Passive cooling might be an overstatement... or not. That'll depend on how this turns out.
My idea is to "DIY" (fabricate, not fabricobble) a heatsink for both my GPU and CPU with a very specific layout to essentially double the cooling capacity - also known as half the airflow needed - aka lower the noise. Of course, if this turns out to be effective enough, I will certainly run the setup passively.
Why you might ask?
Well, I've tried my hand with water cooling, and although it under the right circumstances removes the need for fan noise, it introduces a lot of not needed failure points and another component of noise - the pump. Therefore I've come to the conclusion that doing a custom heatsink solution and using heatpipes to move the heat instead of water might be superior in my case.
The concept as of now looks something like this (keep in mind that it will be a prototype and no final solution, and that the final solution if this works out will be custom built everything including the case):
Keep in mind that something like this will also be the case on the CPU side.
The gameplan as it looks now goes something along the lines of:
I'd appreciate feedback - also tips and experience if any of you have tried similar.
Best regards,
rosinbole.
many edits: gosh darn photo not displaying properly
My idea is to "DIY" (fabricate, not fabricobble) a heatsink for both my GPU and CPU with a very specific layout to essentially double the cooling capacity - also known as half the airflow needed - aka lower the noise. Of course, if this turns out to be effective enough, I will certainly run the setup passively.
Why you might ask?
Well, I've tried my hand with water cooling, and although it under the right circumstances removes the need for fan noise, it introduces a lot of not needed failure points and another component of noise - the pump. Therefore I've come to the conclusion that doing a custom heatsink solution and using heatpipes to move the heat instead of water might be superior in my case.
The concept as of now looks something like this (keep in mind that it will be a prototype and no final solution, and that the final solution if this works out will be custom built everything including the case):
Keep in mind that something like this will also be the case on the CPU side.
The gameplan as it looks now goes something along the lines of:
- Get sample heatpipe.
- Test efficiency in non-vertical orientations.
- Test bending ability and efficiency after bending.
- Create CAD model for prototype.
- Get raw materials for prototype.
- Construct protoype.
- Might be a build log in itself, only time will tell.
- Conduct testing on prototype.
- Evaluate if this concept will work or not.
- and share my new found knowledge with all of you of course
I'd appreciate feedback - also tips and experience if any of you have tried similar.
Best regards,
rosinbole.
many edits: gosh darn photo not displaying properly
Last edited:
