well, and because of it's density, it moves quite easily across even the very slightest of gradients. A drop spilled on a circuit board is almost guaranteed to short something out before it stops moving.

5. Containment.. I don't know that actual procedures used to clean up a mercury spill, but I'm quite certain you could make a plastic tray or such-like to put your tower in that would contain all the mercury if there was a leak. Wont do much if any vapors escape though, and since you'd basically be continuously passing the mercury across a heat source, vapors would be a very real concern.

That should probably be enough to talk anyone out of actually trying it, heh. And if it isn't, it's pretty expensive stuff. $71.50 gets you HALF a cubic inch. Compare that to the volume of your cooling loop (to save you some math, that's enough to fill 2.5 inches of 1/2" ID tubing). I haven't had any luck googling up a price of liquid nitrogen, but I'm positive it's FAR less expensive.

If you really want to get fancy.. go liquid nitrogen. You wont break your bank account irrevocably, and you're MUCH less likely to go legitimately crazy.

I'll concede one thing, though, an Hg cooling loop would look damn sexy!

 

[Igneos79]

I'm a chemichal technician so i can t tell you it could work, but i can also tell you nobody willy manufacture it. It is my beleif you would do well if you went for propane or even butane. It is more widely spread, as good as mercury and it isn't toxic. You can find someone to build it for you easier, and in turn, make some money with your invention.

 

[thaltek]

Yuritau i absolutely agree with you on the points you made. except for one thing.... there are those of us whom see no money limit when it comes to cool.... >_>

as for the limitations of a pump and the whole and heating up too quickly... well i think i would get a pump with a higher flow rate than the usual water pumps made for water cooling and also a pump with a higher torque motor to compensate for the density and specific heat capacity issues.... perhaps a pond pump could be substituted with say 100gph flow rate granted at that point you are now looking at 5/8" or 3/4" tubing and i would suggest getting custom copper water blocks that can handle the higher flow rate....after all we wouldn't want static pressure to build too much in the system....

now to for that pesky thermal expansion.... two potential solutions the first being a reservoir that has a second chamber for overflow like the reserve radiator fluid tank in your car..... the other idea the same thing but with a neat factor in it....... take a clear tube with some size (ie 1/2" or bigger) have one end hooked up to the reservoir and the other end sealed. orient the sealed end vertically above the reservoir in a place you can see, make sure the air in the tube is well below atmospheric pressure. and presto you now have an awesome thermometer for your cooling system....

Igneos79 also makes a valid point but, if one is to build a cooling system with chemical like butane you might as well go all out with a liquid hydrogen system..... ^_^ after all if you are going to lose your eyebrows because of a spark, at least do it with some dignity.....

 

[schizrade]

I think you underestimate the lethality of what you are proposing. When it expands, and it will, it will burst the loop, and as it is heated will release quite a bit of Hg vapor.

This is a REALLY bad idea outside of a full containment environment.

http://en.wikipedia.org/wiki/Mercury_%28element%29

 

[thaltek]

I think you underestimate the lethality of what you are proposing. When it expands, and it will, it will burst the loop, and as it is heated will release quite a bit of Hg vapor.

This is a REALLY bad idea outside of a full containment environment.

http://en.wikipedia.org/wiki/Mercury_%28element%29

i get the feeling you didn't completely read the first post in this thread...... if i remember correctly mercury is toxic as itself but once in the body will bond with chlorine atoms making mercuric chloride which is toxic in that it interferes with the oxygen transfer in red blood cells...... not to mention being a heavy metal has a nasty habit of getting trapped in brain, liver, kidney tissues causing serious long term problems..... so with that said obviously there would have to be a containment system for the potential vapor but that has already been brought up.....

secondly this thread is not intended for the debate of environmental ethics (shudders) but for the debate of actually building a system that uses mercury as a coolant and be creative in solutions that keep the user (human) safe.... the idea is to get people thinking of how to solve a difficult if not toxic problem and then be able to apply gained knowledge or ideas to other possibilities such as liquid butane, or any other possible coolant.......

i will try to get around to posting diagrams later today of some ideas, i have an exam later today so it might not be until tomorrow......

