Bending aluminium

[Rolyata]

I want to make up some bent aluminium components.

The items will be 'U" shape brackets.

I intend to start with aluminium strip that's about 250mm long and 50mm wide and then bend the last 50mm at each end to stand up at 90 deg.

The thing I don't know is how thick can the strip be and still be sensibly bent without special tools ? For my purposes, the thicker the better as it needs to have holes tapped in it, but I still need to bend the stuff.

I don't know exactly what grade it's going to be as it's just the ali strip I get from my ironmonger.

As far as bending is concerned, I usually use an old Workmate with angle iron as jaw protectors/strengtheners. I mark the bend point, clamp it up, place a hardwood block against it and belt it with the hammer ( 2kg ish ).

So my question is really three questions

What's the thickest 50mm wide strip that I'm likely to successfully bend this way ?

Is there a better way of doing it without buying special tools ? I assume that hot bending isn't appropriate for ali, neither is any form of heat softening or hardening a practical proposition.

If all else fails, how much might I have to pay to get four items like that made for me in 6mm thick strip ?

[Set Square]

Forget trying to bend 6mm thick aluminium - 2-3 mm is the best you are likely to achieve. Forget trying to do it with a Workmate too - you need a decent metal-working vice.

Rather than trying to cut threads in thin sheet - which won't work, anyway - why not use threaded inserts such as Rivnuts [See

- but note that you don't need their expensive tool to attach them to sheet metal - you can DIY with a fully threaded bolt, a nut and a washer!]

[Dave Plowman (News)]

You might find you can get 'U' section ally near enough for what you need. And at 50mm wide, it wouldn't cost a fortune for offcuts. George King Metals in Colliers Wood would be my first call if you're anywhere close - they sell by weight and of course a cutting charge.

Annealing ally is easy. Wipe it with soap - the sort you clean your face with - and heat it 'till the soap turns black. Then quench in water.

I'd first try to find suitable U section extruded. It's a delight to work with. I use it for making up things that have to fit in a 19" bay, etc, and it comes in various heights. I've not needed anything 250mm high, but with a bit of luck...

[Rolyata]

4mm would do, 6mm would be better. There's no significant load on the threads, the threaded part just hold another minor part in place.

Inserts would stick out the other side and I need the front & back to be flat so that it can slide into alternative fixing positions - where it's fastened by a bolt through a hole into something more solid.

I don't have a heavy enough bench for a decent vice.

I do have a tool for setting threaded inserts. I don't much like it as the inserts don't always bite very well and sometimes work loose enough to rotate in their hole. The tool looks much like the one you linked to. I get better results by doing what I previously used to do. Cinch nuts, a ball bearing and a hammer !

[Rolyata]

Dave Plowman said -

I rather doubt it.

The width needs to fit an existing part and it's not a convenient round figure either. Extrusions always seem to be whole inch or cm multiples.

That's why I reckon I'll need to make something the right size.

[The Natural Philosopher]

About 5mm at a guess.

Actually, they are. Heating and slow cooling anneals. Heating and fast cooling re-hardens.

The actual heat is not great - 300-400. Not red hot anyway.

Depends on whether you can find an old boy with a fly press or bench folder as opposed to a metal production shop with all CNC machinery and a requiremenmt for 100% loading at all times to pay for the million quids worth of machinery...

[Rolyata]

Thanks to the two of you for the info about annealing and hardening.

I can remember doing that as a kid when I used to make model aircraft undercarts, but had long forgotten the details.

But the aluminium I used to use was an alloy that's much stiffer than the aluminium I usually find in shops.

Is there an easy ( and cheap ) source of an alloy that's more suitable than bog-standard soft ali ?

[Andy Dingley]

Not on aluminium (or most non-ferrous metals, for that matter).

Annealing of aluminium or copper is a different process from steel. For steel you're trying to change the crystal structure of the metal, either by heating it to a temperature where the desired structure is stable, then freezing it in place (hardening) or by slowly reducing it in temperature so that it can re-align to the structure that's stable at low temperatures (annealing).

For aluminium (and pure unalloyed metals in general), you're just shaking the dislocations out of the crystal lattice and putting it back the way it was.

[nightjar]

...

It does, however, normally require a temperature controlled oven to do it in. For example, the annealing information I have for one fairly common free-machining age-hardening alloy is to hold at 410C for 2-3 hours, reduce at 10C per hour down to 260C, then air cool. Age hardening, for T4 temper is soak at 510C for 3 hours, then water quench. Hardening for optimum strength (T8 temper) requires 510C for an adequate time to heat the piece through, water quench, cold working to shape, then holding at 160C for 15 hours before air cooling.

Colin Bignell

[Mary Fisher]

NONONO! Not with Al!

Mary

>

[Dave Plowman (News)]

Ah - right. Your original post suggested the size wasn't critical.

[Dave Plowman (News)]

Phew. I thought my senility was getting worse. ;-)

[Andrew Gabriel]

I think you're describing work hardening. Annealing (heating) is the opposite, breaking the lattice up into small crystal domains with lots of dislocation boundaries so it bends more easily (at the more dislocation boundaries).

[Baz]

Dave Plowmans method for Annealing is the way to go.

"Annealing ally is easy. Wipe it with soap - the sort you clean your face with - and heat it 'till the soap turns black. Then quench in water."

Baz

[The Natural Philosopher]

Yes, with Al.

[The Natural Philosopher]

[Andy Dingley]

No, I'm actually describing the annealing process in unalloyed copper (because it's simpler). I'm describing the thermal process applied to reverse work hardening.

Aluminium is a bit more complex, and so yes I am being somewhat inaccurate here. It's complicated by being an alloy (nearly all of the time) and also because dislocations are inherently more mobile in aluminium than they are in copper. Where dislocations become trapped in aluminium is due to the alloying elements and the domain boundaries they give rise to, not just to dislocations trapping themselves, as they do in copper.

What I describe does go on in aluminium, it's just not the only, or even the most significant process. In pure alumium that's purely work-hardened then it may be. The other annealing process in aluminium (as you mention it) is to do with dislocations that are trapped on these boundaries, or by changing the conditions at these boundaries themselves. This is the hardening that's caused by heat treatment or aging of the aluminium alloys.

[Andy Dingley]

Baz wrote (and you followed up):

So which is it ? Does heating and quenching harden aluminium or soften it ?

The answer is of course both. There _are_ quenching processes for aluminium that strengthen it, and coincidentally make it less ductile. They're not useful "hardening" processes though.

If your bike frame is in 6065-T6 alloy, then that "T6" refers to just such a process.

But for all practical home workshop processes, any heating of aluminium anneals, normalises and softens it - whether you quench it afterwards or not.

[nightjar]

Yes, given the right alloy, equipment and procedures, (see my other post). It is not, however, a home project. Even with a professional metal workshop, I send the stuff away to a specialist heat treatment company.

Colin Bignell

[nightjar]

However, if you want to put a thread in annealed aluminium, use a three-lobe displacement tap, not a convential tap that cuts the threads. Three-lobe taps work harden the material locally, making the thread stronger than if you cut it.

Colin Bignell