a Design for a Socket A WB & i need ur help

[safrout]

i have this simple design for a maze WB which i will use on a Socket A proc

the problem is that i don't know what is the dimension of the WB
i am talking here about the base plate since the upper plate must be big enough to have the holes for mountings on the 4 holes in the mobo

so this WB i think it should be the same dimesions of the proc itself to rest on it but what i don't know the height of the block itself and the thickness of the base on the proc
can you suggest some numbers please

all the numbers are in mm by the way

[jaydee]

If you want the holes all the way through the top and base then you need a peice 2" (50.8mm) wide and 3" (76.2mm) long. I would go .5" (12.7mm) thick with a base thickness of .1875" (4.7625mm). In other words the channel depth would be .3125" (7.9375 millimeter) deep.

[safrout]

i think i'll keep the holes in the top only not the base
so i think i need the base to be 50.8x50.8x12mm with base thickness = 4~5 mm
what about the design is it fine or i can imrpove it alittle

[jaydee]

Originally Posted by safrout

i think i'll keep the holes in the top only not the base
so i think i need the base to be 50.8x50.8x12mm with base thickness = 4~5 mm
what about the design is it fine or i can imrpove it alittle

Depends on your methods of making it. That is an old school design that is not high performance. Easiest improvment would be the inlet in the middle. Like this except for the CPU. http://forums.procooling.com/vbb/sho...&postcount=152

[safrout]

very nice and it is easy for me to change the design to this one
thanks alot

[safrout]

but the user himself in another threads says that its performance is not good and that he want to change it

Originally Posted by Odin

Hi guys, i have a simple watercooling system and my temperatures are not really good....

I have access to a milling machine like this one: http://img.photobucket.com/albums/v285/escher/WC01.jpg

wich kind of blocks i can make using this machine? where i can find some nice projects of cpu and gpu blocks?

my block:

http://img.photobucket.com/albums/v...locknaplaca.jpg
http://img.photobucket.com/albums/v...to/PICT0866.jpg
http://img.photobucket.com/albums/v...to/PICT0818.jpg
http://img.photobucket.com/albums/v...to/PICT0801.jpg

it must be there a simple yet effective design, but if i won't be able to find another design i'll stick with Odin's

[jaydee]

Originally Posted by safrout

but the user himself in another threads says that its performance is not good and that he want to change it



it must be there a simple yet effective design, but if i won't be able to find another design i'll stick with Odin's

What tools do you have to work with?

[safrout]

not sure yet but atleast something like this ( /me won' be the one who will do it )

[jaydee]

If they got a mill then I would suggest a pin grid block.

[safrout]

pin grid ???
can you explain please


and sure thanks for your help & time

[BalefireX]

Jaydee is far too modest. Take a look at one of his DIY blocks that performs quite competitively with the mass produced blocks here:
http://www.procooling.com/reviews/ht...rblock_rev.php

You should be able to make something like that with basic equipment if you have the patience to drill a lot of holes

[safrout]

ya and i am searching now for a CAd diagram for the spec of one of his blocks
anyone have a link or the file please send it to me

[jaydee]

Originally Posted by safrout

ya and i am searching now for a CAd diagram for the spec of one of his blocks
anyone have a link or the file please send it to me

Shouldn't need a drawing. The pins are 1/8" and the channels around them are 1/16". I would post a drawing but I don't use complete drawing to make these blocks. I just use took path drawing which you wouldn't know what your looking at unless you used my specific CNC mill.

All you have to do to make a pin grid block is mill strait lines.

[safrout]

ok thanks for the info and sorry for bothering

[jaydee]

Here is a basic Pin Grid block I just drew up that should be easily done on a manual mill and give decent performance. Can use this a basis for whatever material you have to work with.

[bigben2k]

So JD, you're finally getting handy with Solidworks?

[jaydee]

Originally Posted by bigben2k

So JD, you're finally getting handy with Solidworks?

Have to use it now and then to maintain what little I already know about it.

[safrout]

ok thans alot Jaydee for your help

but i heard somewhere that using Al & Cu together is not a good idea cause being together with water will lead to the corrosion of the Al

is this right or not ?

and for sure the Rad will be made using Cu

[jaydee]

Originally Posted by safrout

ok thans alot Jaydee for your help

but i heard somewhere that using Al & Cu together is not a good idea cause being together with water will lead to the corrosion of the Al

is this right or not ?

and for sure the Rad will be made using Cu

Aluminum? No, don't use aluminum. If your getting that idea from the drawing then grey is the default color in Solid Works. It dosn't represent the material. My laoptop has a crappy vid card in it so it dosn't render all that well. Here it is in copper color anyway.

[safrout]

LOL
no i got it from another member here who used AL to do te top plate

anyway after some reading i know now that brass is the best for the top Plate is " brass "
but i am not sure that i'll be able to find it here in Egypt, so in this case i'll use Cu also for the top

wish me luck and soon i'll be back to post pictures

Thanks alot for everyone who helped me here

[jaydee]

Originally Posted by safrout

LOL
no i got it from another member here who used AL to do te top plate

anyway after some reading i know now that brass is the best for the top Plate is " brass "
but i am not sure that i'll be able to find it here in Egypt, so in this case i'll use Cu also for the top

wish me luck and soon i'll be back to post pictures

Thanks alot for everyone who helped me here

Brass isn't the "best", just easier to mill than copper. Copper is actually cheaper.