Opinions on my design

[p4ocer]

Wuttcha think?

and how thick should the copper be?

how far down should the holes be drilled down?

[peepingdan]

You might want to put the intake directly over the processor core.

[p4ocer]

ya...ive changed the plan so the two circles will now be outputs...and i will use an input above the core....

Would i be better off using a different design?

[p4ocer]

Originally Posted by p4ocer

and how thick should the copper be?

how far down should the holes be drilled down?

any answers??

[Etacovda]

Ive heard that around 2mm from the base of the block is about optimal - thats what im using at any rate.

[buzzby]

The base on my black is 2.5mm thick. i think its a prob a bit think i would go for 2.0mm. You really only need to make sure that the base will not bend under the load required to secure it to the cpu.

Buzz

[EDIT] Messed up. didn't put decimal placed in[/EDIT]

[bigben2k]

This is a variation of the #Rotor block... 3 to 5 mm baseplate (at the thinnest point).

[Gulp35]

If it is going to be a variation of a #rotor block all of the holes need channels connecting them to the 6 holes surrounding it (I noticed this on some of the holes but not others , I'm guessing that it is supposed to be but wasn't drawn)

Also if this is going to be a rotor block, from rotor himself, it is better for the holes to be in a square grid not hexagonal (not enough copper to water ratio, also something about that it doesn't cause as much turbulence b/c there are more ways for teh water to flow)

If you wanted to be slightly different and do a rotor you could rotate a square grid 45 degrees so that it is more of a diamond.


Then again if this is a cascade offshoot you need to look back a year or so in the forums to find the "new design idea" thread by bb2k (so many things started he has).

PS How is the "radius" coming allong? I haven't been checking up for a long while.

[bigben2k]

On hold, pending a volunteer CNC'er. Looks like Jon Fettig is interested, plus I've got some more work for him...

[p4ocer]

ok....new one...Would have 3 barbs....

[#Rotor]

Now, on that last one.... are you going to chill a TEC with it? if not, then remove as much of the holes that are not productive in cooling the core area or acting as defusing/collecting areas for incoming and outgoing fluids.

in other words if you are going for a 2 inlet block the hole pattern might end up looking like an hour-glass, with the thin part over the core....

I would go out on a limb and say that if my block in the test was of such configuration, it might have faired considerably better than the one tested, reason being that the hour-glass setup would much better concentrate the cooling to the location over the core...

[jaydee]

What Rotor said. Here is one of my Rotor variations.

pH has it in hand and will be testing it sooner or later.

[p4ocer]

im not using a tec but i do have a p4 so the area that needs to be cooled is quite a bit bigger than the amds...

Edit--Here is a diff idea that i think may work better...looks like an S

[winewood]

Your working too hard. The only reason you need a pin is to remove heat. Why do you need pins along the entire flow? Jaydee has it right. Simply mill out those unnecessary pins if you want your inlet/outlets further out. (for amd). For Intel, the heat area is bigger slightly, but not the entire heatspreader. The tips of the S or the tops of the I will only add restriction.

[#Rotor]

Originally Posted by winewood

Your working too hard. The only reason you need a pin is to remove heat. Why do you need pins along the entire flow? Jaydee has it right. Simply mill out those unnecessary pins if you want your inlet/outlets further out. (for amd). For Intel, the heat area is bigger slightly, but not the entire heatspreader. The tips of the S or the tops of the I will only add restriction.

true if you have a mill..... but believe me, the pins are much easier if you are limited to a drill-press and dremel....

[#Rotor]

something like this... the modular nature of this manufacturing method lends itself well for customizing the block to fit any core size... that is the beauty of it...

[p4ocer]

So where would i get the best performance...

The hour glass or something like jaydee made like this

[p4ocer]

bump bump bump

[DrCooling™]

Before


After



this pictures it's look like the original idea?
it has approximately 6750 square milimeters of interchange surface
Block made by me

translate by gmod :P

[|kbn|]

DrCooling that looks amazing, and original. Where will the inlets and outlets be though, as I cna see it making a huge difference to flow restriction. Im guessing the barbs should be at the left and right of the block. What will you be cooling with it? It looks to big for a cpu, so tec?

Heres my nf7-s nb attempt at a rotor block: http://server5.uploadit.org/files/kbn2k3-nbholes48d.JPG

[DrCooling™]

http://img12.imageshack.us/my.php?lo...e=DSC00250.JPG

http://img12.imageshack.us/my.php?lo...e=DSC00248.JPG

this block is for a cpu, it has a thick base to transfer heat to the sides and to take advantage of the great exchange surface. it's moderately restrictive but it doesn't matter because it has a great performance

it is compatible with peltier too

[JeePee]

Originally Posted by DrCooling™

he is compatible with peltier too

Why do you say it is compatible with a pelt/TEC? Just place a small copper plate under the block or what?

[DrCooling™]

yes, is the evolution of this model

http://www.drcooling.cl/galerias/cpu...erblock_d6.jpg
http://www.drcooling.cl/galerias/cpu...erblock_f6.jpg


http://www.drcooling.cl/galerias/cpu...es/image30.htm



Moore density, moore barbs in the same surface, is light and cheap, ...etc..

[jaydee]

My most recent block.



Made spacifically for CPU only.