"m" wrote in message
...
You won't get a strong bond to aluminum just by cleaning it; but it may
be
strong enough. The mechanical bond you get from ordinary roughness of the
material actually is pretty good when the joint is loaded only in shear.
High-performance applications use PAA (phosphoric acid anodizing) to get
a
strong mechanical bond; PAA gets a tight grip on the parent metal and it
has
lots of surface roughness for mechanical bonding.

Thanks, this and other posts have been very helpful. I'll research
PAA. Is this a process that most anodizing shops can apply or is it
specialized?

I've never had it done, but you'll find plenty of shops that offer it. My
understanding is that it's quite common.

If you know who Ted Mooney is (an expert old timer I've come to respect; I
used to write the Finishing column for American Machinist, and I wrote a lot
about adhesive bonding to metal, but it was decades ago), you'll want to see
his brief comments about it:

http://www.finishing.com/4200-4399/4379.shtml

The comments in that thread go a little overboard for your application, but
they will give you some perspective. None of this stuff is new, BTW. I
researched and wrote about PAA roughly 30 years ago.


Per my prior post, I am now thinking that I might go for an anodized
finish rather than raw aluminum. This would have the anodizing shop
deal with post-machining/stamping cleanup. We could then simply wipe
the parts clean with alcohol and proceed to bonding. My guess is that
anodizing won't be seriously harmful to thermal conductivity. This is
a heatsink application for electronics.

I don't know about the insulating properties of standard thicknesses of PAA.
Anodized layers are basically aluminum oxide, which, at that crystalline
level, is considered to be a ceramic. But my vague recollection is that its
thermal conductivity is high relative to other ceramics.

I would be wary of it, if you need really good thermal conductivity. But the
layer of epoxy itself may be much more insulating, making the thermal
insulation of the anodized layer moot. It's not something I've studied.

PAA is very effective and reliable but it may be more than you need. Not
knowing the strength requirement or the specific loading on the joint, I
can't give you any specific advice about it. All I can tell you is that it's
the standard in industry for bonding to aluminum, especially in aircraft.

Note the comments about applying a primer within two hours. That's something
to watch for, but my recollection is that it's only relevant to very
high-performance applications. It may have something to do with the
hygroscopic property of anodizing. The chemistry of this is over my head,
but anodizing of many types, maybe all types, chemically absorbs and binds
water. It's hard to imagine that your bonding job is going to be very
sensitive to this, because the cogging bond is probably all you need.

--
Ed Huntress