how can i do this?

[cherokeechief79]

i have to make a baseball trophy.ill make the sphere in the lathe and do the seem and laces on a rotary in a 3 axis mill.i drew the seem and laces flat above the sphere but cannot get it to project properly onto the ball.any ideas?it does project and wraps but not properly

[Guest]

Why not start with something like this, might be easier. You can scale it to the size you need

 

https://dl.dropboxusercontent.com/u/96758767/baseball.STEP

 

 

 

BTW, found on GrabCAD

[cherokeechief79]

thanks ,

im trying to teach my students how to do this too.is there a way to get the laces and rib properly projected onto the sphere?

[cherokeechief79]

i have been playing with that model off and on all day.it is just too detailed for my application.i just want to turn the sphere in the lathe and use a ball endmill in a rotary to cut just a centerline of the laces and the seem.

any ideas?or do you know where i might find a simpler model?

thanks

mark

[cherokeechief79]

is this not possible within mastercam? isnt there someway to project "normal" to the sphere surface?

[brandon b]

There's where you can choose to project normal or view

[CJep]

is this not possible within mastercam? isnt there someway to project "normal" to the sphere surface?

 

That would require a multiaxis toolpath,..

You are trying to cut the seam line and keep the tool normal to the surface of the ball, ideally it would work best with a 5axis setup, the next best would be simultaneous 4axis in an indexer.

Axis substitution is not the answer for this geom.

[Guest]

I think the bottom line here is from a modeling/design standpoint is no, I don't believe Mastercam can do what you're asking and if it can, it is certainly not a simple task.

 

That was why I put up the model of the baseball.

[cherokeechief79]

That would require a multiaxis toolpath,..

You are trying to cut the seam line and keep the tool normal to the surface of the ball, ideally it would work best with a 5axis setup, the next best would be simultaneous 4axis in an indexer.

Axis substitution is not the answer for this geom.

i didnt mean keep the tool normal to the sphere.i wanted the geometry to wrap normal to the sphere.and then use contour 3d with axis sustitution.if it were drawn correctly i think i could get the ball endmill to follow and output code correctly.when i draw the geo above the ball and project the geo onto it it doesnt "wrap" properly.

ill try tha project normal button again but i dont think it worked with that either last time i tried.

[Tyler Robertson]

Can you attach your starting geometry.   The distance from your geometry to the sphere is going to affect how it's wrapped. Depending on the size of your laces, the size of the laces and the distance away from the centre of your sphere will have to be proportional to the radius.  

it may still end up waterfalling in the plane direction

[CJep]

i didnt mean keep the tool normal to the sphere.i wanted the geometry to wrap normal to the sphere.and then use contour 3d with axis sustitution.if it were drawn correctly i think i could get the ball endmill to follow and output code correctly.when i draw the geo above the ball and project the geo onto it it doesnt "wrap" properly.

ill try tha project normal button again but i dont think it worked with that either last time i tried.

 

That's because wrap is setup to roll the geom around a constant dia, axis substitution is going to do the same thing,...

If you want cut this as a 4axis toolpath you should use simultaneous output because of the constantly changing radius of the sphere, the toolpath needs to be able to move on the Y axis.

If you are wrapping geom around the X axis in an axis sub toolpath the Y axis is locked at zero or the center of rotation.

[cherokeechief79]

That's because wrap is setup to roll the geom around a constant dia, axis substitution is going to do the same thing,...

If you want cut this as a 4axis toolpath you should use simultaneous output because of the constantly changing radius of the sphere, the toolpath needs to be able to move on the Y axis.

If you are wrapping geom around the X axis in an axis sub toolpath the Y axis is locked at zero or the center of rotation.

i got it.you are right it would have to be output as 4 axis.

this still seems possible though right?

if i took the geometry from the baseball above and created a sphere in the center of it as a surface could i project the lines onto the sphere ?

ill have to give this a try.

im only persistant on this because i teach cnc at night and my students are asking about this.i dont want to just say we cant do it.

[djstedman]

Its definitely possible to do it, just not with a simple xform -> project to create the geometry.. I think you would have better luck if you were to xform project from at least three different sides of the part and fudge the lines together.. once you have the geometry it should be easier.

 

I will give this a shot and see how it works out..

[CJep]

You would need to do the projection in 3 dimensions, when you're projecting in the T/F/R planes it's a 2D projection associated to the Z axis of that plane.

I have used the ISO plane as a toolplane before to do something similar for some multiaxis tom-foolery,...

[cherokeechief79]

thank you i didnt think of that! ill give it a try!

[cherokeechief79]

ok ive almost got it. it does look like I can do it with just 3 axis.all I need now is an accurate layout of the seam with stiching laid out flat.

anyone have anything?

[cherokeechief79]

just when I thought I had it! I got the seem ok and it looks great but the laces do not project well to the sphere.it seems they have to be originally drawn normal to the sphere surface.that would mean they would have to twist as they are driven along the seem contour.

back to the drawing board.

[Machineguy]

Draw it in Rhino. Machine on a 5 axis with 1 flip.

 

Machineguy

[Tyler Robotson]

just when I thought I had it! I got the seem ok and it looks great but the laces do not project well to the sphere.it seems they have to be originally drawn normal to the sphere surface.that would mean they would have to twist as they are driven along the seem contour.

back to the drawing board.

 

roll the seam close to the ball radius and then project it - will still be distorted but wold be much closer  and might look more 'wrapped'. 

[cherokeechief79]

ok I got it pretty nice looking finally.

the problem im having now is that my rotary is on the right side of the table and the part is in the chuck facing to the left as looking at it from the front.

the simulation looks very good but the rotary is turning tin the wrong direction.im only doing the laces first so its prolly not going to be a problem except that the "V" is going in the opposite direction.when I get to the text ill have to correct this somehow.

is there an easy way to force the "A" to go neg instead of pos?im just using the generic has 4 axis post.

 

[cherokeechief79]

here is the file.

I didn't include a stock file but just create an stl to use for stock from the outer profile.

 

[cherokeechief79]

sorry file was too large to upload

[Teh Bear without Brains kinda idiot]

I did this sort of job on three axis only with hand driven indexing

rotary head

the trick is to split job on angle sectors about 30 degrees use toolpath for ball tip compensation center

You `ll get the right results cause you will always get the same section of the sphere

the principe is the key toolpathes can be different

 

 

HTH

[cherokeechief79]

it came out great.i used the model from grabcad and extracted all of the edge curves. I deleted all but what I needed.then I scaled them slightly larger than the sphere I created for the baseball and projected them onto the sphere surface "normal".

after that it was just countour with a .032 ball and swap y for the rotary.

[Zoober]

Nice job!