tigers
[Top] [All Lists]

Re: Aluminum vs. Copper

To: Richard F Flynn <rflynn@dircon.co.uk>,
Subject: Re: Aluminum vs. Copper
From: Bob Palmer <rpalmer@ames.ucsd.edu>
Date: Sat, 18 Mar 2000 15:43:52 -0800
Richard,

Opinions are free but "rigorous data"? That you'll have to pay dearly for. 
;-) Actually, I could probably write several paragraphs on thermal 
conductivity and a comparison of aluminum versus copper and even throw in 
some data too, the bottom line of which is it doesn't make a nickel's worth 
of difference in a radiator. Before I say why this is so, I might point out 
that the purity of a metal has a very big influence on its thermal (and 
electrical) conductivity. Stainless steel, for example, is a very poor 
thermal conductor compared to pure iron (or pure nickel). I could explain 
this in terms of phonons and quantum mechanics, but you can get a feel for 
this classically by thinking about colliding billiard balls. When the 
billiard balls are the same size, they transfer momentum most efficiently. 
If you have a pure metal, all the nuclei (billiard balls) are the same size 
and it will conduct heat pretty well. If you alloy two very different size 
nuclei together, it really screws up the momentum exchange and the thermal 
conductivity goes to hell. The effect at very low temperatures can be 
several orders of magnitude; in fact, this is a pretty good way to measure 
purity. For completeness, I should also point out that in a typical metal, 
approximately half the heat is conducted by the conduction electrons, the 
same ones that carry electrical current. That's why metals are generally 
better heat conductors than insulators, although there are some very high 
thermal conductivity insulators like diamond and beryllium oxide, for 
example. (I'll resist the temptation to explain this phenomenon. ;-)

The reason the purity discussion is relevant to comparing copper to 
aluminum is that industrial copper is relatively pure, whereas most 
aluminum structural parts are highly alloyed for strength, which makes the 
thermal conductivity of typical aluminum alloys a lot less than copper, 
radiators included. (BTW, an interesting engineering fact about structural 
aluminum I learned a while back is its failure limit is zero; i.e., it will 
fail eventually, no matter how small the  applied stress is. Makes you 
think twice about flying in an old airplane, huh?)

 From the previous argument, it would seem that copper is the best material 
to build a radiator out of. However, the thermal conductivity of the 
radiator is scarcely a factor in its efficiency. You could even make one 
out of stainless steel and it would work just about as well as copper or 
aluminum. The biggest impediment to heat transfer in a radiator is the 
air/surface interface, and to a lesser extent, the coolant/surface 
interface. There is scarcely any temperature drop across the thickness of 
the cooling tubes. Other factors, like weight, are very important though. I 
had occasion to pick up an empty copper radiator for a Tiger, one of the 
really good ones which, incidentally, worked extremely well. It weighed a 
ton! I couldn't believe how heavy it was. I didn't actually weigh it (see, 
you have to pay extra for real numbers), but it must have easily weighed 
twice or three times as much as my aluminum radiator. If you don't mind an 
extra twenty pounds or so on the front of your Tiger, then copper has the 
practical advantage of being pretty easy to repair compared to aluminum.

I understand by private communication that there is a better aluminum 
radiator undergoing testing on a Tiger with very impressive results. It is 
not as thick as the Ron Davis radiator. The source is Fluidyne, who 
supplies radiators to NASCAR, etc. You can check them out at:
                 http://www.fluidyne.com/

Right now, a radiator for a Tiger would be special order, but if there is 
enough interest, maybe one of the clubs like CAT will have them make a 
limited production run. I'm sure it won't be available for this summer though.

Hope this amuses if not instructs.

Well, TTFN

Bob


At 05:34 PM 3/18/00 +0000, Richard F Flynn wrote:

>I hesitate to ask, but does anyone have and rigorous data on aluminum or 
>copper
>is better at dissipating heat?  I have been told by a HVAC type that copper is
>usually the metal of choice for cooling systems in buildings because it is 
>more
>efficient than aluminum, yet people on the list seem to have had improved 
>their
>cooling situations with aluminum radiators.
>
>Any hard data out there?
>
>thanks in advance,
>
>rick


<Prev in Thread] Current Thread [Next in Thread>