Let's be fair, it may not be the glue. It could be the user - me.

he items glued together are the ends of the arm of a cast alluminium patio chair. One of six. The arm is bolted at the bottom to the seat and at the top to the chairback.

The break came at the point where a bolt is screwed through the arm and into the chairback. It simply snapped.

This first happend in the summer. When I first tried to locate the sections, I realised that the joint had been under stress. So after gluing wit superglue, I filed the hole to allow the bolt free movemnt.

I thought that would be the end of it but a couple of days ago, the first when we coud sit with up to six people, it snapped again at the same point. I forgot that I had used SuperglueIt and because it looked clean glued it agin wuth two-part gorilla glue (a bit like Araldite). It was no good and came apatrt again as I bolted it in.

On inspection it hada sticky feel and instead of glinting, the brocken areas are a dull grey. I tried cleaning them with petrol. I am not sure if I was successful.
So before another attempt, I thought I had better garner the benefit of experience.

TIA, Alan

--
Mint 20.04, kernel 5.4.0-42-generic, Cinnamon 4.6.7
running on an AMD Phenom II X4 Black edition processor with 8GB of DRAM.

On Fri, 2 Apr 2021 17:10:53 +0100, pinnerite
wrote:

snip

So before another attempt, I thought I had better garner the benefit of experience.

I think it might have been suggested the first time that the chances
are, something like that would need something in the way or
reinforcing, given that it broke before being further weakened (by
breaking)?

So, is there any way you could brace the joint, even if only on one
side with something suitable (steel strip / sheet / rod etc)?

Cheers, T i m

On Fri, 02 Apr 2021 17:32:34 +0100
T i m wrote:

On Fri, 2 Apr 2021 17:10:53 +0100, pinnerite
wrote:

snip

So before another attempt, I thought I had better garner the benefit of experience.

I think it might have been suggested the first time that the chances
are, something like that would need something in the way or
reinforcing, given that it broke before being further weakened (by
breaking)?

So, is there any way you could brace the joint, even if only on one
side with something suitable (steel strip / sheet / rod etc)?

Cheers, T i m



Sadly the shape and size prohibit this. I am going to try some alterative solvents to remove the remaining glue before trying Araldite. I will try to use a clamp.

Thanks, Alan



--
Mint 20.04, kernel 5.4.0-42-generic, Cinnamon 4.6.7
running on an AMD Phenom II X4 Black edition processor with 8GB of DRAM.

On Fri, 2 Apr 2021 18:41:54 +0100, pinnerite
wrote:

snip

So, is there any way you could brace the joint, even if only on one
side with something suitable (steel strip / sheet / rod etc)?


Sadly the shape and size prohibit this.

What about drilling and pegging each end of the joint? It wouldn't
have to be *that* accurate if you have enough 'meat' to drill into and
use a suitable epoxy.

I am going to try some alterative solvents to remove the remaining glue before trying Araldite.

As (chemically) clean as possible is a good start, as might be a
*little* warmth.

Whilst Araldite used to be the go-to adhesive for many things (esp the
24 hr stuff if it really needs to be strong), personally I would first
try something like JB Weld two part epoxy on the grounds that if that
won't hold it, not much else will.

Std 24 hour Araldite is just a good two part resin, the JB Weld is
reinforced, providing a stronger bond as it's less likely to compress
on the side with compression loads and so tear away the rest of the
bond.

I will try to use a clamp.

Anything that will stop it moving whilst it cures is essential.

Once cured, some of these adhesives also enjoy further curing with
some external heat (but don't go mad). ;-)

Cheers, T i m

On 02/04/2021 17:32, T i m wrote:
On Fri, 2 Apr 2021 17:10:53 +0100, pinnerite
wrote:

snip

So before another attempt, I thought I had better garner the benefit of experience.

Unreasonable expectations of what a glue can actually do.

I doubt if there is any glue that could make a joint that will survive
the load that this will be exposed to short slab casting epoxy onto it.
Reinforced carbon fibre composite might just be able to hack it.

I think it might have been suggested the first time that the chances
are, something like that would need something in the way or
reinforcing, given that it broke before being further weakened (by
breaking)?

So, is there any way you could brace the joint, even if only on one
side with something suitable (steel strip / sheet / rod etc)?

