I was so pi**ed with the results of tube bending (whatever "tricks" or trade secrets I used) that I have built me a mandrel tube bender. I get the most incredible bends with it. Have been on shows with my Ellwe*) and nearly everyone admired my tube bends. Other HSEs came by and asked how I did that and asked me to bend their tubes (willing to pay money).
Bending a 6mm brass tube, bending radius being 6mm (relative to center line), the diameter collapses just for 0.2..0.3mm. Only minute scratches or dents left. Didn't try it with aluminium, but should work the same (adjusting bending radius).
How does a mandrel tube bender work? (I hope it is called that way): You need a roll that forms the inner half of the tube. To prevent the tube from collapsing a mandrel (stick) is in the tube right at the place where the bend takes place. While bending, the tube gets pulled from the mandrel (it if fixed) and thus supports the outer half of the bend. The end of the mandrel that supports the tube's outer wall has a somehow tricky form. It is a intersection of a torus and a cylinder. Adjusting is a bit complicated, but once it is done, I get repeatable results and a bend is just something of less than a minute.
I could make photos, but you wouldn't understand from them how it works. ;-)
FWIW we have a quite expensive mandrel bender at work for bending Titanium tubing. Out limits are that we cannot have more than a 5 percent change in the cross section shape of the tube after bending. Our particular machine uses an internal mandrel that appears as a series of beads on the end of a rod. As the tube is srawn around the curved form, the "beads" provide the support to keep the tube from collapsing.
If strength is not a major factor, you can soot the aluminum tube with a candle flame(a Sharpie or similar marking pen can be used , too), and heat it until the soot is burned off using a propane or Mapp gas torch. This will effectivly anneal the aluminum tao a state that is about as soft as you can get it. If it is a hardening variety, it will naturally age harden to about a -T3 or -T4 state in a few weeks. Soft aluminum is much easier to bend.
Model Engineer had a pretty decent mandrel bender a few years back. A look through one of the online indexes should find the issues involved, and copies of older articles can be bought from the publishers. IIRC the ME mandrel bender was demonstrated bending 1/2 inch inside radius bends in 1/2 copper pipe (a very tight U shape) without collapsing the pipe.
There are also some low melting temperature alloys that can be poured into tubing to keep it from collapsing when bent. Google for Cerro-bend (or Cerrobend). It is one of several alloys that melts in boiling water.
Some extremely fine models there, Nick. I only wish I had not let my German get so rusty (I had to learn some technical German at university and pass an exam, but a long time ago!) as I would have liked to read the descriptions.
Tweeking with my setup, the cross sections area could even be increased. But then it will be oval.
I know this principle, but at that size it would be to complicated to make. Making mine took me a week to develop. Having thrown away 2 prototypes in that time and learning what is important and trying to figure out what was the cause for a pile of badly bent tubes.
Filling the tube will not work. If it is bent, the inner part of the tube is not (or just a tad) compressed, but the outer part is stretched. This increases the volume of the bent section. The volume of the alloy in the tube stays constant, so it can't support the tube that much.
Exactly that principle. Just my mandrels are looking a bit different. The one shown with the single ball at it's end is primitive if the ball can't move. My mandrel's tip has the form described in another post. It is more complicated to make (filing, there is no easy setup that could generate the shape), but has the longest possible support for the outer part of the wall.
Also try Wood's metal; that is either the same as Cerrobend, or very similar. IIRC, they used to sell teaspoons made of it in joke shops. Not a great idea as some of the constituents are not altogether human-friendly.
Various recipes in Henley's Formulas:
Wood's Metal: Tin 2 lead 4 bismuth 5 to 8 parts, MP 66-72C Other similar alloys: Bismuth 7 lead 6 cadmium 1 parts, MP 82C Bismuth 7-8 lead 4 tin 2 cadmium 1-2 parts, MP 65-71C
Several other recipes involving the same 4 metals in different proportions giving different melting points.
Probably best avoid the ones containing cadmium if you can.
I am making an engine cage for a 1/10th scale car. As it is cosmetic, strength is not at all important. So I think I will be sooting , heating and carefully bending!!!. Would love to have mandrel bender, looks like th eperfect solution, but I think its slightly out of my reach.
I notice that the replies are all about bending tubes, and not bar (6mm ish dia). Can I bend the bar in the same way, or is there a problem trying to do that?
Good to hear that the bar should be easier.But how? Just force, or can I use the soot and heat method to make life easier?
Also, please allow me to crawl for a sec: I am a great fan of your web site and its attitude to engine tuning.........and customers . No, Im not saying which I am!! Anytime I hear some one going on about tuning with 'interesting' BHP claims, I point them to your site for some no nonsense info. Those who manage to read , without the aid of pictures, come out changed, sober, men !! :)
Bar is easier to bend than tube, and mandrel benders are for tube specifically. The mandrel is a bar that slides inside the tube and supports it at the point where the bending is taking place. There is little or no opportunity to redo a bend in mandrell bending.
There are a bunch of different bending machines out there ranging from the sublime (DiAcro bender) to the ridiculous.
DiAcro Bender pics
The Diacro's have been around for years. Calling them the sublime may be pushing things, but they are a flexible shop tool. Note the variuos dies that are used for specific sized pipes or bars, each for a different radius.
The other end of the spectrum is something like this.
Not a lot that could not be made at home, given the ambition.
There are a pile of usefull tools out there, between and beyond both ends of the spectrum if you go looking.
I was also taught to use soap when annealing aluminium and I still do it this way; it would seem to be more accurate as it depends only on temperature to blacken, I would imagine soot would only burn off if the flame was played directly onto it. Martin