[hoss2006]

Slashmaster brought this up in a past correspondence and again in this thread.
Now that that project is done, I thought I'd give it a whirl.
I found this PDF that explains the different types.
Apparently there are 4 methods to the Incremental Forming Process.
Negative die-less forming, also known as single point incremental forming (SPIF).
Positive die-less forming, also known as two point incremental forming (TPIF).
Negative with-die forming and Positive with-die forming.
As I said, I plan on trying the SPIF dieless method.
Certainly looks to be the easiest.
I picked up some .025 Aluminum sheet to start with, steel might be too tough to tackle but we'll see.

This video shows how the big boys do it with a die.

[ame="http://www.youtube.com/watch?v=KuR_fnZ9JvI&feature=channel_page"]YouTube - OCAS BV - Sheet Metal Fast Prototyping[/ame]


If it works, could be a cool way to make some custom covers and shields.
I made a couple tools to try, one has a .250 ball and the other .375, Rockwell 56 hard.
I won't bother to make a clamping system with De-Sta-Cos until I'm sure it works.
I'll start the experiment this weekend.
Hoss

[escott76]

Slashmaster brought this up in a past correspondence and again in this thread.
Now that that project is done, I thought I'd give it a whirl.
I found this PDF that explains the different types.
Apparently there are 4 methods to the Incremental Forming Process.
Negative die-less forming, also known as single point incremental forming (SPIF).
Positive die-less forming, also known as two point incremental forming (TPIF).
Negative with-die forming and Positive with-die forming.
As I said, I plan on trying the SPIF dieless method.
Certainly looks to be the easiest.
I picked up some .025 Aluminum sheet to start with, steel might be too tough to tackle but we'll see.

This video shows how the big boys do it with a die.

YouTube - OCAS BV - Sheet Metal Fast Prototyping


If it works, could be a cool way to make some custom covers and shields.
I made a couple tools to try, one has a .250 ball and the other .375.
I won't bother to make a clamping system with De-Sta-Cos until I'm sure it works.
I'll start the experiment this weekend.
Hoss

I've been playing with it a lot lately. Let the tool free wheel, spinning is a disadvantage and will generate heat, can tend to gall. If you are going to spin the tool, do it slowly, and make sure you are rolling the right direction. You want to go in a "conventional mill" direction, or conversely spin the tool backwards for a "climb mill" tool path.
Use oil, and defintely use both the clamp plate and preferably a plate underneath the part with a hole in it for the portion which drops through. It won't turn out all that well without this plate. I wouldn't use those clamps you mention, I cut bars with holes for screws like they show, you want to apply even pressure and lock the sheet down.
CRS and aluminum (5052) work great, 304 didn't do so hot (not surprised). Trying for vertical walls will result in a bad time and broken parts, stick to 45 deg or more to be safe.
It's a good technique, but I too have done it on a much bigger machine than an X2/X3. I'm sure .025 or so that you had mentioned would be fine though. Just be sure to get the support on the bottom.
I use about 40ipm, 200RPM, and .010 3D stepover and it comes out pretty good.

[hoss2006]

Thanks for the tips.:cheers:

[slashmaster]

Hoss, Great to see your going to try this! Looking forward to any video you make of it. What did you make your tools out of? What size collets do they take?

Escott76, just curious, how did you come to the conclusion letting it freewheel is the way to go? I thought creating heat with the friction of it spinning was the whole point because the metal bends easier that way. Every incremental sheet metal video I've ever seen they spin it.

[escott76]

Hoss, Great to see your going to try this! Looking forward to any video you make of it. What did you make your tools out of? What size collets do they take?

Escott76, just curious, how did you come to the conclusion letting it freewheel is the way to go? I thought creating heat with the friction of it spinning was the whole point because the metal bends easier that way. Every incremental sheet metal video I've ever seen they spin it.

Experimental results. Creating heat is not the object of the process. I thought so too at first, but that isn't how it works.

[Geof]

Metal spinning, without spinning.

I would not be surprised if there is a bit of proprietary knowledge about what alloys to use; the material in those prototypes get stretched an awful long way.

[HawkJET]

Metal spinning, without spinning.

Exactly! The definition of a lathe is that the part rotates and the tool is stationary. Whereas, with a milling machine the tool rotates and the part is stationary.

In metal spinning, a lathe is used and a tool presses the sheet metal against a form as it rotates.

In this process, you are merely transferring from a lathe to a mill.

