New water block design...opinions??
 

I have posted this at hard forums...but I wanted to get your opinions... Been reading these forums...but this is my first post:)

We have come up with our preliminary design for our water block. We drew it up in a CAD program to get a good three dimensional view that can be directly imported into the mill.

We were just wanted to get some opinions on our design before we get it milled out. We will probably mill it in a week or so.

Here is a picture of the block:
http://cc.usu.edu/~roblb/Full_slice_web.jpg
The block is about 1/2" thick. The fins and walls separating the channels are 1mm thick. You can see the little 2mm divets on the bottom of the channels. The channel has a 11mm width.

Here is a sliced view of the block...so that you can see the base with the divets better:
http://cc.usu.edu/~roblb/Thin_slice_web.jpg

Here is a pic of the top plate:
http://cc.usu.edu/~roblb/Top_plate_web2.jpg

We made the block to fit both P4 and AMD XP chips.
So what do you guy's think?

That's a good design. You've got fins in the important places.

How do you propose to attach those fins?

we are milling the base plate out of 1/2" solid piece of copper. So we don't have to "attach" the fins. Just mill them out...

Do that if you want your fins:

http://membres.lycos.fr/roscool/forum/waterblock.jpg

More surface and less pressure drop because flow will be better guided into the block (no direct 180°!! killing flow) , a 6-7mm baseplate and it will be very good :)

Roscal, that is a good point. We had actually considered making the fins bend around the turns to help with flow rates. We might implement that yet...

If we did...should we make them all solid fins like the picture you posted...or leave some gaps in the straight-ways to help with turbulence?

BTW: What kind of performance did you get with that block you posted a pic of?

This pics was for a chipset block but a friend has realized the same (an evo of a previous block) but no testing..

He was the first to use this design but his block is better (barbs are different than you and no pressure drop for barbs because not 2 x 90° barbs). Have a look here : http://www.overclockers.com/tips247/

Personnaly, I won't make gaps into fins because it will destroy the flow without give anything better, your divets are good enough to improve thermal coefficient on the baseplate..(and flow is already turbulent (Re high with flow speed).