DIY radiator
 

http://users.bigpond.net.au/volenti/DIY_rad.jpg

http://users.bigpond.net.au/volenti/DIY_rad2.jpg


It consists of close to 2m of hard drawn 1/2'' copper pipe, 10 copper 90 degree elbows, a copper T peice and around 40-50m of 2.5mm2 multistrand copper wire.

I'm using 2x 12cm Enermax adjustable speed fans.

The particular wire I used is standard 240v earth wire (for houses)

It has a built in bleed/fill point, it relys on the mounting to the side of the case to provide a "back wall" to direct the air out through the fine wire.

I'll hook it up to my system now and will report back with the results.

ingenious how you have kept the air flow away from the 'fins',
afraid of icing ?

Speechless
 

awww???? I'm not sure what to think.

someone just looked at the pictures and jumped to a conclusion without actually reading what I wrote didn't they?

http://users.bigpond.net.au/volenti/DIY_rad3.jpg

http://users.bigpond.net.au/volenti/DIY_rad4.jpg

http://users.bigpond.net.au/volenti/DIY_rad5.jpg

http://users.bigpond.net.au/volenti/DIY_rad6.jpg

After cooling the cpu ( 2100+@2.405gig, 1.9v) at full load (cpu burn) for an hour and both fans running at close to full speed the water temp has stablized at 7 degrees above ambient, there's about 500ml of water in the system.

You stripped all that wire by hand?? :eek:

haha, not as painfull as it sounds, secure one end, stretch out a long length and run a sharp knife along the insulation, easy :)

Wow that thing looks like a beast. How do those temps compare to what you've used before? Is it replacing a rad or your evap unit? Also what model is that Iwaki? I've been considering the MD15 but I'm unsure about the size.

peace.
unloaded

it's not really replacing anything, it's just an expriment, I do normally use evaporative cooling, and compared to that it's 10 degrees c worse off;) It'll do for lan parties though.

the pump is a MD 20.

here's the dimentions for the md 15;

From the results it seems not very efficient compared to a heatercore. I'll guess a couple reasons for that:

- It seems to have two water loops, in that case the shorter one is going to take most of the flow.

- The copper wire lenght is too much for the width, the heat is dissipated on the first cms and the rest is at ambient temp.

I think in a better design that has an intermediate stage consisting on severaL smaller parallel tubes running between the inle and outlet manifolds, and the wires soldered to the small tubes that are a lot closer together.

oh those short tubes that look like their a water short cut are only structural, their not part of the water flow path.

though your second comment has me puzzled, if almost all the heat was being dissapated within the first few centmeters I wouldn't be seeing the 7+ degrees above ambient temp that I'm measuring from the built in fill tube, which is almost at the end of the water channel length.

I think :shrug: that he's saying that the wires running between the copper tubes are too long because they serve their purpose entirely (or close to) in the few centimeters next to the tubing.

Maybe, though that makes sense to me.

Is just like the difference between a small radiator vs. a large one. If all conditions are the same, both dissipate the same energy, but the small one requires a larger delta temp.

In your case, the usefull area of the wires is smaller, and all the airflow passing across the "cold" zone of the wires is wasted.

I'd like to know how much benefit you believe you've reaped from the copper wire. I've thought of incorperating some cu wire (god knows I've got enough of the stuff!) into some of my rads, but I didn't think the difference would be noticeable.