Steel and aluminium do not make good friends leading to corrosion. The
repair is best done in aluminium but will cost more than it is worth.

--
Regards,
Martin Brown

On Fri, 2 Apr 2021 20:15:30 +0100, Martin Brown
wrote:

On 02/04/2021 17:32, T i m wrote:
On Fri, 2 Apr 2021 17:10:53 +0100, pinnerite
wrote:

snip

So before another attempt, I thought I had better garner the benefit of experience.

Unreasonable expectations of what a glue can actually do.

+1

I doubt if there is any glue that could make a joint that will survive
the load that this will be exposed to short slab casting epoxy onto it.
Reinforced carbon fibre composite might just be able to hack it.

If it could be wrapped with some woven carbon fibre and resin, that
would probably do it?

I think it might have been suggested the first time that the chances
are, something like that would need something in the way or
reinforcing, given that it broke before being further weakened (by
breaking)?

So, is there any way you could brace the joint, even if only on one
side with something suitable (steel strip / sheet / rod etc)?

Steel and aluminium do not make good friends leading to corrosion.

True, but 1) if embedded in epoxy they probably wouldn't be in direct
contact and 2) unlikely to react unless they got wet (that they
shouldn't if the brace is encapsulated)?

The
repair is best done in aluminium but will cost more than it is worth.

Not if they used ally sheet / strip / rod but that would likely be
more expensive and weaker than steel.

A 6" nail cut to length and drilled and epoxied into each side of the
joint as a dowel might be worth a go, given the CSA of the bench
section and diameter of the nail. ;-)

Cheers, T i m

Its the oxide. If it was not for that coating the metal would simply go up
in a flash and a bang, trouble is its horrendously hard to make stuff stick
as the oxide simply comes away. I seem to recall some kind of etching
adhesive could glue it, but the best fix though it will look bad is
mechanically bolting plates onto the joint I think.
Brian

--

This newsgroup posting comes to you directly from...
The Sofa of Brian Gaff...

Blind user, so no pictures please
Note this Signature is meaningless.!
"pinnerite" wrote in message
news:20210402171053.d019320f5e4e16594ebc2e2a@gmail .com...
Let's be fair, it may not be the glue. It could be the user - me.

he items glued together are the ends of the arm of a cast alluminium patio
chair. One of six. The arm is bolted at the bottom to the seat and at the
top to the chairback.

The break came at the point where a bolt is screwed through the arm and
into the chairback. It simply snapped.

This first happend in the summer. When I first tried to locate the
sections, I realised that the joint had been under stress. So after gluing
wit superglue, I filed the hole to allow the bolt free movemnt.

I thought that would be the end of it but a couple of days ago, the first
when we coud sit with up to six people, it snapped again at the same
point. I forgot that I had used SuperglueIt and because it looked clean
glued it agin wuth two-part gorilla glue (a bit like Araldite). It was no
good and came apatrt again as I bolted it in.

On inspection it hada sticky feel and instead of glinting, the brocken
areas are a dull grey. I tried cleaning them with petrol. I am not sure if
I was successful.
So before another attempt, I thought I had better garner the benefit of
experience.

TIA, Alan

--
Mint 20.04, kernel 5.4.0-42-generic, Cinnamon 4.6.7
running on an AMD Phenom II X4 Black edition processor with 8GB of DRAM.

On Fri, 02 Apr 2021 19:32:47 +0100
T i m wrote:

On Fri, 2 Apr 2021 18:41:54 +0100, pinnerite
wrote:

snip

So, is there any way you could brace the joint, even if only on one
side with something suitable (steel strip / sheet / rod etc)?


Sadly the shape and size prohibit this.

What about drilling and pegging each end of the joint? It wouldn't
have to be *that* accurate if you have enough 'meat' to drill into and
use a suitable epoxy.

I am going to try some alterative solvents to remove the remaining glue before trying Araldite.

As (chemically) clean as possible is a good start, as might be a
*little* warmth.

Whilst Araldite used to be the go-to adhesive for many things (esp the
24 hr stuff if it really needs to be strong), personally I would first
try something like JB Weld two part epoxy on the grounds that if that
won't hold it, not much else will.

