Wow -- long discussion!
On the issue of the thermal efficiency of the original design with the
0.018 aluminum, did you see the test I did of this configuration
compared to a baseline all copper collector with soldered fins?
It seems to me that this is a pretty good test that should give valid
results, and it shows that the aluminum fin fitted around the copper
pipe as shown in the plans with silicone and a paint layer performs at
96% of the level of the all copper collector with soldered fins.
There does not seem to be a whole lot more to gain?
I'm a bit leary of using anything but silicone as the heat transfer
and bonding agent between the copper and aluminum. Silicone stands up
to 500F, bonds well, excludes water well for corrosion, separates the
aluminum from the copper for further corrosion protection, and is
flexible enough to withstand some thermal expansion differential. You
want this thing to still be working 20 years from now -- so choosing
materials is important.
There are some silicone caulks with fillers that are designed to
conduct heat better, but, again, how much of that missing 4% do you
really need to gain?
As Nick says, perfect is the enemy of good.
I think the key to a good thermal bond between the copper and aluminum
lies in forming a groove that fits well, and then really bearing down
when stapling -- you want any gap between the copper and aluminum to
be sub paper thin, and filled with silicone. This with the large heat
transfer area is (I think) what makes it work. To make the stapling
work, the backing board needs to be something like plywood that will
On the issue of fin thermal performance vs width and thickness, some
of the solar design books define a fin efficiency that compares a
given fin to a perfectly conductive fin. It depends on the width,
thickness, and conductivity of the fin material. I'll try to find
this and do the calc for 0.01 -- I'll post the number if I can find
it. I believe that I did it for the 0.018 material and it was OK.
I use a narrower sheet of aluminum under the copper tube that only
goes out about half way to the edge of the fin. You could just use a
full width one -- this gives you a sort of 0.02.
If the aluminum is bare, I would find a zinc chromate primer, and
paint it with that. I believe that some hardware stores carry
chromate containing primers, but if not, its definitely available on
online. Its not very expensive.
As a side note, I got an email from someone who wants to build the
collector with flattened and grooved aluminum pop cans -- they are
only 0.004 wall thickness, but when you flatten it doubles to 0.008,
and you could have a grooved can on top and a flat can on the bottom
-- since the flattened and grooved cans are only about 4 inches wide,
you might get a pretty efficient fin. I tried smashing and grooving a
sample Pepsi can and it takes a very nice groove. Not something
that is going to appeal to everyone, but a really interesting idea.
Thanks for the input,
At this point the plan is to use a layer off 0.010" aluminum on the
spray a thin coat of zinc chromate only where the pipe will touch, then lay
soldered copper pipe assembly on top of the aluminum. And put the formed
aluminum over the pipe.
I'll spray the formed aluminum where it will touch the copper.
I have put together a jig to form my aluminum, works good but I will
tweak it a little bit more.
I would like your input on the thermal flow of the 0.010" aluminum, which
I'm using two layers of. But then you almost did that with your 0.018"
Equation 2.42 on page 27 of the 1998 2nd edition of Schaum's Outline on
Heat Transfer resembles that remark: q = k A n thetab tanh(nL), where
q is heatflow, and k is the fin conductivity, and thetab is the difference
between the fin base temp and the surrounding fluid, and A = wt, and
w and t are width and thickness, and n = sqrt(2hbar/kt).