Multiaxis - surface/solid - triangular mesh

[rgregory]

Good day

I am trying to cut a triangular mesh model of a statute bust with the multiaxis surface/solid triangular mesh toolpath and not getting the desired results. No undercuts are tool pathed, and the results are looking like a 3 axis toolpath converted to a 5 axis toolpath. This is not good. I've stumped the dealer and hope some one here can help. I've tried both X5 and X6 with the same disappointing results. Any help would be appreciated

Bob

[Guest]

The file would be far more helpful than the pic

[rgregory]

The file would be far more helpful than the pic

And what file would you like to see? The X6 file is to big.

Bob

[K2csq7]

And what file would you like to see? The X6 file is to big.

Bob

 

Can you save as a different filename, then delete everything except the associative geo (maybe keep a few adjacent surfs for reference), then post it or put it on the FTP?

[rgregory]

Can you save as a different filename, then delete everything except the associative geo (maybe keep a few adjacent surfs for reference), then post it or put it on the FTP?

Can some one tell me how to post to the FTP site?

Bob

[K2csq7]

7th topic from the top in this (industrial) forum "FTP link".

Welcome to the forum!

[rgregory]

7th topic from the top in this (industrial) forum "FTP link".

Welcome to the forum!

I see no instructions and no hint. Why is this so cryptic?

[Trevor Bailey]

ftp://mastercam:swis...cam-cadcam.com/

[Tyler Robertson]

What toolpath strategy are you using - Parallel cuts?

 

Use a pattern surface for your tool axis and make sure it's on the inside of your part

[rgregory]

What toolpath strategy are you using - Parallel cuts?

 

Use a pattern surface for your tool axis and make sure it's on the inside of your part

I have tried constant Z and constant cusp with tool axis directed to a chain closest point. The chain is vertical passing along the center axis. I would post a file but do not know how to use the FTP site. Every toolpath I have tried fails to reach any under cut areas like the pedestal which is smaller in diameter than the top of the bust. I am using a 5 axis Thermwood router and want to believe X6 can cut this but so far no joy.

Bob

[K2csq7]

Using the FTP.

Click on the link Trevor posted above, browse to the appropriate folder. (mastercam.... X5 or X6 files)

Open a windows explorer window and browse to your file.

Drag & drop or copy & paste your file from your windows explorer window to the FTP folder.

[Guest]

Uploading to FTP using IE

 

or

 

[rgregory]

Using the FTP.

Click on the link Trevor posted above, browse to the appropriate folder. (mastercam.... X5 or X6 files)

Open a windows explorer window and browse to your file.

Drag & drop or copy & paste your file from your windows explorer window to the FTP folder.

Thanks for the help, but I still can not upload. I can get to the X6 files and see 3 other files loaded there but when I drag my file to the page, no joy. Mozilla fire fox just wants to open it and not upload. I've tried also to use internet explorer and get the same results. I seem to be missing something critical

Bob

[Guest]

If you strip the file out, leaving just the single path, then zip it up, if it's under 20 megs you can email it to me, I can place it on the FTP and get a look at what you're trying to do

 

jmparis65 at gmail dot com

[Guest]

It's int he X6 folder if anyone wants a peek

 

SAPPHO FINISH TP.Zip

[Guest]

Here is something I almost never say, I've come up empty on this.

 

STL's atypically brutal to work with and this one is not an exception.

 

Having this as a surface model would make this 1,000,000% easier

 

If I get a few extra minutes I'll try a few more things but not looking good at the moment with this file

[crazy^millman]

Well the biggest thing here is going to be the method of approach. You need to think about the surface to drive the toolpath as the axis controls for the 5 axis tool path. Then the STL needs to be the limits for the toolpath. Using the method you are you are limited to option. By using different toolpaths and using the STL as the collision control you can do so much more. The Multi-Axis Toolpath will work great on this great, just need to think about the underlying surface at the axis control, then the STL becomes the Collision control and you are off cutting with some ease. The next problem is the hole under the chin. Anything will have a fit with this in Mastercam expect Verisurf. I filled the hole using the reverse and then re saved the STL. I will be glad to email you the file if you would like.

 

Making the drvie surface model needs to take consideration the cuts, the travel of the machine and the tagnet of the tool to the surfaces. You may want to break this up to get correct cutting parameters. Again this is not something easily done in Mastercam, but Verisurf cut slice and dice this thing to many pieces if need be. Another approach would be use our slice tool ro get surface slices every .5 along the Z axis. Then use Mastercam offset contour to make a bunch of .5 smaller chains. Now I make a net surface out of that and use that as my drive surface for my toolpath and axis control and I got something really special. I programed Thermwood 5 axis routers some years ago, how I got my start in Mastercam 5 axis programming. I see one toolpath to rough this with say .5 ball endmill .3 step leaving .05 and then coming back with a .25 ball endmill to finish. The use the needed end mills where the detail is needed. That will be where the breaking of the STL will be a huge time saver. You can use containment boundaries to limit the toolpath, but the whole STL would still be used for toolpath calculations make this file become one monster quick. Let me know if I can be of assistance.

