Small mandrel bender

[keith5700]

Whilst on a slow period with my V8 I have been designing my next engine, which will be a 1/3 scale V10.
Remembering what a nightmare the V8 exhaust manifolds were to make, I decided to have a go making a mandrel bender for the V10 exhaust.

The tube size is 14mm o/d x 1mm wall.

Hopefully the principle is self explanitory from the pics.
I cut the half diameters with a 14mm ball nosed end mill, after hogging as much steel away with plain end mills.

The mandrel bit, which goes inside the tube, is just hardened silver steel, sized to be as close a fit to the bore as possible.

I did 4 trial bends, the first without the mandrel inside the tube, and 3 more with the mandrel pushed a bit further past the tangent point of the bend each time.
Pushing the mandrel in any further than the last photo resulted in a lock up of the mandrel in the bend, so that dictated the limit of its forward position.

The dimensions written on the tubes is the thickness after bending, from the i/d to the o/d. I was hoping for the best one to be a bit closer to 14mm, but it's not too bad.

There is still a kink in the i/d at the start of the bend. The mandrel won't prevent this. I still need to see if there is any way of reducing this.

I have just got some ally tube to try. If that doesn't kink then it's just a consequence of the stainless being much harder to pull round the former than the ally.

I know full sized benders have a wiper to reduce the kinking here, but I doubt that's going to scale down very well.

 

[ConductorX]

I am not a metal expert but I have bent and installed miles of stainless and copper tubing. The commerical benders have a limited radius based on the size of the tubing. If you try to make a tighter bend you risk tubing failure or a bad bend. I know the goal here is scale models but you have to work with the limitations of the material as well.

IMHO "G"

 

[Lew Hartswick]

I am not a metal expert but I have bent and installed miles of stainless and copper tubing. The commerical benders have a limited radius based on the size of the tubing. If you try to make a tighter bend you risk tubing failure or a bad bend. I know the goal here is scale models but you have to work with the limitations of the material as well.

IMHO "G"

Some things don't scale well and some (like material properties) don't scale at all.
...lew...

 

[Septic]

ConductorX makes a good point...

I too have spent decades working with tubing of various sizes and compositions and deviating from common practice often introduces weaknesses that aren't immediately obvious.

 

[keith5700]

I hear what you all say.
I know the bend radius is very tight, hence the idea of making the mandrel bender, as opposed to an ordinary type bender, or filling the tubes with low melting point alloy.

I know the bend radius I want, which will look the best, and am now trying to produce that radius.
Actually I'm not that far off what would be acceptable. At the moment there is 16 thou' of crush. If I can get that down to 10 thou' then I'd be happy.

I am going to anneal the stainless, in case that makes a difference.
Slightly tighter fit of the mandrel, ie. bigger diameter.
More grease in the tube so maybe I can push the mandrel further up the tube.
Cheers.

 

[MachineTom]

In large diameter tubing bending, the inner mandrel is a cable with a series or Torus shaped disks, sort of like a doughnut strung on the cable. A hydraulic cylinder tensioned the cable and disks, this allowed the disks to pivot on the cable as the tube was bent.

It seems that you could make up a similar device for you small tubes.

 

[johnny1320]

I have often wondered if using a hydraulic hose under pressure inside the tubing would work for this problem. P.S. Keith5700 awesome workmanship!

 

[Lockstocknbarrel]

Hi Keith,
Impressed with you're craftsmanship, have you thought of filling the tube with fine sand and plugging the ends to limit the distortion, sorry somewhere in the old grey matter of my brain I remember the "Old Gunsmiths" used the method to reduce the problem that you are striking.
Please give it a go and let me know.
Kindest Regards
Beagles

 

[jwcnc1911]

Man that is ingenuity at it's finest there! Necessity is the mother of invention. I saw something similar to this in Modellbau, if you want I'll dig it up and send it to. PM me if your interested.

 

[keith5700]

Some general observations rather than answering specific points:-
The effort required to pull the tube round is far more than I had anticipated. I used a 30" overall length lever, and I had to brace my foot against the bench to stop me pulling it over. I think the forces involved are more than hydraulics or sand could withstand, by a factor of 10 at least.
I have seen the torus shaped discs and they seem to be a good idea, but I think I'd struggle to scale this down into a 12 mm bore, although it may be possible. Maybe a 3mm cable would be enough to drag the ring through.
On a similar theme I wonder if I could force some 12mm ball bearings through after bending.

JW, I would certainly be interested in seeing that feature if you can find it.
Cheers all, hope to have some V8 news soon.

