all copper rig??????

[Cyaneed]

I've been engineering a WC system with a TEC cooler on the RAD and have been wondering if anyone has ever hard plumbed the case with all coper tubing? I would think that more copper surface area would help the water stay cooler and the soldiered joints at the components would surely prevent leaks if done right. I'm planning for the pump and RAD to be external and would take Vinyl tubing to that but inside the case what would be the downside of going all copper? The ridgidness can be overcome with good pipe layout and the fitting that will pass through the case can be the disconnection point should I have to remove said tubing for repairs.

I was also planning on fixing my Tuniq 120 to the hot side of the TEC so my chiller doesn't become a boiler. I'd think a Tuniq could disipate 300w don't you *hehehehehe*

[jdrom17]

I've seen people who used all copper tubing, so it is possible. As long as all the components are copper I don't see what the issue would be.

Sounds like an interesting project to say the least.

[MU71L4710N]

ive seen it done several times on this forum. i dont recall how the results were (i wanna say 1-2c improvement or something?), however i think i remember one of the persons who did it saying he didnt think it was all that practical to do because it makes removing the cpu/gpu/whatever else is blocked a ***** to take out because obviously the copper pipes arent going anywhere so it makes removal of the hardware impossible. so he had to bleed the system and take the pipes apart to do any maintenance on his rig. (there are ways around that though, i can draw u a pic if your serious about this project and want some ideas).

http://forums.extremeoverclocking.co...=copper+tubing

this thread has a pic of that done. im sure the one im thinking of is around here, too many pages come up on search tho

[Teyber]

http://forums.bit-tech.net/showthrea...ght=hard+lined

[Cyaneed]

Thanks for that link. I was coming up with squat searching for just copper. I still need to do some research on flow rates and heat transfer needs for my C2D but so far my vision is to use 3/8's annealed copper tubing with concentric bends to eliminate turbulence.

I've also come to a dillema. It seems there is conflicting opinions whether it is better to buy a waterblock or mill one. Is the D-Tek Fuzion really the best there is and can one that is milled from scratch by a beginner have any hope of beating its performance? Is turbulence, high flow, or jetted impact the best to increase surface area and heat transfer. It seems the review links I follow are a lot of dead links or page not found's.

I'm probably way overthinking this project because I've seen complete idiots get a working system. But I want to build a rock solid system that will keep tempts as steady as physically possible. Maybe even sub-ambient with water. Then again it might be just as well to fill it with Ammonia and go for the phase effect . Yup I'm definitely overthinking this.

[Teyber]

if you have have a mill, definatly make your own! you will get so much bragging rights...

[MU71L4710N]

Originally Posted by Cyaneed

Thanks for that link. I was coming up with squat searching for just copper. I still need to do some research on flow rates and heat transfer needs for my C2D but so far my vision is to use 3/8's annealed copper tubing with concentric bends to eliminate turbulence.

I've also come to a dillema. It seems there is conflicting opinions whether it is better to buy a waterblock or mill one. Is the D-Tek Fuzion really the best there is and can one that is milled from scratch by a beginner have any hope of beating its performance? Is turbulence, high flow, or jetted impact the best to increase surface area and heat transfer. It seems the review links I follow are a lot of dead links or page not found's.

I'm probably way overthinking this project because I've seen complete idiots get a working system. But I want to build a rock solid system that will keep tempts as steady as physically possible. Maybe even sub-ambient with water. Then again it might be just as well to fill it with Ammonia and go for the phase effect . Yup I'm definitely overthinking this.

i seriously doubt a beginner will be able to mill a block that outperforms the dtek, i would just buy the dtek.

you cannot cool sub ambient with water without a chiller or something, just a water loop with a rad wont cool anything beyond ambient, and since its constantly heated the coolest it will really get is 5/10c over ambient.

[Teyber]

the phase effect doesn't have to do with ammonia...

yeah milling a block will not get you better temps then a d-tek, just that it would be fun

[Cyaneed]

Originally Posted by MU71L4710N

i seriously doubt a beginner will be able to mill a block that outperforms the dtek, i would just buy the dtek.

you cannot cool sub ambient with water without a chiller or something, just a water loop with a rad wont cool anything beyond ambient, and since its constantly heated the coolest it will really get is 5/10c over ambient.

