My CAD Waterblock.

[-J-]

this is my CAD waterblock, that i hope to build soon.

 

[-J-]

another1

 

[-J-]

im making it with Al instead of Cu, cause i cant get any nice heatcore or radiator with Cu tubing.

 

[-J-]

tell me what u think about the design, and if u would change something in particular.

 

[Albigger]

nice CAD, what program you using?

how wide are those channels, and how deep? And how thick is the base plate?

Design looks pretty good. With Alum. you need a lot of surface area.

if you are looking at the last pic you posted, is the inlet the left most barb or the right one?


so are you just milling completely through the aluminum middle block? And is the base colored differently in the pics because that is going to be copper?? Or is it just a color thing? because you would get corrosion problems if it is copper and the rest is aluminum, but you seem to know that already from your initial post....

 

[-J-]

the small channels are 2,5mm wide. and the bigger ones are twice that, 5mm wide.

they are 15mm deep.

the inlet would be the left one, so the fresh water goes first over the small channels that are right over de core.

the base of the alu would be 5mm, plus 5mm of the Cu plate.
The Alu and the Cu wont be in contact with water. i will use Artic Alumina (artic silver is way to expensive here).

i thought making the alu base thinner, but this WB design isn't only for 1 waterblock. im going to sell these things. so the Cu plate will be an option.

BTW, im using 3D Studio Max 4.2

 

[Albigger]

ok, i thought the inlet/outlet where they way you said, i just wanted to make sure.

15 mm seems deeper than necessary, and might add unnecessary weight to the waterblock/cold plate assembly. Not that's its not doable, i'm just not sure how far up the 15mm walls the heat is going to travel.

I see now that the Cu won't be in contact with the water, but having direct Cu and Al contact might still not be a good idea. I'm not sure though - so I'll let others chime in here.

I don't know if the copper would offer any advantage over just using the alum. block. with a 5mm base, it should be thick enough to use, even for pelts probably. But you would have to test it both ways - I don't know if just putting the 5mm copper under the waterblock would actually help at all....

 

[grim]

design is nice. maskedgeek made a block like that and i designed one that i never made that was based on a circle instead of a rectangle. the problem with designs like that is that they restrict flow alot because of all the sharp turns.

i dont think the copper plate on the bottom will help. i think a total base thickness of 10mm is way too thick. in many cases 5mm is too thick.

 

[-J-]

mmm will try with/without the Cu coldplate. and see if there's any performance increase.

another option would be, making the channels con the Al block deeper, so that the base thickness aint that big.

thanks 4 your sugestions.

albigger, i think that Alu-Cu together dont have a problem if there isn't any water with them. most aircoolers with Alu-Cu bases dont have any problem.

 

[Albigger]

albigger, i think that Alu-Cu together dont have a problem if there isn't any water with them. most aircoolers with Alu-Cu bases dont have any problem.

oh, duh....... good point!

yeah, i would try with and w/o the copper plate. also I think making the channels deeper into the aluminum so you have a thinner base is a good idea. but also maybe make the aluminum block shorter, so in the end you channels are still 15mm or less, but you have a thinner base.

i guess it doesn't really hurt to have the deep channels - just low velocity going throughout the block then.

 

[-J-]

making the channels deeper so that the base is thinner without the cu plate wont work, a friend of mine made this block, and got terrible temps. the base is only 1mm thick.

 

[JFettig]

thats weird,

ok on the first one, its pointless to add channels if they arent gonna be int he peice touching the cpu. so just make a flow directing channel or something, like you have it will only kill performance...

 

[-J-]

u say i shouldn't use the Cu plate??

 

[JFettig]

I say you should put the channels in the CU
the reason for the channels is for the heat to rize up into them, more surface area, otherwize you will be lacking surface area, someone at AOAforums made a very similar block.
the problem with this design is that these channels with all those large turns will restrict TONS of flow.

 

[-J-]

mmm so i should make it of Cu then. but i have the battery effect problem. nickel plating the Cu would help in some way??

 

[Paxmax]

mmm so i should make it of Cu then. but i have the battery effect problem. nickel plating the Cu would help in some way??

I had a nice mix of metals in my setup.
Brass nipples, steelpiped watercooler, copper nipples, tin solder, pyrolytic graphite block(ok, it doesn't count as metal...).

My tap water inside turned brownish within an hour...
I tried de-ionised water, still turned brown within a day...
Added waterwetter to de-ionised water, and Hey Presto! no more reaction.
I haven't seen any discoloration now for the past 11 months.

Nickel plating would be another barrier between water/waterblock and will make it a bit less performing.

Nice lookin' block though, I would make it a bitt less fancy with straight channels running in parallell. The analogy is parallell = hottest water possible all over the block = better performance. Also straight channels wouldn't impede flow as much as thoose beautiful twists'n'turns.
/Paxmax

/Paxmax

 

[-J-]

i can't get watter wetter here. whats that exactly?? i mean what it used for, its an aditive for the water in the radiator of cars or what??

 

[SemiCycle]

Check this out. It will tell you about water wetter and also chemicals that may be more readily available to you.

http://www.procooling.com/articles/html/chems_of_oc_n_and_super_coolin1.shtml

 

[-J-]

thanks

 

[Rakk]

What radiator are you using?