Mousepots?

[Equinox]

I would like to ask some questions on dimensions of these devices. I can tell you that I think im looking at approaching this a littl different than copper. When I first looked at the device I noticed its made of a very high thermal conductivity. My thoughts are on the wall thickness and not the thermal conductivity of the material. I was looking for a material that has a high thermal cycle in such drastic of an environment. Copper itself has a very high thermal expansion coaff which makes for a low thermal cycle. This makes the contact with the proc not optimal.

I need the dimensions of the inner diameter, the outer diameter, there optimal length "no overkill", and the thickness of the wall between the processor and LN2. Just so you know I wouldn't want to test the item myself. I really don't see me putting the mousepot thru alot of thermal cycles.

I can tell you what material I am looking into using. It's an Aluminum Nitride ceramic which has a high thermal conductivity of 170 W/m K. Copper is around 300. It's thermal Expansion is around 3.1 "10-6 Celsius; (RT-1000 Celsius). As copper is around a 6. Im looking to make the material .03125 to .0625 wall thickness between the CPU and the LN2. I think this is how we will get optimal performance from the LN2. This will also allow me to test the materials for a flow thru style block. All the edges will be radiuses for optimal thermal shock resistance for the highest cycle times.

Feel free to tell me what you think
Eq

[Equinox]

Um noooo this thread needs ot be closed or deleted you could have just asked me!!

Wall thickness and base thickness is what makes a unit or breaks a unit. Pm and ask Kingpin for his new units dimentions as Im sure he'll say the same thing.

I wanted to know what all people think. I apologize if I stepped on any toes for asking this question in the forums. I was just wishing for a broad spectrum view from many people. I noticed you were the one posting ETA's on some mousepots. Are you affiliated with this manufacturer?

[n00b 0f l337]

He IS the manuafacter and he's already stopped making mousepots. He's going to make one more run of tubes sometime I beleive and then he's done. Other than that its his work and his schematics.

[afireinside]

Jason is the designer of the mousepot. There is no company involved. The dimensions are a bit of a secret I believe :)

[Equinox]

Well, Im thinking on .125 wall thickness for tubing and 9 inches tall. The dia will be a littl over the proc full dia. I will start the wall thickness of the base at .125 and just lap down till we have the desired effect. The ceramics manufacture will warrantee the product for a certian amount of cycles. Congrats Jason. That is inovation at it's fullest. Really neat to be talking with the individual who had such forsight!

[harleybro]

Jason's tubes are awesome. I believe the last batch was rev.11 or so. He has spent a few years working with differant designs which is why dimensions aren't readily passed out. :)

[G H Z]

Well I think it's cool to have someone coming in with some different ideas myself. That material sounds interesting but I don't think it will work well, and your base thickness sounds way too thin. Expansion really isnt a problem either as the weight of a container (CU) that size pretty much insures good contact on it's own tho it doesn't hurt to have a firm mounting method.

Still it would be cool to see some other designs put thru the paces, you never know what you might come up with.

[iS.Terminator]

Well I think it's cool to have someone coming in with some different ideas myself. That material sounds interesting but I don't think it will work well, and your base thickness sounds way too thin. Expansion really isnt a problem either as the weight of a container (CU) that size pretty much insures good contact on it's own tho it doesn't hurt to have a firm mounting method.

Still it would be cool to see some other designs put thru the paces, you never know what you might come up with.
ya, he can come up with something that'll make us :slobber:

[ilkkahy]

For what ive read dozen times best thickness is around 1/2".. thinner will make it colder. Most important thing is working surface area. These solid type containers conduct heat very nice on thick pipe walls so they will work good with dryice too. For ln2 use i think you should try getting plenty of surface area close to cpu. It seems you have good machinery to make any type of bottom to container so i think you should try something bit more complex and effective.

[Equinox]

For what ive read dozen times best thickness is around 1/2".. thinner will make it colder. Most important thing is working surface area. These solid type containers conduct heat very nice on thick pipe walls so they will work good with dryice too. For ln2 use i think you should try getting plenty of surface area close to cpu. It seems you have good machinery to make any type of bottom to container so i think you should try something bit more complex and effective.

I cant machine ceramics. Only metals. All exotics. But no ceramics, chit is way to tuff. Y I am doing the thin wall is a calculation of radiant thermal energy and ceramics nature to not expand so much under the conditions. My best calculations say that my thermal expansions will be able to withstand such thin walls. Could just be a massive f**k up thou:( I have a minimal quantity I have to buy:(. What it is really about is that i dont miss any material that would out perform copper. I would hate to get deep into this and find out l8r that ceramics would do so much better. I dont plan to try all kinds of stuff. Just the stuff that stands out on the thermal conductivity charts I have.

[ilkkahy]

Oh I didnt read the first post. You have intresting project going on. Even though im not quite following what kind of system are we talking about. Is this container/block going to be used in some closed Ln2 loop or just like in normal air pressure?

I put pm on you but there i still was under impression that you were making just a normal container.