Okay so milling is out. not a problem since I was kinda worried about that concept anyhow. And I intend to have a chiller of sorts. I was going to put my Tuniq 120 in the reservoir attached to the cold side of a TEC and some cheap HS on the hot side with a couple 120's to blow the heat away. Also had an idea for a TEC Waterfall to add some ambiance to the setup and some white noise to drown out anything else. That idea got way too complicated trying to reduce evap so I nixed it, and I wanted to use what I had already to minimize costs. Not sure how cold I can get the res but I'll be certain to test it before I put it into service.

Originally Posted by Teyber

http://forums.bit-tech.net/showthrea...ght=hard+lined

Awesome link man. I was planning to go more industrial and the piping will be a lot cleaner without couplings but that was the idea I had. So much for original thinking huh. Still a great thread to read and get inspiration from.

[MU71L4710N]

Originally Posted by Cyaneed

Okay so milling is out. not a problem since I was kinda worried about that concept anyhow. And I intend to have a chiller of sorts. I was going to put my Tuniq 120 in the reservoir attached to the cold side of a TEC and some cheap HS on the hot side with a couple 120's to blow the heat away. Also had an idea for a TEC Waterfall to add some ambiance to the setup and some white noise to drown out anything else. That idea got way too complicated trying to reduce evap so I nixed it, and I wanted to use what I had already to minimize costs. Not sure how cold I can get the res but I'll be certain to test it before I put it into service.

Awesome link man. I was planning to go more industrial and the piping will be a lot cleaner without couplings but that was the idea I had. So much for original thinking huh. Still a great thread to read and get inspiration from.

i dont think thats practical, its too much surface area for a dinky little tec to cool, its not gonna work. ive seen people talk about doing that countless times and ive yet to see it done (doesnt mean it hasnt been...i just havent run across it yet) so that must say something.

if your trying to cool your water do it right, use a CHILLER out of a water dispenser or something. and post pics...pics are good.

[Cyaneed]

Well since I dont have a spare fridge laying around, do you have suggestions for a chiller or a link how I can build one without a spare fridge? I had the intention of using a 300w TEC on the Tuniq thinking it would get the fins frosty cold but with the Sintered heatpipes I'm not sure if they transfer cold as efficiently as they to heat. Point is I have a small TEC from one of those auto refrigerators that does get pretty cold surprisingly that I was going to use. I'd want to put that somewhere to assist cooling so suggestions are quite welcome.

As for pics, the best I could do at this point is drawings. I'm an engineer at heart so everything has to be planned out before I start construction. I'd love to get a CAD program to put it all into detail but I'm broke, hence the reusing a lot of parts. I'll keep on reading to get the info I need but links sure do make my job a ton easier. and yes I am a lazy a$$ just in case you were wondering.

[TDOGZZ]

a lil offtopic but... a member here used a coppper coil buried in the ground about 3-4 feet instead of using a rad.(he got good temps considering it was passive)

i can,t remember the members name but i posted in his thread if you want to search threw my old posts(about 2-3 years ago)

that copper plumbing pic in MU71L4710N,s link that wang chung posted is awesome!!

here it is..http://forums.extremeoverclocking.co...ht=copper+coil

[Cyaneed]

holy hell. Geothermal cooling. that belongs in the extreme section doesn't it? Thats just crazy, and incredibly not portable. I'd love that kind of thermal stability but I'm not planning on living in this house forever. At most I thought about getting a cheap A$$ $100 Window AC unit to put the rad in and even that is a bit more permanent and external than i had hoped to be. I'm the type who likes a clean and almost unnoticeable install so the external rad will even be as hidden and inconspicuous as I can make it. Probably mounted on the back in the outflow of the existing case fans. I dunno I'm still working out flow and heat transfer stuff but layout is next on my list of design aspects.