 

[schizrade]

i get the feeling you didn't completely read the first post in this thread...... if i remember correctly mercury is toxic as itself but once in the body will bond with chlorine atoms making mercuric chloride which is toxic in that it interferes with the oxygen transfer in red blood cells...... not to mention being a heavy metal has a nasty habit of getting trapped in brain, liver, kidney tissues causing serious long term problems..... so with that said obviously there would have to be a containment system for the potential vapor but that has already been brought up.....

secondly this thread is not intended for the debate of environmental ethics (shudders) but for the debate of actually building a system that uses mercury as a coolant and be creative in solutions that keep the user (human) safe.... the idea is to get people thinking of how to solve a difficult if not toxic problem and then be able to apply gained knowledge or ideas to other possibilities such as liquid butane, or any other possible coolant.......

i will try to get around to posting diagrams later today of some ideas, i have an exam later today so it might not be until tomorrow......

I did read the first post, and it is a really bad idea.

 

[thaltek]

ok, your thoughts have been noted..... this is just an academic conversation, no hard feelings......

i would also like to add in that i have a math fetish that a topic of thermodynamics can more than satisfy for a couple of hours..... after all the system would be expensive enough, that it would be more practical to see how it behaves by using hydrostatics and fluid dynamics to mathematically understand it......

 

[Yuritau]

Yuritau i absolutely agree with you on the points you made. except for one thing.... there are those of us whom see no money limit when it comes to cool.... >_>

as for the limitations of a pump and the whole and heating up too quickly... well i think i would get a pump with a higher flow rate than the usual water pumps made for water cooling and also a pump with a higher torque motor to compensate for the density and specific heat capacity issues.... perhaps a pond pump could be substituted with say 100gph flow rate granted at that point you are now looking at 5/8" or 3/4" tubing and i would suggest getting custom copper water blocks that can handle the higher flow rate....after all we wouldn't want static pressure to build too much in the system....

Hey if you wanna spend $500+ on the coolant fluid alone (conservative estimate, BEFORE adding in the cost of shipping hazardous materials), you go right ahead and have yourself a good time. Just post pictures! .. and any funny stories about being investigated for buying so much mercury out of the blue.

As for the pump, you'd be better off going for a small, high pressure lift pump. A high flow pump isn't going to have noticeably higher output pressures. You'd get the best performance going for something on the industrial end of things, quick google hunt brings up something like these pumps here.

Using the smallest one of those pumps, you could easily get enough pressure to push mercury through your system, and still maintain a low, manageable flow rate, AND not have to worry about burning out your pump. But again, now we're talking about high quality, expensive as hell stuff. Plus, you'd probably need to have it custom fitted to handle mercury and all it's quirks. Also now you're WAY out of the flexible tubing ballpark. Now you're talking about metal pipe, probably stainless steel, or MAYBE some of the stronger plastic rigid pipes.

By the way, if anyone ever tries to make good on this whole idea.. please record all conversations you have when ordering products. Specifically, the mercury itself, and the pump. I want to hear the reaction the sales rep has when you explain that you'll need to know if any modifications need to be made to the pump before it handles a continuous flow of elemental mercury.

So to summarize: If you have buckets of money that you just don't really need, sure you could pull off a mercury cooling loop. It's still dangerous as hell, and you'll almost certainly have to register yourself with the local Hazmat authorities, in case of emergency. But it could be done. I would think that the inconvenience of having to keep your tower in a sealed environment (risk of mercury vapors, remember) would outweigh any possible coolness factor or cooling benefits, though.

 

[thaltek]

alright since i can see that this thread is going to get "omg! thats not a good idea"'d to death i will drop the matter.... if however someone inquirers me about the subject i will resurrect this thread.....

so at this point the conclusion of this thread is that its too toxic and too expensive...

 

[serpretetsky]

the only advantage i see to mercury is that you can build a completely passive pump for it with no moving parts, other than that, water all the way.

I am curious, though, to see mercury's performance compared to water, i always wondered about the heat conductivity vs heat capacity trade off. I'm definitly not curious enough to play with mercury.

