Machining- CNC help needed for scratch build passive water cooling system.

[mOs_cOww]

Hey Carbonite,

I am starting a new project with the goal being a passive or semi-passively water cooled pc with radiators and heatsinks with an external pump in a sealed reservoir enclosed in a sound proof/dampened housing.

I have these large heat-sinks that I with to use to cool water applied directly to them through a machined block or perspex.

I have made a crude sketchup model to try illustrate my point more clearly.

What I need is the help of anyone who has access to an accurate cnc machine or mill. Persepex being easy to machine and the size of the piece should make for a quick job.

I have never done anything like this prior and regarding the 3d model I am sure I can make an accurate model to input directly into the cnc machine. I have sketchup, would this be able to make an accurate enough model? From what I've seen its a bit crazy, unless my methodology is not the best.

This is really just for fun however realistically I think I can reduce the overall noise to virtually nothing whilst keeping the temps low.

Here are some Images of the heatsinks.

If anyone is interested or wishes to help out or make suggestions please don't hesitate to post on here or drop me a pm.

Thanks

 

[mOs_cOww]

View attachment 18962View attachment 18963View attachment 18964View attachment 18966View attachment 18967

 

[el3venth]

[MENTION=25979]mOs_cOww[/MENTION] Those fins look like active cooling fins. They look too closely spaced together for passive cooling.

 

[mOs_cOww]

Hey [MENTION=145]eleventh[/MENTION] thanks for the comment. They do look like active sinks however they came off these 3 phase ac motor controller things and they had no fans, just passive.

Realistically I will most likely use a fan to pass a small amount of air over the fins. On my new mobo I can reduce the speed of my coolermaster fans to around 400rpm which passing over the fins is dead quiet. Even the smallest amount of airflow from an active source dramatically increases heat dissipation versus completely passive.

I will still most likely be using rads however I think it is possible to adequately cool the water enough requiring only a very small amount of air to pass through the rads to cool my cpu and gfx card.

 

[Deadmeat]

send me your drawings and will help you out, if i can
will pm you my email addy

 

[mOs_cOww]

View attachment 18973View attachment 18974View attachment 18975View attachment 18976View attachment 18977

These are the sketchup mockup of the rough idea I have in mind. This will fit onto the large heatsink.

 

[mOs_cOww]

View attachment 18979

send me your drawings and will help you out, if i can
will pm you my email addy

Hey man thanks alot, I was uploading the images when you posted I hope you can get a rough Idea of what I have in mind, the model I made is not very good. Never tried to do tubing like that in sketchup, there must be a much simpler and better way of doing it.

 

[RavenShield]

Must say i quite like the idea, looking forward to see more pics of this build!

 

[Off-The-Chart]

View attachment 18979

Hey man thanks alot, I was uploading the images when you posted I hope you can get a rough Idea of what I have in mind, the model I made is not very good. Never tried to do tubing like that in sketchup, there must be a much simpler and better way of doing it.

if that is what I think it is, then I have seen the IN and OUT be on the bottom/top because not sure that angled hole would be easy to make.

 

[mOs_cOww]

Hey guys thanks for the interest, it looks like its going to happen

Deadmeat and I are going to make it hopefully during this december. I am going to meat him sometime to discuss what we will do and come up with a good solution.

[MENTION=63]Deadmeat[/MENTION] Do you think it will it be easy to plug the threaded holes on the base of the heatsinks? Im not exactly sure how to go about something like that. They will obviously have to be completely watertight and also be able to get machined into without coming loose. Would there be a way to fuse them in there with maybe a slightly larger screw that is very cold and screw it into the heated heatsink so they fuse together nicely?

[MENTION=8442]Off-The-Chart[/MENTION] as far as the inlet holes are concerned I will need to discuss what is possible with Deadmeat as he is the cnc man and will be able to tell me what is possible. I am looking for a smooth transition as possible from the 1/2" inlets as possible to reduce turbulence. I am thinking of making the water channels 1/2" as well so there is no necking in so hopefully this will decrease overall pressure in the system.

I think smaller diameter water channels will equate to more time in contact with the surface of the heatsink however it will make more pressure and slow the water down. If any of you have any thoughts on the size and shape of the water channels please let me know.

I was thinking maybe flat rectangular channels would work better instead of domed or half tube channels like in my 3d model. The reason being there will be less water in the channel at any one time but there will be overall a greater surface area, with the downside being increased pressure and possible very turbulent.

 

[mOs_cOww]

Hey guys,

Here are 2 more in depth designs.(the in and out holes are purely for demonstration, not sure how we going do it get)

The first is a flat channel machined into the poly carbonate and no machining on the aluminium.

The second design both the perspex and aluminium are machined into tubular channels maximizing water volume and surface area.

Would the first design be more efficient if it featured the negative machined into the aluminium to increase the surface area or will the second design consisting of the tubular channel be more efficient?

My thinking is the first design with the flat channels will have the greatest surface area in contact with the water versus a semicircular channel which will have a greater volume of water but possible less contact with the aluminium.

I am not sure if that thinking is correct. If any of you have any input or ideas Id love to hear them.
View attachment 19151View attachment 19152View attachment 19153View attachment 19154View attachment 19155

 

[Off-The-Chart]

I would say stick to your 1st design's IN and OUT mounting.

as for which is more efficient, tbh in personal opinion I would have to say I wonder if either method would really make such a significant difference in temps, considering the water will be in constant flow.
maybe if it was actually going to be laying still (this is all just personal ramblings ofc as it would actually be interesting to see if it does indeed make huge difference)

would go with the machining of both parts though and just remember to take into account your o-ring around the edges.

again, if you have a look at some of the custom blocks guys make, it comes out most of the time with just basic flat surface machining.