Silent liquid cooled computer...

6 views
Skip to first unread message

Andrew

unread,
Sep 14, 2011, 6:24:57 AM9/14/11
to Robots & Dinosaurs

Luke Emrose

unread,
Sep 14, 2011, 7:43:48 PM9/14/11
to sydney-h...@googlegroups.com
Hmmm, as an "audio engineer" I am definitely looking for something like this, but there are clearly many problems.

I would imagine that the higher viscosity of the oil in comparison to air would have a direct impact on the life of the electric engines, and therefore significantly decrease the life of the entire thing.
However I now know that there are non-conductive oils, something I definitely was not aware of before!

cheers for the link, fascinating idea and implementation.

L

On 14 September 2011 20:24, Andrew <and...@arcadius.com.au> wrote:
http://www.youtube.com/watch?v=6sP45uBj4-k&feature=related

--
You received this message because you are subscribed to the Google Groups "Robots & Dinosaurs" group.
To post to this group, send email to sydney-h...@googlegroups.com.
To unsubscribe from this group, send email to sydney-hackspa...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/sydney-hackspace?hl=en.




--
Luke Emrose
aka evolutionary theory
www.evolutionarytheory.com
www.soundcloud.com/evolutionarytheory
www.reverbnation.com/evolutionarytheory
http://www.facebook.com/pages/evolutionary-theory/110717958952413

Lachlan Horne

unread,
Sep 14, 2011, 7:48:50 PM9/14/11
to sydney-h...@googlegroups.com
A lot of oils are surprisingly non-conductive...

tALSit de CoD

unread,
Sep 14, 2011, 7:52:22 PM9/14/11
to sydney-h...@googlegroups.com
What you really want is Fluorinert which is what they used to cool down the Cray-2 Supercomputer. I was interested in trying this out a few years ago, and I contacted 3M, and after about 2-4 months of calls, I was quoted about $400/gallon, and being a uni student, I dropped my interest in it almost instantly. It did, however, include specialised shipping from New Zealand.

David Lyon

unread,
Sep 14, 2011, 9:11:20 PM9/14/11
to sydney-h...@googlegroups.com

On Thu, Sep 15, 2011 at 9:52 AM, tALSit de CoD <tal...@talsit.org> wrote:
> What you really want is Fluorinert which is what they used to cool down
> the Cray-2 Supercomputer. I was interested in trying this out a few years
> ago, and I contacted 3M, and after about 2-4 months of calls, I was quoted
> about $400/gallon, and being a uni student, I dropped my interest in it almost
> instantly. It did, however, include specialised shipping from New Zealand.

If you really have too much heat, think about doing some energy recovery
using TECs/Peltier ie:

 - http://www.aliexpress.com/wholesale?SearchText=tec

If you want to do serious cooling, which I've never tried but read about in
turn of the 20th century books, they suggest making liquid air. Once you
compress air enough it enough it turns back to liquid it has an enormous
capacity to suck in energy.

Apart from being quite volitile, as in explosive, it's not carcinogenic.

I just noticed that NASA is going to be using liquid oxygen and liquid
hydrogen in their next rocket program for going to Mars. If they are
allowed to play with it, in Australia we should too :-)



Lemming

unread,
Sep 14, 2011, 9:24:21 PM9/14/11
to Robots & Dinosaurs
http://forums.bit-tech.net/showthread.php?t=163649 <--- good build log
of doing a mineral oil cooled one


> If you want to do serious cooling, which I've never tried but read about in
> turn of the 20th century books, they suggest making liquid air. Once you
> compress air enough it enough it turns back to liquid it has an enormous
> capacity to suck in energy.

Making liquid air is not a simple process, also having enough to
immerse your whole motherboard in would be problematic, not to mention
a lot of components STOP working in temperatures that cold.

> Apart from being quite volitile, as in explosive, it's not carcinogenic.

Nothing I have read substantiates this.

> I just noticed that NASA is going to be using liquid oxygen and liquid
> hydrogen in their next rocket program for going to Mars. If they are
> allowed to play with it, in Australia we should too :-)

We are, afaik there are no restrictions on buying either. The problem
is having the money.

