Gluing Aluminium

I have tried gluing / bonding aluminium in the past, with mixed results.
I use an epoxy resin, of the Araldite type, clean & abraid the area (to
remove oxide), apply resin, leave to set.
Sometimes it works fine, others it fails. The joints aren't stressed, they tend to be small boxes etc. Often as not, the resin is more to fill a gap/seal than 'hold together'.
There seems to be no obvious reason for the times it doesn't work.
I need to join/seal some very thin aluminium (paper type thickness). Again, no real stress, but a seal is required and the join will be exposed to mild heat ( you could touch it but probably wouldn't want to).
I've found some resin, JB weld, which is rated for high temperature but how do I improve the likelihood of the joint 'taking' in the first place. The joint will be close fitting, as good as clamped until the resin is set.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

Well this at least I have experience with. Clean the aluminum first until water will wet the surface completely. Clean a stainless steel wire brush the same way. Then when the brush and aluminum are absolutely dry apply the epoxy. After the aluminum is covered with the epoxy wire brush the surface by putting the brush through the epoxy. This removes the oxide layer which prevents good bonds. This layer is quite thin and the little particles that get mixed into the epoxy won't have much of a weakening effect. Every time I have used this method I got excellent bonds. Eric
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Thank you.
I think I can see the logic / chemistry behind this. The scrubbing through the epoxy excludes oxygen and prevents the oxide layer reforming. I was aware this can happen quickly.
I will order a suitable brush and give it a try.
Of course, the time between cleaning and reforming of oxide in my previous tries could well be the unexplained factor in my variable results. Not only could it simply depend on how soon I applied the epoxy, temperature would probably play a part.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
"Brian Reay" wrote in message

Another factor as well as the oxide layer is thermal expansion of the aluminium which puts huge shear forces on the epoxy/aluminium joint. I've no idea what the thermal expansion of Aradlite is, but you'd be very lucky if it's the same as aluminium. In the past I've very successfully used Devcon aluminium filled putty as a bond for aluminium sections, and it has the advantage of being able to be filed, drilled and tapped. If rubbed down with fine steel wool it gives the appearance of the parent metal.
Andrew
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Good points.
Previous applications the temperature hasn't been so much of an issue, although there would always be some change even if just in the ambient temperature.
In the new one, there is a heat source.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 04/02/16 13:42, Brian Reay wrote:

If you can apply the heat without distorting the aluminium (enameling oven, pottery kiln), aluminium solder should do the job.
--
Rusty Hinge
To err is human. To really foul things up requires a computer and the BOFH.
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Wed, 03 Feb 2016 13:57:43 -0800, snipped-for-privacy@whidbey.com wrote:

In boat building, I've read cautions against too much clamping pressure when using epoxy. Apparently, you can squeeze it out to the point of failure. Don't know how that'd apply to aluminum.
Pete Keillor
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

It certainly sounds reasonable.
When I worked in industry (avionics and other military kit), several projects used chassis/boxes which were Redux bonded. I wasn't on the mechanical side of things but, as I understand it, this involved apply a special tape to the joints which were also screwed/bolted. The assembly was then baked and the tape 'glued' the joint. The screws remained but were really just to hold the assembly will the bonding took place. I may have the details wrong put those are the basics. It is cheaper than a casting in shortish production runs. The assemblies were generally cubiods, made of alloy, with hollow sides which contained cooling fins over which air was blown in some applications. The joints were narrow for EMI reasons and the screws also provided electrical bonding. I'm not sure if the tape was 'metal loaded' to make it conductive.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 04/02/16 12:24, Pete Keillor wrote: [...]

You can - more so than with eg wood.

https://www.youtube.com/watch?v
Mkc63EpMQ
Epoxy may not be the best choice for aluminium, as it is very rigid when set. Try a polyurethane.
Also, epoxy doesn't handle heat at all well, it softens at fairly low temperatures.
This applies to so-called "high temperature" epoxies as well as ordinary epoxies - they can give a few more degrees, but any epoxy you can afford is weakened at 100C, and basically useless at temps above 150C or so.
For better heat resistance, try a high temperature "red" silicone adhesive.
(note, adhesive, not gasket sealant, which doesn't bond well).
someone was talking about youtube resources, I was reminded of this:

https://www.youtube.com/watch?v=fK9m5PadmiI&list=PLSnvVtM4lBIVB8APbcRTuUSVrnMhkdOua

-- Peter Fairbrother
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Thu, 04 Feb 2016 16:58:28 +0000, Peter Fairbrother

I used to make rails for circuit boards so the assembly could be slid into an aluminum tube. The rails were aluminum and urethane was bonded to the rails. I made a mold for the rails and the molder just poured the uncured urethane compound in. That urethane stuck VERY well to the aluminum, the only way to remove it was by soaking in hot chlorinated solvent. Eric
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 03/02/16 21:27, Brian Reay wrote:

The oxide will reform before you have time to think about blinking. That is, unless you are working in an inert gas cabinet.
My guess as to why it sometimes works, but doesn't at others, is that there are so many grades of 'aluminium', Dural, Noral, and other alloys.
--
Rusty Hinge
To err is human. To really foul things up requires a computer and the BOFH.
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Tue, 5 Apr 2016 15:16:27 +0100, RustyHinge

Not a problem when you abrade the area _through_ a layer of epoxy that's already been applied.
--
Mark Rand
RTFM

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.