non-conductive fluid

The alternative to direct air cooling

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
nahyah
Posts: 30
Joined: Tue Aug 03, 2004 12:17 pm

non-conductive fluid

Post by nahyah » Wed Nov 17, 2004 3:34 pm

how much would you pay for non conductive, non corrosive, basically fluid designed for watercooling?

TheDarkHacker
Posts: 135
Joined: Tue Jun 01, 2004 10:09 am
Location: Fort Worth, Texas

Post by TheDarkHacker » Wed Nov 17, 2004 3:48 pm

personally or for real. for real it will cost you about $200 a gallon. this is what i would recommend but why not use water

http://www.dtekcustoms.com/index.asp?Pa ... ProdID=139

nahyah
Posts: 30
Joined: Tue Aug 03, 2004 12:17 pm

Post by nahyah » Wed Nov 17, 2004 4:22 pm

yeah thats what i saw and i thought it was a typo, but i guess not.. do they know that people not many people will drop that much money on just fluids?

Straker
Posts: 657
Joined: Fri Jul 23, 2004 11:10 pm
Location: AB, Canada
Contact:

Post by Straker » Wed Nov 17, 2004 4:56 pm

...they do, or the product wouldn't exist. lots of people waste money on kits too.

Rusty075
SPCR Reviewer
Posts: 4000
Joined: Sun Aug 11, 2002 3:26 pm
Location: Phoenix, AZ
Contact:

Post by Rusty075 » Wed Nov 17, 2004 5:56 pm

Why do you care if your fluid is non-conductive? (Conductivity is not the problem with using just plain water in a WC loop)

TheDarkHacker
Posts: 135
Joined: Tue Jun 01, 2004 10:09 am
Location: Fort Worth, Texas

Post by TheDarkHacker » Wed Nov 17, 2004 6:05 pm

well if it is non conductive, if you have a leak, it will do nothing to the system. you could dump this stuff all over your computer compoents and it will still run.

Rusty075
SPCR Reviewer
Posts: 4000
Joined: Sun Aug 11, 2002 3:26 pm
Location: Phoenix, AZ
Contact:

Post by Rusty075 » Wed Nov 17, 2004 6:22 pm

Yes, your machine will continue to run until you've overheated and toasted the CPU. :lol:

There are a variety of non-conducting fluids already available, but they're all either way to expensive (like Fluorinert), or have horrible thermal properties (like Mineral Oil)

Bat
Posts: 279
Joined: Sun Jan 05, 2003 12:09 pm
Location: U.K.

Post by Bat » Thu Nov 18, 2004 1:47 pm

I'm very sceptical about that "Fluid XP" stuff, especially since finding some details (on the manufacturer's site, I think it was) and learning that it's mostly water.

Also see here:
http://www.overclockers.com/articles1028/
If the list of ingredients there is complete, then you shouldn't expect the stuff to inhibit corrosion.

nahyah
Posts: 30
Joined: Tue Aug 03, 2004 12:17 pm

Post by nahyah » Fri Nov 19, 2004 3:19 pm

i think i will trust my engineering ability and use water. wonder how much it will cost to use fluid xp on reserator....

scotty6435
Posts: 100
Joined: Wed Nov 03, 2004 3:02 pm
Location: Salisbury, UK

Post by scotty6435 » Sat Nov 20, 2004 11:06 am

Yeah, fluidXP is basically pre-mixed (and overpriced) distilled water with an anti-corrosive.

Bat
Posts: 279
Joined: Sun Jan 05, 2003 12:09 pm
Location: U.K.