> I would imagine that the higher viscosity of the oil in comparison to air
> would have a direct impact on the life of the electric engines, and
> therefore significantly decrease the life of the entire thing.

The higher viscosity only really affects the fans, and they don't seem
to care that much.

What does happen as mentioned in the build log I linked is rubber and
plastic parts can be softened. So traditional electrolytic caps tend
to fail very quickly.

Ian Howson

unread,
Sep 14, 2011, 9:24:40 PM9/14/11
to sydney-h...@googlegroups.com
One doth not play idly with such materials:

Note that is on a near-empty tank!





--

tALSit de CoD

unread,
Sep 14, 2011, 9:29:43 PM9/14/11
to sydney-h...@googlegroups.com
Those guys are awesome!
I love how they share all (most?) of their materials.
After this test, you can see that all their tethered test start off on stilts and then "land" by hanging off the tether.

David Lyon

unread,
Sep 14, 2011, 9:33:07 PM9/14/11
to sydney-h...@googlegroups.com
is that a (beautiful) pink flame?

Angus Gratton

unread,
Sep 14, 2011, 9:54:41 PM9/14/11
to sydney-h...@googlegroups.com
On Wed, 2011-09-14 at 18:24 -0700, Lemming wrote:
> > Apart from being quite volitile, as in explosive, it's not carcinogenic.
>
> Nothing I have read substantiates this.

Sorry if I'm misreading you here Lemming, but are you saying nothing
substantiates that liquid air would be volatile?

Liquid air would be ~20% liquid oxygen. Nitrogen has a lower boiling
point than oxygen, so if/when you release liquid air that concentration
rapidly goes up.

Even pure liquid nitrogen (inert by itself, obviously) can lead to
volatility issues if cold enough, due to condensing oxygen out of the
air.

Fun times...


Anyhow, don't know if this has been posted yet, but AMD have been
sponsoring ridiculous overclocking of their CPUs by cooling with liquid
nitrogen & (bonus ridiculousness) liquid helium.
http://www.youtube.com/watch?v=UKN4VMOenNM

(Caution: video makes overclocking look 'edgy' with excess red light &
rock music. ;).)

Thanks for Lachlan for originally pointing me to that video. :)


- Angus


David Lyon

unread,
Sep 14, 2011, 10:09:49 PM9/14/11
to sydney-h...@googlegroups.com

On Thu, Sep 15, 2011 at 11:54 AM, Angus Gratton <g...@projectgus.com> wrote:
Sorry if I'm misreading you here Lemming, but are you saying nothing
substantiates that liquid air would be volatile?

try: http://en.wikipedia.org/wiki/Thermobaric_weapon

David Lyon

unread,
Sep 15, 2011, 9:58:56 AM9/15/11
to sydney-h...@googlegroups.com

On Thu, Sep 15, 2011 at 11:24 AM, Lemming <iner...@gmail.com> wrote:
We are, afaik there are no restrictions on buying either. The problem
is having the money.

Well all I can say is that everyday the media bombards us with political
statements saying that Australia is to be moving into 'clean energy future'
(quote - Julia Gillard).

The American space program, with it's Mars rocket, and supercooled
oxygen and hydrogen propulsion system, definitely show what 'clean
energy' projects are underway in the world - and what our politicians
have already agreed (in-principle) to buy.

At the end of the day, this Mars rocket is just water powered. Having
it's basic water molecules pre-processed and then stored. Mixed at
the last minute to release massive amounts of energy as the bonds
come back together to form a stable molecular state.

True - there's no restrictions on buying the technology. It will all be
available on the international market just like solar cells and wind-
turbines. Water Powered rockets and propulsion systems are going
to become billion dollar systems - once the technology is better
development and the kinks worked out.

We'll always be allowed to buy it.. for that there is no doubt.

That's why we're being asked to pay all this money from the
carbon tax.. so that we can buy a new stream of technology that
is going to come out of the current NASA space propulsion
systems..








Jake Anderson

unread,
Sep 15, 2011, 10:12:58 AM9/15/11
to sydney-h...@googlegroups.com
you do realise that pretty much all production hydrogen comes from oil/gas/coal right?
nobody makes hydrogen from water on an industrial scale.

