New LN2 tube design.

[alexio]

Well how does this look?

http://img.photobucket.com/albums/v342/AlexKok33/PIC-001.jpg

This is a sideview. The base will be square with the dimensions of 75*75mm, the picture speaks for itself. The base will be 5mm thicker at the edges than at the middle, this has to do with the tools I use to cut away the copper (20-30cm diameter cutting blade, suitable for metals like copper).

I still have to deside bewteen 15mm/20mm (middle/sides) or 10/15mm (middle/sides).

The base has a total of 36 pins and the total surfase area without the tube itself is about 550cm2. The weight would be 1,85kg without the tube if I use the 15/20mm baseplate and a bit lower with the thinner base.

Any suggestions?

Thanks in advance,

Alex

[k|ngp|n]

I like it. The increased surface area/mass right above the core area is key. IMO if you could combine a design like that with a solid type design, it would be killer.

[Waus-mod]

I still think a round base would be better. as i already told you ;)
But still a great desing, with lots of surface to cool.

@ kingpin: Would that be possible to do with a cnc?

[alexio]

I like it. The increased surface area/mass right above the core area is key. IMO if you could combine a design like that with a solid type design, it would be killer.

Thank you for your possitive comment. What do you think is better, a 10/15mm base or a 15/20mm one? Remember, the block is huge (75*75mm).

[Jack]

Such a massive base will result in a very slow fall in temps and you'll be waiting a long time before it hits low temps. But it'll hold the temp very well.
So the base will always be a compromise...

What do you mean with 15/20mm? can't figure out what you're trying to say with that...

Making a solid container, thus making that base with the tube on it, out of a solid piece of copper isn't going to happen.
Solid containers are made on a lathe, so you're stuck with circulair groves on the bottom.
If you decide to make the container out of 2 pieces, then you could mill the base and your design won't be a problem. It'll take 2 houres to make it, but it can be done :D

what i forgot to mention,
Dont you think 75x75 is a little oversized? You have to add atleast 1/2" of insulation with LN2 and you'll end up with a 10x10 base. I don't think that'll fit on any motherboard.

[Waus-mod]

The size is to big indeed. a 54mm dia is way better. Alexio means a 15mm base or a 20mm (thickness). I would say about 10 to 15mm is the best! 20mm is to thick.

[G H Z]

I like it. The increased surface area/mass right above the core area is key. IMO if you could combine a design like that with a solid type design, it would be killer.

I have an idea for a solid container incorporating complex surface area designs, I might try to make one in the next month or so.

What are your thought's on the base thickness here Kingpin? How thick is mousepot base?

[k|ngp|n]

I have an idea for a solid container incorporating complex surface area designs, I might try to make one in the next month or so.

What are your thought's on the base thickness here Kingpin? How thick is mousepot base?

I'm not sure..I think somewhere between 5/8 and 1/2. There are a few changes I want to make for a solid container after using this one quite a bit....one being a bit more complex in design behind the cpu incorporating a raised feature in the center for a rod/arc setup. Other things are less mass in total (especially on the sides as this container takes FOREVER to come down to full ln2 temps, and subsequently if there is a problem you are waiting for a very long time for temps to come back up.), differently designed base, and possibly an aluminum version of the final design.

[speed bump]

I personally am leaning towards trying an aluminum tube again but if you have to weld make sure you do it very well becuase the on I built cracked when I LN2 tested it.

[k|ngp|n]

It will be solid so no welds.... just like the solid mousepot. When you tested aluminum, could the aluminum hold a good load like copper? I assume copper will be better, but still need to see for myself.
btw sorry for crapping out your thread alexio..I am still looking forward to seeing your designs.

[alexio]

It will be solid so no welds.... just like the solid mousepot. When you tested aluminum, could the aluminum hold a good load like copper? I assume copper will be better, but still need to see for myself.
btw sorry for crapping out your thread alexio..I am still looking forward to seeing your designs.

I believe the specific heat of Al is actually higher than that of copper. In other words you need more heat to up the temp of a certain amount (in weight) of Al by 1 degree than copper. Cooling it down is just the same effect but reversed.

I believe copper is the best because it has a lower thermal resistens. I will look up the thermal properties of both materials and post them here. The specific heat of Al per kilo is higher for sure but the density is also lower so the calculation is a little more complex.

I'm actually thinking of making something of silver. The price is about 200 euro/KG in the form of a blok but you can just get the value of the tube itself and of the silver powder back at a jeweler.

What if I would make you a silver container kingpin? It would be a little pricey but it could be done.

[MaxxxRacer]

alexio, the specific heat of aluminum is 899 J/KG*C while copper is 387 J/KG*C so it takes aobut 2.3 times the heat to bring aluminum up to (or down) a certain temperature.

You are also right that the thermal resistance of copper is MUCH lower than that of aluminum and does transfer heat faster than aluminum, thus making it the better choice for heat pumping.


On a side note. heat prefers to transfer through metals vertically rather than horizontally. With this in mind it would behove you (alexio) to make the center section (above the die) of the tube have more surface area in the vertical direction. Doing so will net you better heat transfer.

[Disposibleteen]

And the most eloquent post of the week goes to Maxxxracer. With all kidding aside he is correct, i doubt alum would hold load temp very well.