Post by Bat » Mon Nov 22, 2004 10:54 am

With an "anticorrosive"? Not according to the link I gave above:
http://www.overclockers.com/articles1028/
According to the Material Safety Data Sheet (MSDS) given to me by Integrity PC, FLuid XP+ contains

* Dihydrogen Oxide, [Water]
* Xanthan Gum, [CP Kelco: Keltrol-T &/or Keltrol-T622]
* 1-Dodecanol [Lauryl Alcohol]
* 1,2-Propanediol [Propylene Glycol]
* 1,2,3-Propanetriol, [Glycerin]
To be honest I'm a little puzzled: so far as I can see the main effect of those additives would be to make the stuff a little gloopier (more viscous). Perhaps it's just a marketing things: one reviewer guessed it was some sort of light oil, based on the "non conductive" sales pitch and the way it poured and felt on his fingers.

scotty6435
Posts: 100
Joined: Wed Nov 03, 2004 3:02 pm
Location: Salisbury, UK

Post by scotty6435 » Mon Nov 22, 2004 11:05 am

Bat wrote:With an "anticorrosive"? Not according to the link I gave above:
http://www.overclockers.com/articles1028/
According to the Material Safety Data Sheet (MSDS) given to me by Integrity PC, FLuid XP+ contains

* Dihydrogen Oxide, [Water]
* Xanthan Gum, [CP Kelco: Keltrol-T &/or Keltrol-T622]
* 1-Dodecanol [Lauryl Alcohol]
* 1,2-Propanediol [Propylene Glycol]
* 1,2,3-Propanetriol, [Glycerin]
quote]

It's designed to make it less able to support metal ions in solution (galvionic corrosion). C'mon, a site that lists water as dihydrogen oxide is gotta be a n00b fisher. The alcohol (layryl means it is clear) and glycol is antifreeze and the glycerin lubricates the pump (not sure how good that is in the long run). Xantham gum is a stabiliser for the rest of the ingredients. It basically stops them separating out onto your pipes, blocks and pump when the system is powered down.

What you'll find is that after a month or two it will have to be changed as it breaks down (common prob with antifreeze). Waterwetter or equivalent + distilled water is still by far the best method.

Bat
Posts: 279
Joined: Sun Jan 05, 2003 12:09 pm
Location: U.K.

Post by Bat » Mon Nov 22, 2004 2:52 pm

scotty6435 wrote:It's designed to make it less able to support metal ions in solution
If the alcohols were there as a major proportion of the mixture, that effect might start to become significant. However, if metal salts are less soluble in the solution than in water, that just means the corrosion products will end up as solid rather than in solution sooner... except that the xanthan gum stabilises colloidal suspensions. I don't see it having a big effect one way or the other, but I'm not sure what they were thinking.
scotty6435 wrote:(galvionic corrosion).
The question of whether the corrosion is accelerated by mixed-metal "galvanic" effects is irrelevant here. ("Galvanic" not "Galvionic".)
scotty6435 wrote:C'mon, a site that lists water as dihydrogen oxide is gotta be a n00b fisher.
As I read it, the manufacturers used that term in the data sheet they supplied to the people running the overclockers.com site. Presumably they were hoping to impress someone: we agree there.
scotty6435 wrote:The alcohol (layryl means it is clear) and glycol is antifreeze
First, the word. The etymology of "lauryl" (not "layryl") is from laurel; my guess is that lauryl alcohol or lauric acid was originally extracted from that plant. Why do you say it means it's clear? Anyway, nowadays "lauryl" just refers to a chain of twelve carbon atoms.
1-Dodecanol is an oily liquid, with a melting point of 24-27C, and it's insoluble in (immiscible with) with water. (Presumably the gum and/or glycerin helps to stop it from separating out and floating to the top.) Being insoluble, it won't affect the melting/freezing point.
The propylene glycol will lower the freezing point: its own melting point is -59C and it is miscible with water.
The lowering of the freezing point will be small given the small concentration of additives. Anyway, does it matter at all? How many people want to chill their water below its normal freezing point?
scotty6435 wrote:and the glycerin lubricates the pump (not sure how good that is in the long run).
The mixture might be a little more effective as a lubricant than plain water (but remember that increased viscosity does not mean a better lubricant). I'd imagine that if there were any benefit the makers would be shouting about it though.
Oh, and (I'd have to check but) I'd expect it to lower the freezing point unless there's an awful lot of it: it's miscible with water and its melting point is 17.8C. Also it helps to stop the 1-dodecanol from separating out.
scotty6435 wrote:Xantham gum is a stabiliser for the rest of the ingredients. It basically stops them separating out onto your pipes, blocks and pump when the system is powered down.
That's only needed for the lauryl alcohol. It also increases the viscosity, has some lubricating ability and stabilises suspensions.
scotty6435 wrote:What you'll find is that after a month or two it will have to be changed as it breaks down
It wouldn't surprise me, but what makes you think it does? (The makers quote a shelf life of five years but say nothing about life in use.)
scotty6435 wrote:(common prob with antifreeze).
Really? It lasts for years in cars. What makes you think so?
scotty6435 wrote:Waterwetter or equivalent + distilled water is still by far the best method.
I'm not sure that water wetter helps. Certainly some kinds are reported as causing problems. I don't think I've seen any tests showing improved performance.