Lemming

unread,
Sep 15, 2011, 4:41:55 PM9/15/11
to Robots & Dinosaurs
Liquid air in and of itself is not explosive.

Adding something else to it will make it potentially explosive.

Liquid air cannot ignite, it can be used to oxidize things that are
already alight, but chances are the large amounts of nitrogen present
would preclude a detonation.

And yes I am well aware of Thermobaric weapons. Thermo being heat,
which liquid air has a distinct lack of for a start.

Also the most common Thermobaric weapon as mentioned on the first line
of that wikipedia article is a "Fuel-Air Bomb" or FAE. As the "fuel"
in the name stipulates, you have to add fuel to it. Liquid air !=
fuel.

On Sep 15, 12:09 pm, David Lyon <david.lyon.preissh...@gmail.com>
wrote:

Lemming

unread,
Sep 15, 2011, 5:05:26 PM9/15/11
to Robots & Dinosaurs
> At the end of the day, this Mars rocket is just water powered. Having
> it's basic water molecules pre-processed and then stored. Mixed at
> the last minute to release massive amounts of energy as the bonds
> come back together to form a stable molecular state.

Hydrogen and Oxygen are stable molecules in their own right. Mixing
liquid Hydrogen and liquid Oxygen together, even in the right (2:1)
proportions will not suddenly transform it into water. You would end
up with a mixture of the two liquids, and that is all.

To turn them into water the hydrogen must be combusted in the presence
of oxygen, this will release "massive amounts of energy" and also
leave you with some water left over. the input to output ratio is not
1:1 though, far from it.

For a start, the amount of energy required to get the the hydrogen in
the first place is a lot more so than what you get out of it.
Especially so if you are trying to turn water into Hydrogen and
Oxygen. Then to get water, and only water, out of the reaction you
have to burn it in the presence of pure oxygen (more energy used to
extract and liquefy the oxygen). If you burn hydrogen with air (or
liquid air) a lot of the hydrogen will actually react with the
nitrogen present in the air to form some fairly toxic compounds. The
most well known of which being Ammonia (NH³).

Hydrogen is not a viable fuel for our worlds economy, not until we
come up with sustainable, controllable, and most importantly clean
nuclear fusion.

David Lyon

unread,
Sep 15, 2011, 7:15:46 PM9/15/11
to sydney-h...@googlegroups.com
The fuel for Thermobaric weapons is just liquid air. Nothing else.

The wiki page is worded in such a way that it is reasonably truthful
but doesn't give people the idea that they could make such a device
themselves. They wouldn't want people going onto hackaday.com
and pulling this:

 - http://hackaday.com/2010/06/10/making-liquid-nitrogen-at-home/

removing the filter and voila.. but they could..

As weapons, those devices are best left lost in history.

But in propulsion and energy systems, completely different story.

What would be nice to see is the Hunter Valley turned into the
Australian Las Vegas. I mean where they generate Hydroxy and
hydrogen gas from Solar Cells. And you can buy it if you don't
make it yourself.

As for what is and what isn't a 'viable-fuel', that's a completely
subjective term.

And I don't know what country you work in, but as an Australian
I am not allowed to work on any nuclear fusion project in this
country. It's forbidden by Law.

Sadly Lemming - "we" won't be coming up with anything. What
you talk about is just not allowed to be worked on here.




Jake Anderson

unread,
Sep 15, 2011, 7:41:35 PM9/15/11
to sydney-h...@googlegroups.com, David Lyon
On 09/16/2011 09:15 AM, David Lyon wrote:
The fuel for Thermobaric weapons is just liquid air. Nothing else.
[citation needed]

http://www.globalsecurity.org/military/systems/munitions/thermobaric.htm
you are *wrong* there is no conspiracy, the simple answer is in order to have a chemical reaction you must have chemicals that want to react.
air is air is air, gas or liquid won't suddenly make it explode otherwise the atmosphere of the planet would explode on a regular basis.
I have played with liquid air, LN2 and LOX
none of them exploded even a little bit.




