[ATM] glued aluminium mirror cell update

["Dream - Telescopes & Accessories, Inc." <shane@dreamscopes.com>]

Andreas,

Sorry to hear your epoxy bond is failing.

I don't see (on your web site/description) that you used sandpaper on the 
surfaces of the aluminium prior to bonding two pieces together with 
epoxy.  I highly recommend this as it sands down to raw, unoxidized 
aluminum, thus giving a better bond.  The rough surface obviously gives the 
epoxy much more to grab onto as well.  Oxidation however is typically the 
main issue with bonding metals.  They will oxidize immediately so you want 
to bond them as quickly as you can.  A vacuum chamber filled with nitrogen 
or... one trick is to do your final sanding "wet."   Use some epoxy instead 
of water though.  This prevents the newly exposed raw aluminum from 
oxidizing.  Well, it greatly reduces it.

The other thing I noticed on your site is that you cooked this epoxy in an 
oven.  I see two potential problems with this.  Can the epoxy take that 
kind of heat (356F!/180C)?  Normally off the shelf epoxies are meant for 
room temp cures and only have Tg's (Glass Transition points) of 
120-140F.  The high temp epoxy I use on a daily basis can be cooked to 350F 
but, and here's the other potential issue, it has to be cooked at a very 
specific schedule.  Initial ramp of only 5-7F/minute to 150F, then soak for 
1 hr.  Then up to 200 at the same 5-7F/m, then soak for 1 hr.  And so 
on.  Final cool down is the same, no faster than 5-7F/m.  If you cooked it 
in your home's cooking oven, controlling the ramp temp would be next to 
impossible, as ovens turn power on full to get up to the set temp as 
quickly as possible.

Did the epoxy say to cook it in the fashion that you did?  There are resins 
that I know of that are meant for more instant heating but I've never heard 
of an off the shelf epoxy that can take high temps.  Plus the high temp 
resins that I'm familiar with are very expensive, starting around $100/gallon.

My guess is that your problem lies somewhere within these three main issues:
- you overheated the epoxy.
- you heated it far too rapidly.
- the bond joint needs different preparation.

As far as CTE's, typically resins have far higher CTE's than even 
aluminum.  So if your bond was weak to begin with and temps swings did 
cause a great enough stress due to differential CTE between two materials, 
it was probably your resin shrinking more than the aluminum.

The bottom line though is that a decent epoxy should be able to withstand 
this sort of environment without issue.  37F is not really that 
cold.  Unless the telescope was sitting at 225F a few minutes prior...

West Systems epoxies have been used in the boating industry for 
decades.  These are used in environments similar to ATM use: cold, wet, 
dry, hot, you name it.  If you want flexibility in the resin, you can 
either use specific resins that have that property (although this doesn't 
sound good for an opto-mechanical application where flexure is trying to be 
minimized) or you can cure your resin in certain manners that helps produce 
more flexible epoxy.  I feel the issue is one of the three factors listed 
above though.  Spaced-based applications have been bonding aluminum to low 
CTE composites for decades and those see huge temp swings compared to most 
ground-based use, about +225-250F to -225-250F.  Some systems see these 
extremes every 90 minutes, +250F, then -250F...

Thank you.

Sincerely,
Shane Santi - Owner
Dream - Telescopes & Accessories, Inc.
http://www.dreamscopes.com
610 - 365 - 2833


Hey everyone,

I recently introduced my ATM 10" travel dobson (URL: www.duda-derwahl.de), 
with one of the features being a glued mirror cell made from aluminium 
square tubing (www.duda-derwahl.de/mirror_cell.htm).

After a couple of observing nights with cool, but not too cold 
tempereatures (plus 3 - 5 degrees celsius) I had another failure of two of 
the glued connections and I came to the decision to not pursue the glueing 
any further. Though I cannot exclude mechanical stress as cause of the 
failure, my guess is that the glue I used (epoxy brandname "Uhu Endfest 
300") is not flexible enough to acommodate for the thermal expansion of the 
aluminium.

In my opinion it does not neccessarily follow, that glueing cannot work - 
as far as I know, glueing aluminium is sucessfully used in automobile and 
window making industry. But the way I did it, is obviously not suitable to 
form a longterm stable connection.

There's still room for improvement, though. Either using a more suitable 
glue (more flexible or with matching thermal expansion), increasing the 
overlap area of the aluminium parts, or reducing the gap between the 
glueing surfaces.

But since the mirror is relatively lightweight (2.4 Kg) and I have some 
scrap birch plywood at hand, I will make a new cell from wood that 
hopefully won't have these problems.

Just thought I'd let you know in case someone want's to go along this track...

Andreas Derwahl

_______________________________________________
ATM mailing list http://www.atmlist.net/