Posted by Solar Mike (NZ) on February 24, 2009, 8:14 pm
The rate that heat is moved to the column of water flowing through the
riser pipes is some what proportional to the thickness of the aluminum
0.010 or 0.25mm fins will require a inter pipe spacing of 100mm or 4 "
to get effective transfer; this means lots of riser pipes and
associated welded connections to the headers.
Going to 0.5mm metal allows a riser spacing of 150mm with good
efficiency and less copper.
If you have lots of cheap 0.010 metal then sandwich the risers between
2 pieces pop riveted together, this will make a good thermal bond
between the fins and copper risers, put some heatsink
compound, eg high temp graphite grease works well to fill any air
voids along the contact line. Spray paint the copper with thin coat of
etch primer first, this helps prevent dissimilar metal electrolysis
over the life of the panels (20 years +)
Mike Scaife (NZ)
Posted by amdx on February 24, 2009, 11:35 pm
>The rate that heat is moved to the column of water flowing through the
I appreciate you looking into this.
Hmm! the original designer used 0.018" alum and 6" inter pipe spacing,
so by your numbers he got it just about right with the minimum materials.
I have contemplated putting aluminum above and below the copper pipe, That
give me the .020" (0.5mm) you mention.
High temp graphite grease? That's a thermal compound I haven't seen
Do you have any thermal comparisons: air, silicone, thermal mastics, etc.
Ok, I hear you about using the etch primer to prevent electrolysis, I am
how the primer would effect the thermal bond between the alum/copper?
Posted by Solar Mike (NZ) on February 25, 2009, 8:41 am
In this application provided you press a groove in the aluminum sheet
to give a large contact surface area when clamped to the risers the
thermal compound doesn't have to be hugely conductive, all its doing
is filling the small air voids and anything will conduct heat better
than air. There are specialized compounds that contain conductive
materials eg as used on heatsinks in the electronics industry but when
you are wanting perhaps a liter of it to do your panels its just too
expensive. Silicon rubber mastic works well also, although it doesn't
conduct heat very well, in this instance the film thickness is so thin
when clamped between the fins and risers that a reasonable heat
transfer occurs; although it sets quite quickly so you have to get the
whole deal clamped together before this occurs.
I mentioned high temp graphite grease as its cost effective, readily
available and like the silicon doesn't set hard thus allowing some
expansion movement to occur.
If there isnt a groove in the fins then something very conductive must
be used as the surfaces in contact are very small, I would suggest
this is not a good way to go.
Thin coat of etch spray primer from a can is 0.0002 inches thick, will
have no measurable effect on the heat transfer.
Best of luck.
Posted by barefoot on February 25, 2009, 8:57 pm
The (linked) design at
relies on pressing the aluminium around the pipe to achieve a large
contact area. Thickness of the aluminium shouldn't matter much as heat
is received through the top face and lost through the bottom where it
contacts the pipe. The aluminium will have a gradient ie it will be
more or less at water temperature where it is in contact with each
pipe and hottest halfway between the pipes. The thickness of the
aluminium will effect this final temperature. Decrease the spacing
between pipes for more efficiency. The effect of corrosion between
dissimilar metals will be very important - the original design showed
a powdercoated (enamelled?) aluminium which would resist corrosion
possibly to the detriment of heat transfer?). I've never bought it
myself but heat transfer paste is available in tubes for use in a
"silicon" gun. Try and find a refrigeration display case company that
makes butchers coldpates with pipes under a stainless steel bed as
they will use it. Stainless has quite poor conductivity.
Posted by Joesepi on February 27, 2009, 3:35 am
At 120 dgerees C how much grease conductor will remain and not drip out the
bottom of the troughs? Welding or soldering would be a trick. I thought I
would try it but 300 feet of tubing soldering convinced me otherwise. I
couldn't afford the 100 pounds of lead...LOL