For inhibiting corrosion, I would recommend "life extender" additives for car engine coolant, or additives intended for domestic central heating systems (the kind with hot water circulating through radiators).

Note that the makers/sellers of this XP stuff say that the liquid itself is not corrosive but they don't say it inhibits or protects against corrosion. Also they don't mention growths of algae, bacteria or fungi, but they do say the additives are all food-grade so presumably living things will do just fine in it. (No, they're not used as preservatives in food.)

scotty6435
Posts: 100
Joined: Wed Nov 03, 2004 3:02 pm
Location: Salisbury, UK

Post by scotty6435 » Mon Nov 22, 2004 3:00 pm

Jesus! You produce this stuff or are you just a freelance chemist browsing the forums? Basically, the point of the post was to say that it really isn't a wonder product and a similar, or better efffect can be made much cheaper by adding a concentrate to distilled water. (took me 5 mins to write my post, how long to write yours... :wink: ) And it's a bit picky corecting a spelling mistake....

p.s. achieved an E grade in A-level chemistry and proud of it....it's rubbeshhh (watch little britain if you're confused :D )

Edward Ng
SPCR Reviewer
Posts: 2696
Joined: Thu Dec 11, 2003 9:53 pm
Location: Scarsdale, NY
Contact:

Post by Edward Ng » Mon Nov 22, 2004 4:09 pm

Bat wrote:...Also they don't mention growths of algae, bacteria or fungi, but they do say the additives are all food-grade so presumably living things will do just fine in it. (No, they're not used as preservatives in food.)
I got growths in my very first water system, which was filled with 100% Fluid XP. I don't know if that was the fault of the coolant or my fault (most likely mine), but I got growth within the first week, and it didn't seem to me that the Fluid XP was doing anything at all to prohibit it. The growth was a feathery, white looking streamy material that clung to my tube walls. Multiple hydrogen peroxide flushes of the system turned the feathery white streamy gunk into gross-as-all-hell orange particulates. My cousin, who was helping me with the flushing process, was about to puke from seeing it.

It didn't smell too great, either.

-Ed

Ice Czar
Posts: 20
Joined: Tue Nov 16, 2004 9:22 pm
Location: [H]igh Altitude Lab

Post by Ice Czar » Fri Nov 26, 2004 8:11 pm

Rusty075 wrote:Why do you care if your fluid is non-conductive? (Conductivity is not the problem with using just plain water in a WC loop)
direct contact like in Spray Cooling
basically a direct die phase change
(or within a block but then it wouldnt need to be non-conductive)
that fluid however wouldnt be a good choice with a boiling point of 99C
so just imersion for that fluid

the Heat Transfer Coefficient (W/cm² °C) of Spray cooling is far superior to other methods
Image

http://www.spraycool.com/flash/WP_Datas ... entals.pdf

Rusty075
SPCR Reviewer
Posts: 4000
Joined: Sun Aug 11, 2002 3:26 pm
Location: Phoenix, AZ
Contact:

Post by Rusty075 » Fri Nov 26, 2004 8:45 pm

Ice Czar wrote:direct contact like in Spray Cooling....
:lol:

You don't need non-conducting fluid for direct die cooling. Plain old Di+antifreeze will work just fine.