 

[thaltek]

the only advantage i see to mercury is that you can build a completely passive pump for it with no moving parts, other than that, water all the way.

I am curious, though, to see mercury's performance compared to water, i always wondered about the heat conductivity vs heat capacity trade off. I'm definitly not curious enough to play with mercury.

i was under the impression that mercury didn't react with copper......
http://en.wikipedia.org/wiki/Amalgam_(chemistry)

i was wrong...... iron components all the way.... iron seems to be the only metal so far that would work for containing the toxic brew.....

 

[Yuritau]

I am curious, though, to see mercury's performance compared to water, i always wondered about the heat conductivity vs heat capacity trade off. I'm definitly not curious enough to play with mercury.

A substance with high heat conductivity and lower heat capacity (copper) is great for the block itself. But for coolant fluid, I believe that heat capacity is going to be the better property. I suppose you could design an effective system for either though. You'd just need higher flow rates (usually means higher pressure too) if the heat capacity was lower.

i was wrong...... iron components all the way.... iron seems to be the only metal so far that would work for containing the toxic brew.....

Ouch, having to use all iron is going to cut down on the CPU block's efficiency. Iron's thermal conductivity is only 80.4 W/mK (Wikipedia:iron), vs 401 W/mK for Copper (Wikipedia:copper).

 

[thaltek]

hmm maybe not... i will have to look into it but several options may exist such as using a diamond thermal interface with the processor....... or several other ideas..... i need to mull them over in my head before i start writing about them though.....

 

[Yuritau]

hmm maybe not... i will have to look into it but several options may exist such as using a diamond thermal interface with the processor....... or several other ideas..... i need to mull them over in my head before i start writing about them though.....

This just became my favorite conversation EVER. lol

 

[thaltek]

as some of us are aware diamond is one of the most heat conductive materials known to man...... and i am flattered my nerdyness made your day......

 

[stainremover]

i'm pretty sure it's one of those no can do types of things. i mean aside from the coolness factor, i don't think it's even going to do that much better than water. good wc loops keep the components a 10-15 degrees above ambient under load. so you switch to mercury, get industrial pumps, fabricate your own iron pipes and blocks - for what? keep your chips at ambient? 10 degrees lower under load?

and if everything's iron, you won't even be able to see the mercury. you'll just see plain iron pipes.

 

[jrdonnaruma]

You could potentially use high grade PEX tubing, like used in some lab environments. As for the diamond blocks, lab-grade synthetic diamond would likely be your best idea, as you could have them fabricated to the right specifications, and they would cost considerably less than cut diamonds.

I was going to suggest using Gallium for a safer choice, but then i remembered that it doesn't turn to liquid until the high 80s low 90s, not really ideal.

 

[BrainEater]

The liquid metal you are after is called Galinstan.

When it comes to liquid metal loops , you are not limited to just Iron.Any of the Refractory metals will work fine.

Galinstan would be a much better choice from a safety standpoint , with similar performance to mercury.

--------

Here is your diamond supplier :

http://www.usapplieddiamond.com/

A 1 inch diameter , 1mm thick , Thermally oriented lattice (1800W/(m-K)) diamond 'coldplate' is about 10k$...

 

[thaltek]

The liquid metal you are after is called Galinstan.

When it comes to liquid metal loops , you are not limited to just Iron.Any of the Refractory metals will work fine.

Galinstan would be a much better choice from a safety standpoint , with similar performance to mercury.

--------

Here is your diamond supplier :

http://www.usapplieddiamond.com/

A 1 inch diameter , 1mm thick , Thermally oriented lattice (1800W/(m-K)) diamond 'coldplate' is about 10k$...

perhaps i could get them to lower the price too about 5k$ if the disk is 0.25mm thick and 0.75in diameter.

as for visability of the mercury i wouldn't rule out the idea of using glass tubes wrapped in an clear anti shatter material.

 

[jrdonnaruma]

Man, that would be incredible to have as a cooling system *Starts planning his all diamond plate system for when he is filthy rich..* Even on just water!

Galinstan looks like a very promising substitute to mercury, if you dont want to have to worry about containment, just from my quick reading on MSDS info from Geratherm. Seems like a it might even out do mercury in cooling ability

 

[Mr. Pedantic]

Erm...yeah.

