Overclock.net banner

1 - 20 of 43 Posts

·
Premium Member
Joined
·
5,573 Posts
Discussion Starter #1
I posted this on another thread. But, wanted feedback on the idea. It seems sound. And inherently leak proof. Please read and fling your opinions at me.

1) Take standard 1/4" copper tubing and form a CPU sized spiral. The center of the spiral is where the tube bends straight up and away.
2) Melt a load of solder onto the bottom of the spiral.
3) Belt sand the solder side
4) Lap the solder side
5) Take the stock HSF from intel (for example) and tear out the copper core leaving only the spring clip.
6) Put some AS5 down and sandwich the copper spiral between the CPU and the spring clip.
7) Clamp on some hoses and feed them to an el-cheapo pump and reservoir.
 

·
Registered
Joined
·
2,049 Posts
the solder MIGHT melt if anything goes wrong lol, that would suck

its worth trying however, you would have to get really good conductive solder tough.
 

·
Banned
Joined
·
3,194 Posts
no it woulnd melt

no was it would get hot enough

It would probably work better if you melted a bunch of copper on it instead of solder

BTW who runs a 1/4 CPU block unless they have a cheapo kit
 

·
Premium Member
Joined
·
5,573 Posts
Discussion Starter #4
Quote:

Originally Posted by thehybridpyro
the solder MIGHT melt if anything goes wrong lol, that would suck

its worth trying however, you would have to get really good conductive solder tough.
Anyone know the thermal conductivity of silver vs copper?
 

·
Premium Member
Joined
·
22,477 Posts
Hi Franken, The only problem might be in the heat exchange of it...You want the water to stay over your chip for a short time to maximize the heat dissapation..The thinner the bottom of the WB in general is the way to go..Thats why a pure impingment WB cools the best but ruins the flow...An example is the Swifty Storm...
 

·
Premium Member
Joined
·
5,573 Posts
Discussion Starter #6
Quote:

Originally Posted by born2killU
no it woulnd melt

no was it would get hot enough

It would probably work better if you melted a bunch of copper on it instead of solder

BTW who runs a 1/4 CPU block unless they have a cheapo kit
High temp silver solder wouldn't melt. But you would need a butane torch to braze it on. 1/4" is the copper pipe diameter. The diameter of the finished spiral would be about 1 - 1.5 inches in diameter.
 

·
Premium Member
Joined
·
5,573 Posts
Discussion Starter #7
Quote:

Originally Posted by ira-k
Hi Franken, The only problem might be in the heat exchange of it...You want the water to stay over your chip for a short time to maximize the heat dissapation..The thinner the bottom of the WB in general is the way to go..Thats why a pure impingment WB cools the best but ruins the flow...An example is the Swifty Storm...
Yeah...so you would have to belt sand the block far enough to cut into the tubing and you would not know how dangerously thin that wall would be.

Not a high performance solution. I guess the question is, would it be better than an heatsink/fan solution?
 

·
Premium Member
Joined
·
22,477 Posts
Quote:

Originally Posted by FrankenPC
Yeah...so you would have to belt sand the block far enough to cut into the tubing and you would not know how dangerously thin that wall would be.

Not a high performance solution. I guess the question is, would it be better than an heatsink/fan solution?
I cant answer that I'm not sure...Might make an interesting experiment to someone...
 

·
Registered
Joined
·
2,049 Posts
Quote:

Originally Posted by FrankenPC
Yeah...so you would have to belt sand the block far enough to cut into the tubing and you would not know how dangerously thin that wall would be.

Not a high performance solution. I guess the question is, would it be better than an heatsink/fan solution?
i might give it a try
what does copper melt at? and if i cant melt it at home what kind of place can melt copper?

also high silver solder should have better thermal conductivity then regular solder, not as good as aluminum or copper though
 

·
SNERP!!!
Joined
·
4,183 Posts
Quote:

Originally Posted by FrankenPC
Yeah...so you would have to belt sand the block far enough to cut into the tubing and you would not know how dangerously thin that wall would be.