[Equinox]

Oh I didnt read the first post. You have intresting project going on. Even though im not quite following what kind of system are we talking about. Is this container/block going to be used in some closed Ln2 loop or just like in normal air pressure?

I put pm on you but there i still was under impression that you were making just a normal container.

It is a normal container. It will research for the material needed for the final stage. final stage isn't a closed loop but a burn off method described in my other post. I think im lettin chilly1 have first crack at being my littl testing ho. "no offence chilly1 :), j/k" So, he has told me some info that i feel should be kept confedintial out of respect for him. but the mousepot is just to test materials. I surely couldn't surpass Mickeymouses tech because he is out to achive different goals. Thanks for all the intrest. Very cool. And I hope I can get a few of these out by 4th quarter!

[gloatlizard]

my friend i want to help you in your goal with a little help, for that system IF YOU CAN FIND IT you must use a single crystal of silicone, i did some researches some months ago, i need to refind them in order to let you look 'em;) i remember that some studies show how copper was inferior to the silicone single crystal at that cryogenic temperature;)

ps.it was a reaserch upon the cryogenics heatexchangers;) however i think that material will be very difficult to find:(

[gloatlizard]

however copper is very used also with liquid helium temperature so it will be enough for that xtreme use you want to do;)

[gloatlizard]

this is the incredible result of the research i did

http://www.nasatech.com/Briefs/Mar98/NPO30021.html

[G H Z]

Interesting, use silicon to cool your silicon chip. Wonder what that would cost :D

[gloatlizard]

yeah for sure it will have an incredible price...better copper:D they are cpu we have not to cool the hell:D

[speed bump]

^^^well time for someone new to step up to the plate I guess.

All the advancement in tube designs and now feed control remind of the last ten years in drag racing and N2O setup and design.

[[XC] moddolicous]

Whats colder: LN2 or liquid oxygen. I always heard the liquid O2 was really cold, but never thought about it for this use. I have NO LN2 or DI experience whatso-ever, so I really cant say much on this matter.

[zabomb4163]

pc ice had an idea even beyond the ln2 drip. single stage methane cooling system with the condensor sitting in an ln2 bath. then a normal evap could be used and achieve temps even lower than ln2 (assuming the chip could take it)

[alexio]

Whats colder: LN2 or liquid oxygen. I always heard the liquid O2 was really cold, but never thought about it for this use. I have NO LN2 or DI experience whatso-ever, so I really cant say much on this matter.

O2 condenses at between -100 and -150 at 1 bar, I don't know exactly but you can search this up in a sec.

[[XC] moddolicous]

It looks like Liquid O2 and LN2 have the same temps basically:
http://www.chem.t.u-tokyo.ac.jp/appchem/labs/kitazawa/SUPERCOM/65/65_e1.html

[TheMeatFrog]

Good luck with your endeavor, it sounds very cool!

The people who design these units put way too much time/effort into it to just hand out the specs. :cool:

OFFTOPIC
gloatlizard: You should try using the edit feature instead of posting 3 times in a row. :nono:

[the_chad]

Yeah good luck with your project, its great to see more and more people coming up with different designs, i cant wait to see some of these units (in the flesh/pictures)

exciting stuff

[snowwie]

ALSO THE PERSON THAT MAKES THE BEST UNITS ARE NEVER MIT GRAD'S BUT THE GUY THAT'S IS A FEILD RESERCHER OR IN THIS CASE A WORLDCLASS BENCHER .;) LOOK AT THE WORLDS BEST LN2-DRYICE BENCHERS THEY ALL HAVE DESIGNED THERE OWN UNITS
um, MIT grads don't bench

so jason, if I wanted to buy one of your containers I would have to promise not to talk about its constuction? these things are almost 200 bucks a pop, i can't believe the information this guy is requesting should be kept secret

[zabomb4163]

ALSO THE PERSON THAT MAKES THE BEST UNITS ARE NEVER MIT GRAD'S BUT THE GUY THAT'S IS A FEILD RESERCHER OR IN THIS CASE A WORLDCLASS BENCHER .;) LOOK AT THE WORLDS BEST LN2-DRYICE BENCHERS THEY ALL HAVE DESIGNED THERE OWN UNITS

Same for cascades. Chilly, Pc ice and others will tell you Math and calculations only get you so far the rest if from listeneing to what the user saids is needed here or there.



http://www.research.ibm.com/journal/rd/466/schmidt.html

"In late 1986, ETA Systems, Inc., a subsidiary of Control Data Corporation, shipped the first liquid-nitrogen-cooled CMOS computer system. This machine utilized direct-immersion cooling of single-chip modules immersed in liquid nitrogen"

thats right. a company was shipping units colder than your mousepots 19 years ago.


"circuit speed increased at a rate of approximately 1.4% for every 10°C reduction in chip temperature"

[Equinox]

Many people have asked me about certain projects they needed machinist help on. I started a thread in speak freely that you can get in contact with several Machinist, worldwide.

http://www.xtremesystems.org/forums/showthread.php?t=74780

Eq