Bear in mind, Mercury has a viscosity about twice that of water. In addition, it also has a density about 28 times as great as that of water. These 2 statistics mean that even if you were to put your pipes perfectly horizontally, your pumps would have to be abnormally powerful. And since you probably wouldn't have your pipes completely horizontal, then your pumps would have to be even more powerful, to counteract the force a vertical column of mercury of practical dimensions would exert on a conventional pump.

Also, while Mercury has a marginally lower coefficient of expansion, since the specific heat capacity of mercury, being a metal, is much lower than that of water, you'll see a mercury system reaching much greater temperatures than a comparable system of water. Not only will this reduce the efficiency of the cooling system, since thermal conductivity depends on the temperature difference between the two materials (though it would still conduct much better than water), but because of the temperature range, it would also expand. This would lead to quite major problems with leakage and such, especially if the unit were designed for long-term use.

It would also mean that your contact plate would probably need to be something non-conducting, just in case.

 

[LittlePud]

Forget about the technical challenges; what about legal problems and liability?

- Having that much Mercury in one place probably violates some sort of local ordinance if not some federal (EPA) regulations.
- If you have a leak (whether vapor or actual metal) and a 3rd party gets injured, you will most definitely be looking at a civil suit if not criminal prosecution (negligence causing bodily harm?).

IMHO risking a trip to federal PMITA prison isn't worth a few extra MHz or GHz.

 

[thaltek]

Forget about the technical challenges; what about legal problems and liability?

- Having that much Mercury in one place probably violates some sort of local ordinance if not some federal (EPA) regulations.
- If you have a leak (whether vapor or actual metal) and a 3rd party gets injured, you will most definitely be looking at a civil suit if not criminal prosecution (negligence causing bodily harm?).

IMHO risking a trip to federal PMITA prison isn't worth a few extra MHz or GHz.

meh......

pedantic: such pumps do exist, please bear in mind this is a hypothetical thread...

littlepud: perhaps you should look up said regulations....... and let us know...... i would love to sit for a couple hours and research that but >_< classes are eating about 19 hours of each day and the rest is for sleep!!!!!

 

[LittlePud]

http://www.epa.gov/mercury/index.html
http://www.epa.gov/mercury/regs.htm
http://www.epa.gov/mercury/spills/index.htm

"Spills of More than Two Tablespoons (One Pound)
Any time one pound or more of mercury is released to the environment, it is mandatory to call the National Response Center (NRC). The NRC hotline operates 24 hours a day, 7 days a week. Call (800) 424-8802. Note that because mercury is heavy, only two tablespoons of mercury weigh about one pound."

2 tablespoons is 30 mL to a pound (~454 g). A liter of the stuff would weigh over 15 kg (33.3 lbs) -- this stuff is (IIRC) denser than lead. Toxicity aside I don't think it would be practical (even in a lab environment) to try and build a loop around a fluid this dense -- there's no way your piping/gaskets would hold up, and there's probably no reasonably sized/priced pump that can move the stuff.

 

[thaltek]

little pud if you don't mind i will refer to you as "pd"....

thank you for that important information pd......


pumps:
http://www.precisionfluidpower.com/c/Vickers/Vickers.html?gclid=COatmpyFkp0CFdFL5QodhDRA1A

lines and fittings:
http://www.discounthydraulichose.co...t=21&Click=2&gclid=COfmhLmFkp0CFeEN5QodXCP68Q

 

[osrk]

magneto pumps anyone? I haven't done any research but being mercury is a metal i think magneto pumps would be badass instead of you standard centrifugal pumps that everyone else uses. ENGAGE THE CATERPILLAR DRIVE!!!

I'm also +1 on the whole "really bad idea" thing and if i was your neighbor i'd probably file some sort of court injunction to stop you from doing this....by the way where do you live?

 

[jrdonnaruma]

so you want to use a magnet based pump.. Inside of a computer? Did you even think about what you just said before you posted? Thats just the dumbest idea I have ever heard.

