bobkoure wrote:Why aluminum? You do realize that you'll need to run corrosion inhibitor in your coolant even if you have only aluminum contacting it?
Or is this a weight issue? Looking at alu-to-air heat transfer efficiency? As I remember, coolant-to-cu is enough better to compensate, but I can't remember the specifics - and am too lazy to look it up for you.
Just wondering - it's your PC, you do what you like...
Both aluminium and copper will oxidise in water (unless it's perfectly non-ionic distilled water with no dissolved oxygen).
Aluminium oxide is a white substance that gives aluminium its distinctive milky grey appearance. Most of the naked aluminium that you see knocking around is oxidized. If you scratch some, you'll see that it is a bright shiny metal, and that the dull milky sheen returns rapidly.
Aluminium oxidises very quickly in the presence of free oxygen, but it generally only oxidises to a depth of about one aluminium oxide molecule - a layer that then effectively protects it from further oxidation (barring the presence of powerful oxidising agents or ionic/galvanic potential).
Copper, too, will oxidise - with a greenish layer sometimes called 'verdigris' - but will then be protected by its own oxide layer: just check out some of the old green copper domes and lighting conductors on some elderly churches and similar buildings.
These very thin layers of oxide will not vastly compromise a water (or air) cooling system.
In a watercooling scenario, of course, corrosion is a serious issue in the event of mixed metals. If I remember rightly, there's a smashing article on this phenomenon, galvanic corrosion,
hereabouts somewhere.
I've not done much research for the following further than my own head, so bear with me if it's utter tripe:
- Copper is a better conductor of heat than aluminium - about twice as good.
- Copper costs on average three times as much as aluminium, as a material out of which to manufacture a heat sink.
- Copper is three times a dense as aluminium, which has a mechanical impact in designing a water-block (it'll be heavier).
(some references
here)
I've heard it suggested that copper's higher specific heat capacity means that that aluminium is better suited at transferring its heat into the cooling medium. That is to say that while copper will conduct heat away from a body-in-contact more rapidly than aluminium, it will retain that heat for longer (which makes it great for bed-pans). So far, my research on this particular aspect of the discussion has revealed lots of noise and very little signal.
Hope that helps