Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Aluminium/Copper Corrosion, does it stop?

5 views
Skip to first unread message

Anti-Spam

unread,
Nov 3, 2009, 6:11:42 AM11/3/09
to
I made a solar panel with copper tube risers and aluminium heat
collecting fins. The fins were formed with a semi circular slot (the
same diameter as the riser) running down the middle. and two of these
fins were clamped around each riser using pop rivets.
The panel works very well, but a white powder corrosion has formed
where the two metals meet. Since making the panel, I have read that
these two metals should not be used togther, but its too late.Does
anybody know what I can expect to happen to the panel, will this
corrosion eventually stabalize? The heat output from the panel does
not seem to be impaired.

Anti-Spam

unread,
Nov 3, 2009, 6:34:07 AM11/3/09
to
Sorry further to this question, what metal should I have used for the
Fins?
Do you recommend I take the panel apart and replace the Fins with
another metal?

Josepi

unread,
Nov 3, 2009, 7:26:01 AM11/3/09
to
See if you can insert some anti0oxide greese. Electrical people are familiar
with these products.

"Anti-Spam" <An...@Spam.com> wrote in message
news:m750f5p5n09ltr7at...@4ax.com...

dow

unread,
Nov 3, 2009, 2:34:06 PM11/3/09
to

Ideally, you should make the fins from copper, i.e. the same metal as
the tube risers are made from.

The white powder is aluminum oxide. The aluminum fins are being
corroded. The risers should not be affected. So, since you're going to
replace the fins anyway, you might as well wait until the performance
of the device diminishes significantly. It may continue to work
satisfactorily for quite a while.

dow

Anti-Spam

unread,
Nov 4, 2009, 4:23:48 AM11/4/09
to

Thanks for that, not sure I can get copper sheet here easily (also
sounds expensive), I can get tin probably, will that work reasonably
well, as I don't have a massive gap between risers, about 5 inches?

Solar Mike (NZ)

unread,
Nov 4, 2009, 2:34:56 PM11/4/09
to

> The panel works very well, but a white powder corrosion has formed
> where the two metals meet. Since making the panel, I have read that
> these two metals should not be used togther

I have constructed a number of such panels and have had no problems,
painted the copper risers and fins with an etch primer and allowed to
harden before clamping together. The primer provides isolation between
the two metals preventing the galvanic reaction.

Cheers
Mike (NZ)

dow

unread,
Nov 4, 2009, 2:45:50 PM11/4/09
to
> well, as I don't have a massive gap between risers, about 5 inches?- Hide quoted text -
>
> - Show quoted text -

Are you sure you can get real tin? Most of what is called "tin" is
actually steel thinly plated with tin.

Tin will also corrode, by galvanic action, if it is in direct contact
with copper in an environment where moisture might be present at the
junction. The only metals that won't corrode are copper itself, and
ones that are less active, chemically, than copper. These tend to be
expensive, like silver, gold, or platinum!

As others have suggested, you can stop the galvanic action by putting
an electrically-insulating layer between the metals.

dow

Josepi

unread,
Nov 4, 2009, 3:27:12 PM11/4/09
to
You may be able to stop the corrosion by insulating the return path for the
galvanic action from the takeoff piping and not bother with the contact
surfaces.

It takes a complete path to complete an electrical circuit. Breaking it
anywhere will stop the current.

"dow" <william...@gmail.com> wrote in message
news:862169cf-2d19-4779...@r27g2000vbp.googlegroups.com...

phorbin

unread,
Nov 5, 2009, 8:12:02 AM11/5/09
to
In article <eg30f5t8f62sj2jsv...@4ax.com>, An...@Spam.com
says...

...galvanic corrosion.

Someone has already suggested keeping moisture out... For that solution
to work well the joints should be dry first. Getting them dry could be
interesting.

I can't find my museum conservation textbook but...

My guess is that the aluminum is the anode and will break down. The
copper will probably remain intact.

Someone with more metallurgy than I will have to speculate on how much
the oxide film will insulate the connection.


help.http://www.dossert.com/technicalinfo/corrosion.htm

Anti-Spam

unread,
Nov 7, 2009, 3:37:59 AM11/7/09
to

That sounds like a very good idea, I will be making another one soon
for somebody else, so will employ that method, thanks.

In the meantime, all the while my one is relatively efficient (we
don't have to try so hard, down here in sunny spain), I think I will
just leave it to see what happens. Thanks to all responders.

Mark in Spain.

Jean Marc

unread,
Nov 9, 2009, 2:35:21 AM11/9/09
to

"Anti-Spam" <An...@Spam.com> a �crit dans le message de news:
4acaf55ar9s5vu17k...@4ax.com...

> In the meantime, all the while my one is relatively efficient (we
> don't have to try so hard, down here in sunny spain), I think I will
> just leave it to see what happens. Thanks to all responders.
>

It is not going to stabilize, and will continue 'till no contact occurs. You
may want to spray oil (or grease) at the joint to insulate electricaly and
slow the process.

HIH
JM


0 new messages