 

[LittlePud]

little pud if you don't mind i will refer to you as "pd"....

thank you for that important information pd......


pumps:
http://www.precisionfluidpower.com/c/Vickers/Vickers.html?gclid=COatmpyFkp0CFdFL5QodhDRA1A

lines and fittings:
http://www.discounthydraulichose.co...t=21&Click=2&gclid=COfmhLmFkp0CFeEN5QodXCP68Q

The problem with using metal pumps/piping with Mercury is that even in the absence of any chemical reactivity, Mercury readily forms an amalgam (i.e. dissolves) numerous other metals. Think of it as trying to contain water with a pipe made of sugar or salt crystals.

 

[LittlePud]

magneto pumps anyone? I haven't done any research but being mercury is a metal i think magneto pumps would be badass instead of you standard centrifugal pumps that everyone else uses. ENGAGE THE CATERPILLAR DRIVE!!!

I'm also +1 on the whole "really bad idea" thing and if i was your neighbor i'd probably file some sort of court injunction to stop you from doing this....by the way where do you live?

Mercury is a metal but is not ferromagnetic. Magnetic caterpillar pump wouldn't work.

 

[thaltek]

magneto pumps anyone? I haven't done any research but being mercury is a metal i think magneto pumps would be badass instead of you standard centrifugal pumps that everyone else uses. ENGAGE THE CATERPILLAR DRIVE!!!

I'm also +1 on the whole "really bad idea" thing and if i was your neighbor i'd probably file some sort of court injunction to stop you from doing this....by the way where do you live?

next door...... that will remain my answer until orsk decides to read the first 5 posts.....

The problem with using metal pumps/piping with Mercury is that even in the absence of any chemical reactivity, Mercury readily forms an amalgam (i.e. dissolves) numerous other metals. Think of it as trying to contain water with a pipe made of sugar or salt crystals.
Today 02:01 PM

this was already discussed.... iron and most high iron content alloys are nonreactive to mercury.... i was thinking of using a hydraulic pump to compensate for the density seeing as how a centrifugal pump would just spin in place and not pump effectively..... as for a magneto pump i needs moar pics and info!

 

[BrainEater]

so you want to use a magnet based pump.. Inside of a computer? Did you even think about what you just said before you posted? Thats just the dumbest idea I have ever heard.

erm , never taken apart a harddrive I see.Hardrives have VERY strong magnets in them....Magnets from a pump (of any kind) are not an issue.


-------

Mercury is a metal but is not ferromagnetic. Magnetic caterpillar pump wouldn't work.

Wrong.

EM pumps don't require ferromagnetism , it's conducting fluid that matters.EM pumps work just fine for mercury/galinstan etc.

-----

I'll see if I can dig up some pics/info on EM pumps.

Try googling 'nanocoolers'. Thats a company that used to sell EM pump based liquid metal (galinstan) cooling.

 

[jrdonnaruma]

erm , never taken apart a harddrive I see.Hardrives have VERY strong magnets in them....Magnets from a pump (of any kind) are not an issue.

Apparently you have never put another magnet near a hard drive... Having a pump that works through magnetism is not something I would want anywhere NEAR my hard drives. Not even in the same case. If you were going to do it with a magnetic pump, i sure hope you are using all solid state drives within the case.

 

[ann0yanc3]

wonder how well a TRU Diamond HSF would work... would be pretty baller to say that you cool your computer with diamonds.

 

[Zoson]

Apparently you have never put another magnet near a hard drive... Having a pump that works through magnetism is not something I would want anywhere NEAR my hard drives. Not even in the same case. If you were going to do it with a magnetic pump, i sure hope you are using all solid state drives within the case.

Motors are magnetic. Due to that fact, unless you're running a combustion engine to run your pump, all pumps are magnetic.
My MCP655 is right next to all five of my disks, it's been there for over a year without issue.
The idea that your pump will mess up your disks at all is a complete fallacy. You should really know what you're talking about before you call someone stupid.

 

[BillParrish]

OK apparently no one here knows how magnetic fields and Faraday cages work. Stop it.

The answer is a strong focused magnetic field will screw up a HD in a second.