[Teyber]

why would you put the rad in it? my friend read some stickies on phase change.
http://forums.extremeoverclocking.co...d.php?t=101902
Please read that

[Cyaneed]

Thanks for that Teyber. sounds like a chiller is a lot more than I want to deal with. If anything I'll forget about water and just keep it on air with the AC unit to reduce ambient. Kind of like when I opened the window in the wintertime i can get nice clocks and steady RAM timings just with all my fans. I'm currently running 4 120mm 3 40mm and the built-in on the GTS. There is a lot of airflow going on. My original plan was to just cover the CPU and the GTS until I can afford to upgrade to a Blitz Extreme. The more I talk to you all the less prepared I feel to do this project. Taking this from concept to reality might be a huge mistake on my part.

Also is it better to have more water in the system or for the water to move through the system faster? I was thinking of doing 1/4 tubing and a higher pressure pump to compensate. Would give me more room for routing and bending. would also allow me to blend in with the heat pipes and make the cooling system look like its supposed to be there. I cant help it, I'm a perfectionist.

[Teyber]

i thin kyou should go watercooling. Just your image is a little off on sub-ambient temps.

[MU71L4710N]

very much correct. sub ambient is not gonna be just hooking a tec up to a res or putting a res in a fridge (that will kill a fridge extremely fast). if alot of work and modification is out of your window then definately stick to strait water unless you wanna spend alot of $$ on one that someone has already done.

what i was talking about is ive seen someone take the chiller out of one of those water dispensers (that uses the 10-15~ gallon plastic water jug on top of it that u see in offices and ****).

but if your wanting to cool a GPU also you dont want to go sub ambient, if i remember right GPU's have a hard time running at sub ambient temps.

[Cyaneed]

Originally Posted by Teyber

i thin kyou should go watercooling. Just your image is a little off on sub-ambient temps.

I was going for sub ambient not sub-zero freezing where I'd need to use Alcohol or something with a lower freezing temp. A cool down to 0C after the RAD would be fine for what I was planning. I'm not going for records, just something to show off and get more out of this chip than an 800MHz OC at best. then again when my new RAM gets here I might try for higher because that is what I think was holding me back before. I'm optimistic that I understand this board quite a bit better than when i first put it all together.

Additional Comment:

Originally Posted by MU71L4710N

what i was talking about is ive seen someone take the chiller out of one of those water dispensers (that uses the 10-15~ gallon plastic water jug on top of it that u see in offices and ****).

but if your wanting to cool a GPU also you dont want to go sub ambient, if i remember right GPU's have a hard time running at sub ambient temps.

I had that idea about the water cooler too. A 5 Gal reservoir and a nice source of cold refreshment to boot if I can get it to stay potable through the loop *hehehe* What would the cooling properties of Hinkley Springs mountain spring water be?

I did not know that about the GPU and since that is the second, if not first, major source of heat in my box thats kinda important. According to SmartDoc, it runs normally at 57c when my proc runs full load at 40c. And that is after I applied AS5 to all the GPU heat sources.

[MU71L4710N]

umm...ur trying to squeeze a dollar out of a dime.

your gonna get mad bacterial build up in your loop if you try to do that...theres a reason we use water additives in loops.

[Cyaneed]

It was a joke. but now that I think about it running a coil around the edges of the cannister would give me enough cooling I'd think and I wouldn't have to mod anything. Now to just find one cheap enough for me to talk the wife into letting me buy one. That is always the hardest part of any computer project.

Additional Comment:

Okay, I'm trying to decide the size of the tubing for my project and I need to know which is better for heat transfer. Volume of water in the loop or speed the water moves through the loop. I'm trying to decide between 3/8" or 1/4" because 1/4 is more workable in small spaces but if 3/8 is better regardless of flow speed then I'll just work around the bulk. Plus if I go 1/4" then I might do better to run stainless tube with compression fittings on the blocks. I can always anodize it with brass or copper for the shinny look or leave it steely for a brake line look. In this 3D Aurora case I definitely want to keep a metallic theme.

I also wanted to know if anyone had used valves to control and balance flow rates in parallel loops and if it has any benefits over a series loop. I read that flow rates and different pressure drops are the reason parallel loops aren't desirable but if you can control the flow and balance each channel then I'd think each component getting fresh cool water would be the ultimate. Again maybe I'm over thinking things.