I have long been of the opinion that, if you have to rely on glue for
attaching something to aluminum, you should rethink your design.
I am changing my mind.
A little over a year ago, an artist friend of mine asked me to help her out.
She had been having photographs printed large (24"x32"), having them mounted
to a sheet of aluminum. and attaching a frame made of 1" sq alum tube to the
back to act as a stand off when hung on a wall.
The outfit she had been using in Milwaukee to do the work, got shut down on
an OSHA beef and she wanted me to do the final assembly.
She tells me she has that she has bought the special glue and dispenser that
they were using.
"Wait a minute!" I sez, "Aluminum and glue don't mix"
But somebody else has already done it and this isn't a life safety
application. So, I call an engineer at Lord Adhesives to ask him about
I ask him if I have to sand the metal, or etch, or use some special primer,
and he sez "No,".
"This stuff is specially designed to work with dirty metal. That's why it
works well on aluminum. You don't need any special surface prep. You don't
want chunks of dirt or dripping oil, but finger prints and light oxides are
not a problem."
Well it worked a treat.
Months later I have another aluminum job. This one needs me to rivet the
short edge of some 1/2"x 1.25" anodized alum glass channel to 6 ft long
1.75" anodized aluminum angles. The channels are attached perpendicular to
the angle and I used the adhesive to glue them temporarily to the angle
while I drilled the rivet holes.
There was only a 1/2"x1" glue surface on each one (2 per angle, 144 angles)
but despite rough handling not a single one got knocked off.
Last week I had another aluminum job.
A small museum without a lot of money, needs some stands made. After
talking the designer out of some grandiose ideas, I point to the aluminum
angle falloff from the other job, and say, "I can make you some out of that
The design we ended up with was a bunch of simple, flat, angle frames about
The corners were simple lap joints with a short piece of angle mounted in
the inside of the corners to act as a gusset.
I figured that I would glue it and rivet it like the other assembly. but,
when the glue was set, the joints were so strong that the rivets were a
waste if time.
I did some more research and this glue has a shear strength of 475 psi. That
means each glued corner has roughly the same strength as if it were put
together with 16 standard strength 1/8" solid alum rivets. I have a spare
frame here, that I have been sitting on and rocking back and forth. The
angle flexes, but no sign of glue failure.
Anyway, this stuff was Lord #406-19, a two part acrylic adhesive. It costs
about $13 bucks a tube and because of a 4-1 mix ratio you need a special
glue dispensing gun.
I am not affiliated with the company, just an impressed consumer.
Paul K. Dickman
I watched Modern Marvels last night. A program I had recorded sometime ago
about adhesives. They showed this really miraculous 3M adhesive tape for any
metal to metal or other adhesion that would knock any number of rivets out
of the ballpark. Something something 8 --- need to sharpen the brain.
Airliners are glued together, specifically wings.
Epoxy cement on clean aluminum has a tensile strength of 4,000 PSI.:
LePages 24 hr cure, hardware store, type stuff.
Would I trust my life on a home-baked glue joint? Probably
not....well, maybe if properly tested.
On Sat, 29 Sep 2007 23:36:18 -0500, with neither quill nor qualm,
Richard J Kinch quickly quoth:
I've been using the industrial acrylic adhesive-backed velcro for my
glare guard product and it's simply amazing stuff; ten times as sticky
as the old stuff. The expensive stuff is surely ten times more stable
than this, too. I'm sold on acrylic adhesives. But on plane shells?
Shiver me timbers!
Exercise ferments the humors, casts them into their proper channels,
throws off redundancies, and helps nature in those secret distributions,
without which the body cannot subsist in its vigor, nor the soul act
with cheerfulness. -- Joseph Addison, The Spectator, July 12, 1711
unless you are glueing in an inert atmosphere after removing all
surface oxide you are actually just gluing to the oxide layer.
how strong is that oxide bonded to the underlying aluminium?
They don't glue to the oxide. The standard treatment for gluing airplanes
together, since the late '70s, has been a pre-treatment of PAA, then a
high-strength epoxy, and rivets at the edges where there is any chance of
peel or cleavage loads ("rivet-bonding"). The rivets don't contribute
significantly to shear strength; they're just there to prevent the
initiation of peel. The epoxy is bonded to the anodizing.
There are some places where they use no rivets at all. This has been going
on for more than 25 years. The shear strength of the epoxies used in
aircraft bonding often run up around 8,000 psi or more.
FWIW, the Lotus Elise chassis is made of aluminum extrusions that are bonded
together with epoxy.
I wrote articles almost identical to those 25 years ago. IIRC, the L-1011
had a carbon-fiber tailfin, as well. And the other story is the same old,
same old for the high-performance composites business. Very little appears
to have changed.
A lot of people don't realize how much epoxy is used throughout the
structure of an airliner. They probably don't want to know. d8-)
However, they also don't know how much fatigue becomes a problem in
all-aluminum aircraft that were designed over the last few decades. DC3's
are still flying because the engineers didn't know what a reasonable safety
margin was. Now they know, and the life of those planes is finite.
I was told by late father, who was in the aerospace industry in the UK
and later US as a stress analyst, maybe 25 years ago that one of the
reasons the DC3 kept flying because you could still get fuselages and
wings. When the life of the wings was up you bought new ones and the
same with the fuselage. I expect like many planes the airframe life is
also re-evaluated after actual service conditions have been experienced
and extended or otherwise revised.
I think that's true, but I flew in DC3's in Canada's Northwest Territories
that still had fabric-covered control surfaces. Those were *old* DC3's, and
it was only 20 years ago that I flew in them.
They just stood up a lot longer. They also had severely reduced load
capacities because of the overbuilding.
And there is replacement in newer aircraft that extends their lives. Still,
everything is built closer to the limits in newer planes.
Yes, that's also true. But the big factor was that the safety margins for
aluminum in fatigue were not well established, and commercial planes were
I also read a story by an old-time pilot who said that the best plane to be
caught in during a storm was a DC-2, because it was so overbuilt, and its
wing loadings were so low, that it was ridiculous. It would stand up to
violent storms that would tear a DC-3 apart.
I'm sorry. I wasn't clear enough.
475psi wasn't the yield point. It was the UL rated strength for sign
applications, presumably derated with a suitable safety factor for loads
hung over the public way.
The yield point in lap shear is 2650 psi.
The peel strength is 25 pli.
Paul K. Dickman
I was building a homebuilt that had a number of aluminum fittings
bonded into the epoxy-glass structure. I had to sell it when my wife
found out what the engine was going to cost, but I did get to ride in
and fly another of the same design. As in any gluing operation,
cleanliness is next to godliness :-)
If fabric covered what was the structure of the control surfaces made
of?. I don't know much about wood structures so don't know if they
suffer from fatigue.
In the science museum in London there is a section of a 747 fuselage.
It's quite surprising how thin the outer shell is, looks to be about
2.5mm from memory. Not that I have a problem with that as with a
background in engineering I know some damn good people design these
things and the 747 is a strong aircraft judging from the bits that has
fallen off them and they still kept flying.