The answer is motor magnetic fields are focused and the field strength falls off rapidly even if any of it escapes the motor housing the HD case will shield sufficiently from any leakage. Even an inch of space and or any other grounded ferrous metal will knock the field strength down significantly.

Take a 5 lb horseshoe magnet and set it directly on top of an HD and its turned into a raw drive instantly and you will be lucky if it low level formats properly again.


Dont make me have to stop this car. Carry on.
About the Mercury. You will shoot your eye out.

 

[BrainEater]

Thx Zoson , for saving me half the rant.

Apparently you have never put another magnet near a hard drive...

Yes in fact I do.All the time.So does anyone with more than one harddrive.I have 6 hdd's in a stack.........they don't erase each other.

The magnets inside harddrives are Neodymium Iron Boron (NIB) magnets.They are many,many times more powerful than the magnets inside any pump,including MHD pumps.....And even these don't affect the magnetic data.....There's shielding to consider as well..


-----

Take apart a harddrive !

The NIB magnets are insanely powerful....fun stuff !

Be careful tho , they are quite brittle , and shatter like glass , becoming very sharp.

----------

The primer on how MHD (em) pumps operate : http://en.wikipedia.org/wiki/Magnetohydrodynamics

Old Techpowerup article on Nanocooler stuff :

http://www.techpowerup.com/?3105

------

Bottom line :

Yes a mercury/galinstan liquid metal cooling system is not only possible , but people have tried to make money from it already.

The technological challenges are not easily overcome however .It really is just a cost thing....to make it operate reliably and safely costs many thousands of dollars.

 

[BrainEater]

OK apparently no one here knows how magnetic fields and Faraday cages work. Stop it.

Take a 5 lb horseshoe magnet and set it directly on top of an HD and its turned into a raw drive instantly and you will be lucky if it low level formats properly again..

Eh...

Jam on the brakes all ya want.

I did'nt want to get into the entirety of magnetic field theory.Please feel free to teach the lesson , I could use the help.

I've run several harddrives on top of large speakers.And next to pumps, etc....

There just isnt the field strength to affect the data....a 5 pound horseshoe magnet , well maybey , but the harddrive enclosure itself is it's own shield...

Ever actually try that ?

I've got some HDD's lying around , and well , really good magnets too...I admit , Mr Parrish , you've got me curious just what it takes.

 

[thaltek]

Thx Zoson , for saving me half the rant.

Yes in fact I do.All the time.So does anyone with more than one harddrive.I have 6 hdd's in a stack.........they don't erase each other.

The magnets inside harddrives are Neodymium Iron Boron (NIB) magnets.They are many,many times more powerful than the magnets inside any pump,including MHD pumps.....And even these don't affect the magnetic data.....There's shielding to consider as well..


-----

Take apart a harddrive !

The NIB magnets are insanely powerful....fun stuff !

Be careful tho , they are quite brittle , and shatter like glass , becoming very sharp.

------

Bottom line :

Yes a mercury/galinstan liquid metal cooling system is not only possible , but people have tried to make money from it already.

The technological challenges are not easily overcome however .It really is just a cost thing....to make it operate reliably and safely costs many thousands of dollars.

you are my new freind!!!!!!!

and i have a small collection of hd magnets.... they are fun to play with....

 

[Zoson]

OK apparently no one here knows how magnetic fields and Faraday cages work. Stop it.

The answer is a strong focused magnetic field will screw up a HD in a second.

The answer is motor magnetic fields are focused and the field strength falls off rapidly even if any of it escapes the motor housing the HD case will shield sufficiently from any leakage. Even an inch of space and or any other grounded ferrous metal will knock the field strength down significantly.

Take a 5 lb horseshoe magnet and set it directly on top of an HD and its turned into a raw drive instantly and you will be lucky if it low level formats properly again.


Dont make me have to stop this car. Carry on.
About the Mercury. You will shoot your eye out.

Bill, notice I focused ONLY on motors... Not on actual E&M. "The idea that your pump will mess up your disks at all is a complete fallacy." I didn't say that magnetic fields couldn't mess up your disks.

On another note, do HDD's have faraday cages in them? I don't see one through the window of my WD RaptorX. Or is the steel casing just enough to act as one?