[Tyler Robertson]

I do this kind of thing all the time, and constantly have to deal with STL files. if I get a chance this afternoon I'll take a stab at it

[rgregory]

I do this kind of thing all the time, and constantly have to deal with STL files. if I get a chance this afternoon I'll take a stab at it

Thanks to all who made suggestions and offered advice. I am cutting foam now using the multiaxis > surface/solid > triangular mesh toolpath. Most of the 3D models I am tasked with cutting are triangular mesh models. Solids or surfaces would be better, but that's probably not going to happen. Here is what I did to make a all most ok toolpath

1) Built a solid column 1" in diameter capped with a 1" dia dome positioned in the core of the bust.

2) Created a rectangular confinement box around the bust in the xy plane

3) Using the constant cusp pattern, selected the solid column and dome cap as my machining surface

4) For tool axis control selected surface with lead/lag and side tilt as ortho to cut direction at each position

5) For collision control uncheck machining surface and select check surface. Select as the check surface the mesh model.

6) For collision control advanced parameters select and check the Extends tool to infinity box

Bingo and I now have a toolpath with undercuts.

 

Anti-Bingo

1) The toolpath is not normal to the mesh surface, its normal to the machining surface so some desired details are machined away

2) The stepdown value selected will be accurate for the machining surface and Not the mesh surface. So as the mesh surface angle changes the visible stepdown changes. A solution is to break the machining into smaller operations and change the stepdown values to compensate for the changes in surface angle.

 

Question? If I use a lollipop cutter instead of a ballend mill and changing the toolpath stratigy can I get the undercuts normal?

Bob

[Ron Branch @ 5th Axis]

That is where a offset surface like a blob from the stl like a filtered stripped down shape, close, but smooth enough to give the angles and undercut control you are looking for. What I was looking to do was slice the stl and then offset the contours smaller. Smooth out that to make a nice smooth shape. Then make a rules surface out of all of those. Now it would be a good surface to use for the normal. Sounds like you got it going, but yes this can be a task to get what you are looking for.

[dan@advancedmold]

Hello Guys,

 

I have almost the exact same problem as above. My toolpath looks and tilts exactly how I want it to using triangular mesh, constant cusp, but will not machine the undercuts even though there is plenty of tilt to make them a positive angled surf. I am not using stl, I have surfs to drive from, and like I said it looks great but no undercuts. Do I need to play with the limits or is there some other trick to get this tp to work 5axis. The undercut is minimal so I finished it with the mesh tp anyway and it posted full 5 and cut fine but no undercuts, I used an offset contour and morphed 2 curves to go back and finish the job but still would like to know how to get undercuts from the triangular mesh. Anyone know what I can do to get this to work next time?

 

Thanks

[Tyler Robertson]

I'll take a look at this - I've found a decent way to do this with STLS that actually solves the 'Anti-Bingo' problems from the last message on this thread.

 

What I do is either use Parallel cuts (constant Z) , or Cuts along curve ( line profile along the model), or any other advanced multiaxis toolpath that gives you the stepover control you want

I select the mesh surface for machining, and let it do a 4-axis just basic Z-level cutting

- set it to 'one way'

- Set linking parameters to 'move to clerance area' between slices.

 

What you get is a toolpath that just creates rings around the model, doesn't tilt/cut very well, but gives you good slices on the STL surface.

 

- Backplot, save the toolpath as geometry to a separate level

- delete all rapid moves

 

now you're left with just a bunch of wireframe rings.

 

- Create Surface > Ruled Lofted

- Select all the wireframe created by the backplot save (mind the start points

 

This should create a lofted surface that more or less resembles your STL, it depends on your original toolpath stepover. This now can be used as the surface for the tool axis control giving you good normals control. Then you can use the STL as your pattern comp surface, or use it with collision check and 'move tool away, along surface normal' to match the STL.

 

Kablammo!

[Tyler Robertson]

I think constant cusp waterfalls kind of by definition, at least when you're using a silouhette boundary - are you using that or 2D? I'm going to try to make it do what you want, but I think another toolpath is the key.

You probably want constant cusp because it scallops...

[dan@advancedmold]

Thanks for the reply Tyler,

 

I am using a silouhette with a.220 offset,(1/2 ball) meaning the path could almost warerfall down the whole side or vertical walls which I dont want to machine anyway as they are done already. The toolpath has near perfect motion (scallop, its a mold) it just wont see the undercuts. Will it recognize them if I use STL as well. I really like the motion and control of the path right now I just think I am missing something. BTW I have a 5axis trunnion. I am use a chain below my surfs for tilt control. I realize this path is modeled after a scallop, I just hoped with the tilt angle of the tool the machining surfs would become machinable as they are no longer a negative draft angle. The part is finiished now, I am just trying to use all the tools in my toolbox; if its the right one or not I guess I want know till I try it.

 

Dan

[dan@advancedmold]

_{Also I have collision control off, the path didnt need it, the motion was good.}