Besides, what percentage of WC'ers are doing that? I bet you could gather all the PC direct die cooling users in the world into one room. And it wouldn't be a big room. :wink:

Ice Czar
Posts: 20
Joined: Tue Nov 16, 2004 9:22 pm
Location: [H]igh Altitude Lab

Post by Ice Czar » Sat Nov 27, 2004 9:27 am

oh I dont know
think the application is slight larger than you might imagine
starting with the US Army :p


spray cooling is used for thermal control on alot of "ruggedized" equipment, and is making inroad to high density enterprise computing, of course in the high density application direct die isnt necessary or even perferable

but "spray cooling" itself is one of those thermal technologies that is just going to become more prevelant, sort of like heatpipe technology is now, about four years ago there where no consumer level products employing heatpipes (but they where readilly available to enterprise)
now they are everywhere

its actually a super simple technology with just a micropump to go wrong
but like most phase change its all about getting the appropriate balance
putting it outside of the DIY category currently
its been around for a long time as well, but outside of supercomputers wasnt adopted for quite awhile

while a nonconductive fluid isnt necessary for direct die, it is perferable
(ask SurlyJoe at Spodes) and for typical enterprise immersion or direct component cooling is universally employed, generally Perfluorocarbon (3M's Fluorinert)

OC enthusiasts are always "rediscovering" applications\issues that thermal enginneers are familiar with, from Bong (Evaporative) Cooling, Heatpipes, Immersion, TEC, jet impingement, pump heat, indirect phase change, and cascade phase change to direct die
Last edited by Ice Czar on Sat Nov 27, 2004 11:32 am, edited 4 times in total.

Tyrdium
Posts: 272
Joined: Sat Jun 12, 2004 7:29 am
Location: Boston, MA
Contact:

Post by Tyrdium » Sat Nov 27, 2004 9:32 am

IIRC, spray cooling was used (dunno if it still is) in a number of supercomputers (I know for a fact that Cray used it). I believe they used Fluorinert for it; I have no idea how much it costs, though.

Ice Czar
Posts: 20
Joined: Tue Nov 16, 2004 9:22 pm
Location: [H]igh Altitude Lab

Post by Ice Czar » Sat Nov 27, 2004 9:42 am

Its ungodly expensive :p

imagine just the cost of canned pumps and welded stainless stell tubing
not to mention the Fluorinert itself

seals need to be at 4 atmospheres of water pressure IIRC

Tyrdium
Posts: 272
Joined: Sat Jun 12, 2004 7:29 am
Location: Boston, MA
Contact:

Post by Tyrdium » Sat Nov 27, 2004 9:48 am

Ouch, point. Ah well, worth a try... :P

Hmmm... Oy, anybody have access to cheap liquid nitrogen (purity probably doesn't matter too much)? I'm wondering if it would be possible to run a computer fully immersed in the stuff. :twisted:

Ice Czar
Posts: 20
Joined: Tue Nov 16, 2004 9:22 pm
Location: [H]igh Altitude Lab

Post by Ice Czar » Sat Nov 27, 2004 11:12 am

direct LN2 immersion wouldnt work too well
(too much of a good thing :p )
but that doesnt mean you need to actually pump the Fluorinert
like in a supercomputer

check both mission submersible articles by Ramil Tranquilino
http://www.octools.com/index.cgi?caller=supercool.html
(they are quite old now)

an old forumM8 of mine has the best longterm solution Ive seen
a -111 degree Cascade Phase Change

actually there are a few lower than that
ugly as hell but Lardarse was never one to let esthetics stand in the way of function :p

Rusty075
SPCR Reviewer
Posts: 4000
Joined: Sun Aug 11, 2002 3:26 pm
Location: Phoenix, AZ
Contact:

Post by Rusty075 » Sat Nov 27, 2004 3:32 pm

Maybe I should have phrased it more carefully, "how many home PC users are doing direct die cooling?" :lol:

I don't see it really taking off for end users either, for many of the same reasons that watercooling will never be mainstream.

And don't even get me started on submersive cooling....it's a bit of a sore spot with me....we have a bit of a track record. :lol:

Tyrdium
Posts: 272
Joined: Sat Jun 12, 2004 7:29 am
Location: Boston, MA
Contact:

Post by Tyrdium » Sat Nov 27, 2004 3:36 pm

Hmmm... What would happen if beer were used in a liquid cooling system? That'd be a fun article to read... :D

Edward Ng
SPCR Reviewer
Posts: 2696
Joined: Thu Dec 11, 2003 9:53 pm
Location: Scarsdale, NY
Contact:

Post by Edward Ng » Sat Nov 27, 2004 3:46 pm

The fundemental question remains...