Not a high performance solution. I guess the question is, would it be better than an heatsink/fan solution?
It would depend on your pump & rad, and it would depend on how well built the home made block is. Also silver is better than copper for thermal Xfer
 

·
Premium Member
Joined
·
5,573 Posts
Discussion Starter #11
Quote:

Originally Posted by thehybridpyro
i might give it a try
what does copper melt at? and if i cant melt it at home what kind of place can melt copper?

also high silver solder should have better thermal conductivity then regular solder, not as good as aluminum or copper though

You are basically doing what a plumber would do. Braze high silver content solder onto copper. It's how water piping is done in homes.
 

·
SNERP!!!
Joined
·
4,183 Posts
Quote:

Originally Posted by thehybridpyro
i might give it a try
what does copper melt at? and if i cant melt it at home what kind of place can melt copper?

also high silver solder should have better thermal conductivity then regular solder, not as good as aluminum or copper though
you are making my brain hurt...read up on some stuff please. you don't just melt some random metal onto another metal and expect it to stick.
 

·
Premium Member
Joined
·
5,573 Posts
Discussion Starter #13
An add on thought: Place the coil of copper between two wood plates and put it in a vice. squeeze the round shape into more of a square shape and fill the gaps. Then you could use less solder and expose more of the copper surface to the CPU.
 

·
Registered
Joined
·
2,049 Posts
wow 1084.62 °C (1984.32 °F) for a melting point

and Snerp, this is just preliminary info gathering here
im not looking at specifics yet
 

·
Premium Member
Joined
·
5,809 Posts
Quote:

Originally Posted by FrankenPC
Anyone know the thermal conductivity of silver vs copper?
Silver is about 15% better than copper.
 

·
Premium Member
Joined
·
4,746 Posts
In response to your squashing idea, 1/4" is pretty darn small to begin with, would it be wiser to take 1/2" copper pipe and do the vice trick and then have like 3/4"x1/4" copper ovular tubing? That would allow more water to pass through faster and might give you more surface area...

What ya think?
 

·
Banned
Joined
·
3,194 Posts
but with 1/2 in tubing it would only be able to go in one little circle unless you wanted a huge waterblock

EDIT: 3/8 might be a better solution
 

·
Premium Member
Joined
·
5,573 Posts
Discussion Starter #18
Quote:

Originally Posted by kidwolf909
In response to your squashing idea, 1/4" is pretty darn small to begin with, would it be wiser to take 1/2" copper pipe and do the vice trick and then have like 3/4"x1/4" copper ovular tubing? That would allow more water to pass through faster and might give you more surface area...

What ya think?
Not a bad thought. This weekend I might just put my money where my mouth is and try both 1/4 and 1/2.

Also, after squashing, prior to soldering...I'm going to take a rounded pin punch and make dimples all over the bottom side of the spiral. This will introduce turbulence into the flow.

This is beginning to sound viable an do-able without a milling machine.

EDIT: Yeah, 3/8th " sounds better. But, 1/4 inch is readily available. I'm thinking "poor man's" solution.
 

·
Premium Member
Joined
·
22,477 Posts
Quote:

Originally Posted by FrankenPC
Not a bad thought. This weekend I might just put my money where my mouth is and try both 1/4 and 1/2.

Also, after squashing, prior to soldering...I'm going to take a rounded pin punch and make dimples all over the bottom side of the spiral. This will introduce turbulence into the flow.

This is beginning to sound viable an do-able without a milling machine.

EDIT: Yeah, 3/8th " sounds better. But, 1/4 inch is readily available. I'm thinking "poor man's" solution.
If it doesn't cool well at first you could also pinch the tubing in some after the heat sink to slow down your flow some more...Let us know how it does..
 

·
Premium Member
Joined
·
4,746 Posts
Quote:

Originally Posted by FrankenPC
Not a bad thought. This weekend I might just put my money where my mouth is and try both 1/4 and 1/2.

Also, after squashing, prior to soldering...I'm going to take a rounded pin punch and make dimples all over the bottom side of the spiral. This will introduce turbulence into the flow.

This is beginning to sound viable an do-able without a milling machine.

EDIT: Yeah, 3/8th " sounds better. But, 1/4 inch is readily available. I'm thinking "poor man's" solution.
3/8" would be the optimal I think, but with 1/2" I wouldn't bother with a spiral but something more like this:


With more rounded corners of course... I dunno... Looks a bit harder now lol
 
  • Rep+
Reactions: FrankenPC
1 - 20 of 43 Posts
Top