Std 24 hour Araldite is just a good two part resin, the JB Weld is
reinforced, providing a stronger bond as it's less likely to compress
on the side with compression loads and so tear away the rest of the
bond.

I will try to use a clamp.

Anything that will stop it moving whilst it cures is essential.

Once cured, some of these adhesives also enjoy further curing with
some external heat (but don't go mad). ;-)

Cheers, T i m

Thank you for your advice. Believe it or not there is oly ond syockist of JBWeld in the Greater London area. I am going to try the two-part Gorilla again, having checked all the comparative reviews.

I di think of drilling and pinning but holding a small piece of solid Alli and drilling through 3mm or so would require the kind of gear that I don't have. My Dad could have done it. He was fine engineer but I no longer have access to all his tools.

Despite my initial cleaning attempts, first with petrol and next with hand gel, I am still not confident that the edges are clean. I need another solvent to get rid of what I suspect was poorly cured adhesive.


--
Mint 20.04, kernel 5.4.0-42-generic, Cinnamon 4.6.7
running on an AMD Phenom II X4 Black edition processor with 8GB of DRAM.

On Sat, 3 Apr 2021 10:34:14 +0100, pinnerite
wrote:

snip

Thank you for your advice. Believe it or not there is oly ond syockist of JBWeld in the Greater London area.

Plenty on eBay?

I am going to try the two-part Gorilla again, having checked all the comparative reviews.

Erm, but you know it will fail and you will just contaminate the joint
again / further?

Glues are one of those things you get to 'know' and in all my
experience of gluing things, there isn't much to beat JB Weld (and
there are different varieties)

I di think of drilling and pinning but holding a small piece of solid Alli and drilling through 3mm or so would require the kind of gear that I don't have.

A heavy weight or helper to hold each piece still on something and a
an electric drill?

My Dad could have done it. He was fine engineer but I no longer have access to all his tools.

The key is taking your time and setup prep. Given you aren't in a
position to apply a genuine engineering solution the next best thing
is to do your best.

eg, If you drill as square as you can into each section (ideally more
than one point, equally spread apart) using a smaller bit (say 1mm),
then see how close you are using a panel pin with the head cut off
and then re drilling slightly larger, making any angle adjustments as
you go, you should be good to go. Remember you want a bit of slack
around the final dowel(s) to proved a decent layer of epoxy.

Notch the dowels up with some pincers or a hacksaw, chemically clean
everything, warm, mix adhesive, apply to holes and dowels, join parts,
wipe off excess and clamp / weight / tension (Spanish windless /
elastic bands / ratchet straps) the best you can, leave for a day.

Despite my initial cleaning attempts, first with petrol and next with hand gel, I am still not confident that the edges are clean.

Ideally you really want a clean metal to metal joint. Any impurities,
especially at the edges will introduce a weakness, limiting the
strength of the final solution.

I need another solvent to get rid of what I suspect was poorly cured adhesive.

Acetone / a hot air gun and wire brush?

The point is and given the csa of the joint and the potential load on
it, even someone who knows what they are doing (like your Dad) may
have believed to be onto a hiding to nothing and so any attempt at
that sort of repair would have needed to be done to the best / most
likely standard / solution.

Eg, not using the most likely to succeed glue , not cleaning fully,
not keeping dead still whilst curing or then exposing to excessive
forces during reassembly ... were all likely to guarantee it failing
quickly.

You can often get away with less when the odds are better (it only
being aesthetic or the break in a non load-bearing area or a long
split etc) but this doesn't sound like one of those. ;-(

Cheers, T i m

On 03/04/2021 10:34, pinnerite wrote:
Believe it or not there is oly ond syockist of JBWeld in the Greater London area.

JBweld = car body filler for most practical purposes


--
"Women actually are capable of being far more than the feminists will
let them."

On 03/04/2021 10:34, pinnerite wrote:
Despite my initial cleaning attempts, first with petrol and next with hand gel, I am still not confident that the edges are clean. I need another solvent to get rid of what I suspect was poorly cured adhesive.

Petrol: A mix of chemicals of various volatility. Some evaporate at
once, others take longer. It may even have traces of oils.

Hand gel: A mixture of water, various alcohols, and a gelling agent.

Both of those _will_ leave contamination on the surface. I suspect you'd
do better with meths. Or vodka.

Andy