How silently can these alternate extreme cooling methods be done?

-Ed

Tyrdium
Posts: 272
Joined: Sat Jun 12, 2004 7:29 am
Location: Boston, MA
Contact:

Post by Tyrdium » Sat Nov 27, 2004 3:53 pm

I suppose you could use a Resterator for the non-direct liquid cooling, so fairly quiet...

Edward Ng
SPCR Reviewer
Posts: 2696
Joined: Thu Dec 11, 2003 9:53 pm
Location: Scarsdale, NY
Contact:

Post by Edward Ng » Sat Nov 27, 2004 3:57 pm

Tyrdium wrote:I suppose you could use a Resterator for the non-direct liquid cooling, so fairly quiet...
What'd different about this from conventional liquid cooling (which I did not mean to include under alternate extreme cooling methods, thus the word, "alternate")?

-Ed

Ice Czar
Posts: 20
Joined: Tue Nov 16, 2004 9:22 pm
Location: [H]igh Altitude Lab

Post by Ice Czar » Tue Nov 30, 2004 7:05 am

Edward Ng wrote:The fundemental question remains...

How silently can these alternate extreme cooling methods be done?

-Ed
very quite
one small micropump being the only pump and its a canned magnetic drive
(correction its 10 atmospheres) a cut and paste
-------------------------------------------------------------------------------
fluorinert spray cooling is available for the desktop
just a few links
19" & 23" SprayCool Rackmount Enclosures (and workstations) pdf
Micromachined Spray Cooling (NASA)
Spray Cool Technology PDF
Magnetic Drive Gear Pumps



its an area of active research for me
they even have spray cooled "waterblocks" for lack of a better word
the trick with all this is actually dealing with fluorinert which will leak through a seal that would withstand 10 atmospheres of pressure if it was water, the proper selection of the flourinert type for the phase change, the proper selecction of the pump, and they aint cheap pumps, the proper selection of the atomizers\nozzles which impact the phase change

think of it as a hybrid water cooled phase change rig
its both simpler in operation, but far far tougher to determine the right component selection
Right now I know of no overclocker employing it, and have yet to determine the lowest possible temperature it could reasonably be used for
(again that would depend on a number of factors)
but Im very interested in it

few more
http://misspiggy.gsfc.nasa.gov/tva/thermal/index.htm
http://www.vita.com/vso/vso200002/apw.pdf

beli ninja
Posts: 25
Joined: Sun Jan 14, 2007 5:42 pm
Location: Zagreb, Croatia
Contact:

Post by beli ninja » Sun Mar 23, 2008 3:27 am

just an update on SprayCool:
It's a general principle of electrical engineering that volts and fluid don't mix, which is why one is ill-advised to balance a Pepsi can on one's CPU. I first saw SprayCool technology at the Navy League show in Washington a year ago, and the sight of a computer in a transparent case, with a gentle mist of fluid spraying across the circuit cards, was impressive.

I kept waiting for the normal flash-POOF-evil-smellsequence, but it didn't happen. The SprayCool people explained that the fluid being used was nonconductive and noncorrosive (which eliminates Pepsi on two counts), and that they were looking for some tough cooling problems to solve.

This quest has borne fruit. Northrop Grumman has chosen SprayCool technology for the Airborne Signals Intelligence Payload (ASIP) system that is being developed for the U-2 Dragon Lady.

It's a difficult problem because the ASIP package is installed in an unpressurized area on the U-2 - and at altitudes pushing 80,000 feet, there's not a lot of air to cool anything with. Moreover, the electronics need to be cooled under some circumstances and heated in others. The SprayCool chassis has been flight-tested on the U-2 over the past year and has now been selected for production and sustainment.

SprayCool chassis are also used on the Predator and Global Hawk. Not only is the system the only way to cool modules above 70,000 feet in an unpressurized environment (other than built-in liquid cooling) but it is also closed-loop, keeping airborne contaminants away from the electronics.

EDIT BY ADMIN: DO NOT POST LONG URLS!!!!!!

Post Reply