[alexio]

On a side note. heat prefers to transfer through metals vertically rather than horizontally. With this in mind it would behove you (alexio) to make the center section (above the die) of the tube have more surface area in the vertical direction. Doing so will net you better heat transfer.

Yes this is basicly true but far more important in a waterlock for example than in an LN2 tube. Solid tubes with no increased surfase area in the middle like the ones we see these days do seem to perform very well. In this case the cold is giving to the tube itself for a big part rather than through the base.

The design I made was just a design that is fairly easy to manufacture using basic tools (I don't like CNC, want to make stuff with my hands).

I made many complicated designs in my head, and even made some cool scetches after a few beers but I have desided that keeping it simple is the best thing. I believe that you don't need so much surface area right above the core and you don't need so much surface at all infact. I believe that a big mass and only a moderate amount of surface area is best and that's what I based the drawing in the first post of this thread on.

Just remember that making a good design isn't that hard, making a good design that is machinable is much harder.

[ilkkahy]

alexio, the specific heat of aluminum is 899 J/KG*C while copper is 387 J/KG*C so it takes aobut 2.3 times the heat to bring aluminum up to (or down) a certain temperature.

Yes but copper weights 8.91kg/dm^3 where aluminium only 2.70kg/dm^3 so they are pretty even when considering thermal capacity.

If someone is going to make aluminium Ln2 container just make it from some proper heatsink and put plastic pipe around it. Aluminium welding may crack in -200c temps and with this desing pipe doesnt add any cooling power.

GlacialTech Igloo 7200 Pro would make a fine efficient Ln2 base for example.

[SynGroW]

I think it's possible to use a ALU container, the japs sometimes use them afaik

I also found some pics of the solid copper containers from japan, might be usefull for designing a container.

http://v-tec.hp.infoseek.co.jp/ARC_REV62-11.jpg
http://v-tec.hp.infoseek.co.jp/ARC_REV62-21.jpg
http://v-tec.hp.infoseek.co.jp/ARC62/ARC_REV62-3.jpg

http://v-tec.hp.infoseek.co.jp/ARC72/ARC_REV72-3.jpg
http://v-tec.hp.infoseek.co.jp/ARC72/ARC_REV72-2.jpg
http://v-tec.hp.infoseek.co.jp/ARC72/ARC_REV72-1.jpg

There are more pics somewhere on that site, but the only way to find them is just trying by changing the rev. nr. in the URL (72 stands for rev7.2 for example) so we need to figure out where the pics are from the rev14.x because that seems to be the latest :p:

[speed bump]

It will be solid so no welds.... just like the solid mousepot. When you tested aluminum, could the aluminum hold a good load like copper? I assume copper will be better, but still need to see for myself.
btw sorry for crapping out your thread alexio..I am still looking forward to seeing your designs.

It got cold extremely fast(even with a 1 inch base) I never put it on a cpu but with Dry ice I got the base to stay stable -65°c pretty easily.

[Waus-mod]

Alex, when ı come back we talk about all the stuff... Greetz from alanya!

[G H Z]

I also found some pics of the solid copper containers from japan, might be usefull for designing a container.

Look at this ram cooling from v-tec, does anyone know if it is possible to get some of this stuff in the US?

http://v-tec.hp.infoseek.co.jp/ARC/system3t_02.jpg

Looks like aluminum to me.

Hey Alexio have you prototyped a base yet?

[alexio]

Look at this ram cooling from v-tec, does anyone know if it is possible to get some of this stuff in the US?

Looks like aluminum to me.

Hey Alexio have you prototyped a base yet?

No not yet, i have too many things to focuss at. I'm also busy to get the tools needed to start the build of my first single stager.

I will get a 3D render of my new design soon and I will post it here. It will be a solid tube with performance that has never been seen before. One of the tools needed for machining this container I'm going to have to make myself.

I'll see if I have the time to make this thing. School is starting very soon and I'm doing 2 years in 1 so I'll have enough on my mind.

[Jack]

alexio, what city are you from?

[alexio]

alexio, what city are you from?

Krimpen aan den IJssel, close to Rotterdam.

You can add me at msn for a conversation if you like: alexkok33 hotmail com

[Trouffman]

Look at this ram cooling from v-tec, does anyone know if it is possible to get some of this stuff in the US?

http://v-tec.hp.infoseek.co.jp/ARC/system3t_02.jpg

Looks like aluminum to me.

Hey Alexio have you prototyped a base yet?

Wonderfull, this ram-block is very impressive, but the container of ln2 is too small :(.

It's possible ti build an other design for this block ;). I will search for this in few days ;)

[speed bump]

Thats actually deeper than it looks I remember when they did that for the break 19secs and the pictures they had in the thread you could see the bottom its about 4inches deep.

[Jack]

Thats actually deeper than it looks I remember when they did that for the break 19secs and the pictures they had in the thread you could see the bottom its about 4inches deep.
here's a pic from a different angle:
Still doesn't look so deep tho'

http://v-tec.hp.infoseek.co.jp/ARC/system3t_01.jpg

[harleybro]

does anyone know if it is possible to get some of this stuff in the US?
Try and PM PRteamjapan. I have in the past. I did find there is a waiting list for parts and I am not sure which ones they will sell or not sell. It's worth a shot if your interested. ;)

[Waus-mod]

There stuff is really great, mayb if you look good at there design its possible to clone them. But a pm is mayb also very good as harleybro said