The wiki page is worded in such a way that it is reasonably truthful
but doesn't give people the idea that they could make such a device
themselves. They wouldn't want people going onto hackaday.com
and pulling this:

 - http://hackaday.com/2010/06/10/making-liquid-nitrogen-at-home/

removing the filter and voila.. but they could..

As weapons, those devices are best left lost in history.
Yes best people use nukes to blow stuff up.



And I don't know what country you work in, but as an Australian
I am not allowed to work on any nuclear fusion project in this
country. It's forbidden by Law.
[citation needed]

As somebody who is interested in this field, there are no laws precluding it.
There are laws regarding the creation and disposal of radioactive materials which is fair enough.



Sadly Lemming - "we" won't be coming up with anything. What
you talk about is just not allowed to be worked on here.
sure, that's the reason, its a big government conspiracy with thousands of men in black enforcing it thats the most obvious answer.

Nick Johnson

unread,
Sep 15, 2011, 8:25:33 PM9/15/11
to sydney-h...@googlegroups.com
On Fri, Sep 16, 2011 at 7:05 AM, Lemming <iner...@gmail.com> wrote:
> At the end of the day, this Mars rocket is just water powered. Having
> it's basic water molecules pre-processed and then stored. Mixed at
> the last minute to release massive amounts of energy as the bonds
> come back together to form a stable molecular state.

Hydrogen and Oxygen are stable molecules in their own right. Mixing
liquid Hydrogen and liquid Oxygen together, even in the right (2:1)
proportions will not suddenly transform it into water. You would end
up with a mixture of the two liquids, and that is all.

To turn them into water the hydrogen must be combusted in the presence
of oxygen, this will release "massive amounts of energy" and also
leave you with some water left over. the input to output ratio is not
1:1 though, far from it.

You get out exactly as much water as you put in.
 

For a start, the amount of energy required to get the the hydrogen in
the first place is a lot more so than what you get out of it.

This, however, is true, thanks to thermodynamics.
 
Especially so if you are trying to turn water into Hydrogen and
Oxygen. Then to get water, and only water, out of the reaction you
have to burn it in the presence of pure oxygen (more energy used to
extract and liquefy the oxygen).

'Extracting' the oxygen takes place at the same time as you 'extract' the hydrogen. Electrolyze water and you get hydrogen and oxygen, in exactly the ratio needed to combust it. As you point out, it takes more energy than you can produce by burning it, of course.

 
If you burn hydrogen with air (or
liquid air) a lot of the hydrogen will actually react with the
nitrogen present in the air to form some fairly toxic compounds. The
most well known of which being Ammonia (NH³).

Hydrogen is not a viable fuel for our worlds economy, not until we
come up with sustainable, controllable, and most importantly clean
nuclear fusion.

Lemming

unread,
Sep 16, 2011, 8:31:15 PM9/16/11
to Robots & Dinosaurs
> 'Extracting' the oxygen takes place at the same time as you 'extract' the
> hydrogen. Electrolyze water and you get hydrogen and oxygen, in exactly the
> ratio needed to combust it. As you point out, it takes more energy than you
> can produce by burning it, of course.

Extracting it is one part, compressing it and cooling it to liquefy it
is another.

David Lyon

unread,
Sep 16, 2011, 11:32:21 PM9/16/11
to sydney-h...@googlegroups.com

On Sat, Sep 17, 2011 at 10:31 AM, Lemming <iner...@gmail.com> wrote:

Extracting it is one part, compressing it and cooling it to liquefy it
is another.


Well, most of these designs have some sort of commercial plan that
underpins their development.

The seperation of the gas increases safety dramatically. Oxygen and
Hydrogen tanks are sortof safish. If kept apart.

The main problem with hydrogen as a propellent is the low energy
density. But if you have a tank of oxygen and a tank of hydrogen
then the energy density issue completely changes (towards being
practical). A tank of Nitrogen (NOS) - lol - also boosts power..

Perhaps in the future, they may be thinking that they can sell
precompressed oxygen and hydrogen in tanks in a similar way
to what LPG/LNG is sold now.

It would certainly be an easy thing for big business to do. And very
easy to bring about the damage to our coal industry that Julia is
already telling us will happen..

many many hurdles.. but they're working on them...







Reply all
Reply to author
Forward
0 new messages