My guess is that if you are experienced in metal spinning, you would have a leg up with this process.

the material in those prototypes get stretched an awful long way.

You will notice that the parts of the example that get stretched the most are made as a male portion of the form. If it were the opposite, and those features were in a cavity on the form (instead of a core), the perimeter of the cavity would constrain the perimeter of the feature. This would limit the amount of material available for stretching.

As you can see, the method used, allows for a much larger area of metal to stretch to produce the features. This makes the wall thickness more uniform throughout the part and reduces the stress on the material.

[hoss2006]

I only had time for a quick experiment but it looks promising.
Just did a 1 inch circle pocket with the Mach wizard, .200 depth, 1000 rpm, 7 IPM, 15% stepover, .010 depth per pass.
Read thru a bunch more pdf's on the subject.
Read that faster spindle rotation and smaller tool size improve formability so I used the .250 tool.
There was too much flex in the thin aluminum plate so I reset with a piece of high density foam as a backer.
Worked ok but it left a little warpage.
A die would work better to give it a sharper form.
I'll try that, maybe make a fixture to hold the die and plate together.
I want to try oil instead of coolant too, might make a smoother finish.
More to come.
Hoss

[ame="http://www.youtube.com/watch?v=7spGdUamJPk&feature=channel_page"]YouTube - Incremental Sheet Metal Forming SPIF 1st Experiment[/ame]

[diycnc]

i doubt that you will end up with great results without having a die.

if your looking to make simple boxes and covers out of sheet metal then you should do what i did with my cnc controller. start with .080 alum and vnotch the back of the bend .040 deep, at 45 deg with .031 flat. then you can bend over the edge of a table and end up with a perfect looking bend and nice tight radius. i did this using a table saw with the blade at 45 deg. if you had a cnc machine it would make lift that much easier.

[slashmaster]

Great first experiment! Was there any slop between the sheet and the backer? Did that pdf tell you that you can't run higher rpm with faster feedrates so that's why it took 39 minutes?

[escott76]

i doubt that you will end up with great results without having a die.

You would be wrong then. Backer plate needed yes, but you can get fantastic results with no die. This process is of almost no use in making 90 degree walls, as the way that the material stretches while being formed will not handle 90 degree walls. That kind of thing is better done with a brake, or the method you suggest.
An alternative to let you use this tech to make a 90 (you could overbend as well with a back cut bottom) degree bend is to cut out the sheet, clamp the part over something as simple as a 123 block or similar, and then use incremental forming to "wipe" the edges down. You of course need to keep in mind minimum radii as well with all of this, there are charts available

[diycnc]

please link to some video or pictures of this method with no die making anything even remotely complicated with good results. i am very interested, i may even make a machine for this myself.

this is a cool technique but i don't see it serve much usefulness past making some simple embosses.


also if your using a 123 block you should realize that that becomes your die..

[escott76]

please link to some video or pictures of this method with no die making anything even remotely complicated with good results. i am very interested, i may even make a machine for this myself.

this is a cool technique but i don't see it serve much usefulness past making some simple embosses.


also if your using a 123 block you should realize that that becomes your die..

First make a "simple emboss" tool. Use 45 degree sides to make it easy, and go 10x material thickness for the depth of emboss, you can make the tool round to make it "easy" for you. You've got to get the sheet around to stay flat too.
It isn't that easy to do a simple emboss, and what I described above is extremely difficult at best with traditional tooling. A "simple" emboss requires a lot of work in tooling, and if you screw up that tooling, you may have to scrap it all and start over. Change material thickness? There's another tool. This feat (and I've done over 20x material) is easily achievable with this method. It has many advantages over traditional sheet metal techniques in certain applications, especially when you only need a few of something, and you don't happen to have a press brake handy.
I regret that I cannot post pictures or video of the parts I have done. They are customer parts, and done for my employer, so I am not at liberty to do so. When I get some my mill at home set up for this, I will post pictures of those parts.

[BobWarfield]

Fun stuff!

[hoss2006]

please link to some video or pictures of this method with no die making anything even remotely complicated with good results. i am very interested, i may even make a machine for this myself.

The link I gave in post #1 has pics of parts made using the dieless method.

I not trying this for any specific task in mind, just experimenting to see what my mill can do.
To be honest, this is seeming like it would take way too long to make anything of decent size.
I'll try speeding things up but it's like 3D milling, a lot of lines of code to run.
Hoss

[hoss2006]

This project will be on hold for a while, I have to work on some parts for a buddies robot.
Hoss