Experimental radiator \build log!!!

[toolmaker_03]

well here are the radiators, my problem is the configuration of them, simply too many possibilities, so if you could find the time to give your opinions, I would really appreciate it.
I also need advice on a type of shroud for the radiators, the purpose is to inshore good air flow through the radiator.
and lastly how to attach the fans to the radiators?



all advice is welcome and thank you

 

[rubix_1011]

They look nice; what is the tubing ID size? The only thing that might limit performance is round tubing for more laminar instead of turbulent flow.

Where did you order those through or have them made?

 

[Motopsychojdn]

Shiny!
Moto

 

[toolmaker_03]

it's 1/4 in OD on the Rads

 

[toolmaker_03]

They look nice; what is the tubing ID size? The only thing that might limit performance is round tubing for more laminar instead of turbulent flow.

Where did you order those through or have them made?

it's 1/4 in copper so I was planning to use 3/8 in OD 1/4 in ID plastic tubing what would you suggest?

Ebay $ 10

 

[rubix_1011]

Might be rather restrictive; you might consider running 2 in parallel at a time, if so. How many do you plan on running in all...all of them?

 

[toolmaker_03]

Might be rather restrictive; you might consider running 2 in parallel at a time, if so. How many do you plan on running in all...all of them?

I have 12 now, I didn't think that 6 was enough. so your saying 6 in parallel with 6, first split the main line coming in into two lines, than recombine after the radiators, as it travels back to the main case.

 

[rubix_1011]

I'm still wondering how restrictive all those radiators will be in any configuration.

Remember, when a normal watercooling radiator has flow through it, the inlet tank collects the incoming water and then splits it to many parallel flowing channels, or tubes. It then flows into the end tank, 'around the corner' at that end and back through an equivalent number of channels before collecting in the outlet tank.

These radiators have a single path all the way through the rad, bending and twisting along a single channel. Then repeat several times more for as many you have in the setup.

It would be comparable to a 6 lane super highway and a single lane road, both trying to carry the exact same amount of traffic at the same speeds. The 6 lane superhighway will allow cars to move at high speeds along many parallel channels in the same direction. If you attempt to have the same amount of traffic on a single lane highway, there is a lot of restriction and all traffic is dependent on what is in front in order to progress.

 

[noob2222]

Might sound funny, but blow through one and see how hard it is to move air through them, then consider the pump will be trying to push water through that many. They look like the thermaltake rads.

 

[toolmaker_03]

Might sound funny, but blow through one and see how hard it is to move air through them, then consider the pump will be trying to push water through that many. They look like the thermaltake rads.

so than your saying have the line split 12 times, 1 split for each rad than recombine after the rads back to 1 line.

 

[rubix_1011]

You can try, but your flow will drop quite a bit. However, this will also mean your restriction will be far lower as a result.

What pump are you going to use?

 

[noob2222]

Splitting will definately help, how many will be trial and error. Those rads are usually quite effifient, id suggest trying 6 first.

 

[noob2222]

Something like this would look cool for the manifold to split and rejoin the rads.

http://koolflowmanifold.com/zencart/index.php?main_page=index&cPath=3&zenid=aljbm5omcl6k6g28ql801980q7

 

[toolmaker_03]

You can try, but your flow will drop quite a bit. However, this will also mean your restriction will be far lower as a result.

What pump are you going to use?

swiftech mcw 655

 

[toolmaker_03]

Something like this would look cool for the manifold to split and rejoin the rads.

http://koolflowmanifold.com/zencart/index.php?main_page=index&cPath=3&zenid=aljbm5omcl6k6g28ql801980q7

I could make a block out of copper like that than Sauder hard lines with fittings for the tubing

 

[toolmaker_03]

ok, I went and did the math, and I would need 4 to get equivalent cubic inch as a 3/8 in ID line. so I built a couple of proto types to see how it would flow, and putting 5 rads together gives the same return pressure as a single 3/8 in ID line. putting 6 rads together it drops a little bit, but at 4 its worse. go figure.

 

[toolmaker_03]

Just thinking, go with the gigabyte OC edition X58 board... orange with those copper rads would be something

I have been discussing that very same topic, sort of. but this is going on a crosshair II formula MB, but still a lot of copper on this board to.

 

[toolmaker_03]

Very true

hay, would happen to know about fluid dynamics, or the physics surrounding that topic.

 

[Lutfij]

did anyone notice that the openings of the radiator is on its side and not on its face - this would cause the channel of airflow to be quiet limited ... or maybe attach 3 rad ontop of each other and have a fan attached to its side.

those rads look - coppery!

 

[toolmaker_03]

did anyone notice that the openings of the radiator is on its side and not on its face - this would cause the channel of airflow to be quiet limited ... or maybe attach 3 rad ontop of each other and have a fan attached to its side.

those rads look - coppery!

thats a good idea, sense it looks like I will be paralleling 6 on 6 maybe even 6 high with 2 fans. but how to attach the fans?

 

[Lutfij]

working with copper is quiet tricky as its maliable and can also stain cos of the oils your skin produce. attaching it will be a challenge. zipties maybe but thas my suggestion - other will def have a better scope of delivery.

 

[toolmaker_03]

working with copper is quiet tricky as its maliable and can also stain cos of the oils your skin produce. attaching it will be a challenge. zipties maybe but thas my suggestion - other will def have a better scope of delivery.

I am considering a liquid rubber to bond the fan frames directly to the radiators, I will need to build a form to keep the rubber from running while it hardens do you have any suggestions on types of rubber i could use or good material for the frames that the rubber won't bond to?

 

[Lutfij]

i dunno, acrylic maybe but liquid rubber aint a good idea, what will you do when the fans need to come off for cleaning, like i said before - the copper is malleable so will bend at the slightest tug/pressure/while prying the fans off.

 

[toolmaker_03]

i dunno, acrylic maybe but liquid rubber aint a good idea, what will you do when the fans need to come off for cleaning, like i said before - the copper is malleable so will bend at the slightest tug/pressure/while prying the fans off.

my other idea is to make circles out of copper tubing jest large enough to catch the 4 corners of the fan, than soldering the circle to the radiators making a pop in fitting for the fans, no screws needed. than cutting rubber frames that fit the bottom of the fan frames. than attaching that to the fans with some silicone, easy to remove later so I can replace the fans. but hard to make, without crushing the copper tubing in the process, it's a tight turn.

 

[photonboy]

What is the purpose? I could not find it here.

Is it for a single CPU?

Keep in mind that radiators still need FANS to blow through them to be effective. I'm trying to visualize a setup that makes sense and I simply can't with 6 coolers.

If I was going to water-cool a CPU I'd likely use a Corsair H100 (2x120mm) with a supported case.

I think you will find this problem much more difficult than you thought, but I applaud you for trying.

(You may wish to invest in a rubber mat that prevents water from going on the floor.)