 

[Scott_M]

Hi Keith5700

I remember seeing an episode of "How it's Made" where the were making Trombones and the method they used, now keep in mind this was thin wall brass tubing. They would fill the tube with melted pitch (I think) and let it solidify and then mandrel bend. They did not need to cap the ends. After bending they would melt the pitch out and clean the bent tubes. Here is the cool part , They would put the bent tubes , in this case a "U" shape, inside of a two piece mold with the proper O.D. dimensions of the bent tube. This mold then went under a lever operated press. They would press steel balls through it, all coated with an anti-seize lube. I think they used a series of 3 different size balls. And to get them back out they would press in the smaller sized balls to push out the last big ones. It got those mandrel dents out nicely.

I have no idea how this would work with 1mm wall S.S. tube. Food for thought.

Scott

 

[johnny1320]

Hi Keith5700

I remember seeing an episode of "How it's Made" where the were making Trombones and the method they used, now keep in mind this was thin wall brass tubing. They would fill the tube with melted pitch (I think) and let it solidify and then mandrel bend. They did not need to cap the ends. After bending they would melt the pitch out and clean the bent tubes. Here is the cool part , They would put the bent tubes , in this case a "U" shape, inside of a two piece mold with the proper O.D. dimensions of the bent tube. This mold then went under a lever operated press. They would press steel balls through it, all coated with an anti-seize lube. I think they used a series of 3 different size balls. And to get them back out they would press in the smaller sized balls to push out the last big ones. It got those mandrel dents out nicely.

I have no idea how this would work with 1mm wall S.S. tube. Food for thought.

Scott

Scott you read my mind, I was just coming here to post it and saw your post LOL :wall: they used one ball the diam of the inside and smaller ones behind, but it was also in a die the shape of the "u" bend so it would be hard to make a die the shape of the header pipe, it may still work but it might straighten out the bend.

 

[Scott_M]

but it was also in a die the shape of the "u" bend so it would be hard to make a die the shape of the header pipe, it may still work but it might straighten out the bend.

I am not sure if Keith has CNC mill, If he does it shouldn't be a problem.
If he doesn't, It would be pretty simple to make a 2 piece mold/die with some fast drying epoxy or metal filled Devcon.

Scott

 

[///]

I think the forces involved are more than hydraulics or sand could withstand

Hydraulic should be no problem, but I agree on the sand part.
I tried sand once and the grains just crushed, too course so too many voids between the individual grains.
I thought to try a fine grade of silicon carbide or aluminium oxide, say 1200 mesh or even 3 micron powder, but never got around to it.

 

[gabby]

Hi guys, my 2 bobs worth, In the refrigeration trade and plumbing, we use heavy bending springs that allow for multiple bends in a short length of tube (mainly copper though) but this concept could be adapted to suit our needs as long as the spring used is strong enough and won't crush with the tube then you are guarantied to get a perfectly formed (round profile/ cross section) bend.
It works very well with hard drawn copper as well which would be comparable with light wall Ali.
Like I said, just my 2 bobs worth.
Cheers
Graham :hDe:

 

[gabby]

Sorry, the springs are an internal sizing and don't mark the tube at all.
Cheers
Graham

 

[dr_kingston]

Hi Keith5700,
To remove the deformity in pipe bends, you can blow it out with CO2 or welding grade argon etc, set to about 500 psi, then capping the ends and gently heating the effected area until the deformity has disappeared. If you over heat the deformed area you can blow a bubble which is sure to get your attention.
Regards Dale

 

[keith5700]

I've just got sidetracked again so haven't tried anything to reduce the kinking.
I think the first thing will be to try and force some balls down the bore, but without the outer die.
It would be a major project to make 10 different 3D outer dies, even with a CNC, which I haven't got.

Internal springs are good, but with this bend radius, and with stainless tube, I'm not convinced the spring would ever come out again,
unless the tube was overbent quite a lot, and then straightened up a bit. If I had some snug fitting 12mm O/D spring I'd give it a go though.

Heating with internal gas pressure sounds interesting, but I wouldn't have the nerve to try it.

I'll post back when I'm on it again.
Cheers.

 

[dr_kingston]

Hi Keith5700,
Don't let fear hold you back.
I have made tapered pipes by this method and repaires expansion chambers after removing all of the combustibles from inside the pipe.
Pressure and temperature control are the only skills that are required.
Basically the metal is drawn by the heat and the internal pressure so be carefull as it is easy to over do it.
Cheers.

 

[lohring]

It's a lot safer to use water rather than gas to expand out dents. A pressure washer or plumber's pump can supply the pressure. If something bursts there's a lot less energy to be released than with compressed gas. By the way, that mandrel bender is a great idea. I've not seen any nicer results in model race engine exhaust headers even from commercial benders.

Lohring Miller