Help with Generating Tool Paths for a Specific Model

[roerich_64]

Hi,

here are an example how you could engrave the big campfers:

Keyboard_I.jpg (163.87 KiB) Viewed 1053 times

Keyboard.FCStd

(299.81 KiB) Downloaded 34 times

When you spend your milling machine a 4th-Axis, then it is easy to mill the faces.

Have fun,
Walter

[Thorhian]

Hi,

here are an example how you could engrave the big campfers:

Keyboard_I.jpg

Keyboard.FCStd

When you spend your milling machine a 4th-Axis, then it is easy to mill the faces.

Have fun,
Walter

So what you are saying is that FreeCAD can only do this if you have a 4th axis on your Mill? That isn't very practical, especially since this is going to be a 3 axis machine. A normal end mill for a rough pass creating steps in the stock and then a ball mill to finish it to a nice surface would work just fine. I've seen people create paths for curved surfaces in FreeCAD in the waterline/3d surface split thread, but for some reason, I can't seem to apply that to these chamfers on the top of the model.

[roerich_64]

Hi Thorhian,

So what you are saying is that FreeCAD can only do this if you have a 4th axis on your Mill?

You can do what you want with FreeCAD. It is in your mind

That isn't very practical, especially since this is going to be a 3 axis machine. A normal end mill for a rough pass creating steps in the stock and then a ball mill to finish it to a nice surface would work just fine.

isn't it practical?
I think yes. It saves time

I've seen people create paths for curved surfaces in FreeCAD in the waterline/3d surface split thread, but for some reason, I can't seem to apply that to these chamfers on the top of the model.

This one?: https://forum.freecadweb.org/viewtopic. ... 20#p387234

Ok here is the first shot of your keyboard without 'ball-end-mill clearing'

Keyboard_II.jpg (88.83 KiB) Viewed 1017 times

[Thorhian]

Hi Thorhian,

So what you are saying is that FreeCAD can only do this if you have a 4th axis on your Mill?

You can do what you want with FreeCAD. It is in your mind

That isn't very practical, especially since this is going to be a 3 axis machine. A normal end mill for a rough pass creating steps in the stock and then a ball mill to finish it to a nice surface would work just fine.

isn't it practical?
I think yes. It saves time

I've seen people create paths for curved surfaces in FreeCAD in the waterline/3d surface split thread, but for some reason, I can't seem to apply that to these chamfers on the top of the model.

This one?: https://forum.freecadweb.org/viewtopic. ... 20#p387234

Ok here is the first shot of your keyboard without 'ball-end-mill clearing'

Don't get me wrong, I'd love to make a 4th and maybe even a 5 axis attachment to the machine, but that's just out of my budget at the moment (hence why it is not practical for me at this time lol). Just facing those chamfers would be easier that way if I did have a 4th axis.

[roerich_64]

Hi Thorhian,

Have a nice weekend and be patient for the next step
Will see, what i can do.

Sincerely
Walter

[Russ4262]

Okay. This problem, or lack of capability, with MillFace has been around way too long! One solution is in the PR cue as Path: MillFace - Add `Face Region` method to boundary shape options #3534.

We should not have to create helper pads for milling the tops of the posts (for example) in the model presented in the OP by @Thorhian. This is what MillFace should, and will soon, be capable of doing.

This proposed `Face Region` boundary shape option is for selected faces and is smarter by being partially model-aware. Refer to the attached images, specifically the face adjacent to the vertical wall of the recessed area in which it is located. Also this allows for usage on faces where the bit is larger than the face area, or will not fit into the face because of face geometry that is narrower than the cutter at some point, or in its entirety.

Should the community so desire, an additional adjustment property can be added to reduce the excessive/clearing performed. In the image the left face with a straight edge needs a much larger adjustment factor to clear the entire face than required by the circular face on the right. The current hard-coded adjustment factor is set so that all faces will(should) be cleared, regardless of shape.

Happy milling,
The PathWB Dev Team

[chrisb]

I'm not sure if I like what I see. Would this mean that I mill the part above the posts first and afterwards the complete pocket down to the bottom?
Although it seems to be so clear what I would do manually, I see that it is by far not an easy task to be done automatically.
For generating the GCode manually I would use a big mill and clear everything down to the post top level. Then I may change the tool and mill the rest of the pocket.

[roerich_64]

I'm not sure if I like what I see. Would this mean that I mill the part above the posts first and afterwards the complete pocket down to the bottom?
Although it seems to be so clear what I would do manually, I see that it is by far not an easy task to be done automatically.
For generating the GCode manually I would use a big mill and clear everything down to the post top level. Then I may change the tool and mill the rest of the pocket.

+1

I found to clear the inner room of this keyboard was a predestined job for the new waterline-function.

With them we can do 'mill the room above the posts' first and with a next waterline clear the rest to the bottom.

Example: https://forum.freecadweb.org/viewtopic. ... 81#p403063

[Russ4262]

Morning Gentlemen,
Clearly I miscommunicated the point, which is improved functionality of MillFace. How a user chooses to mill a part for best quality and efficiency is there choice.

The part presented here is a great example for demonstration of the new feature.

Say for discussion's sake that a user for this part started with a MillFace to clear stock down to the top of the part. Some users will then logically choose to apply a PocketShape operation selecting the bottom face of the large recessed area because there is some logic there. Such an operation would leave all the material above the posts. With current FC operation set, and the same 3/8in endmill cutter, FC fails to, or is unable to, access the material over the posts because the cutter is larger than the face area (and without adding addition assistive geometry). Herein is the problem.

The reality is that the cutter really will fit into the available space of the part in order to mill away the remaining material, but MillFace needs this new strategy to understand how to work with an area smaller than the cutter diameter. With this new MillFace improvement our user will then be able to take the logical next step in their progression and clear the material above the posts as their line of thinking suggests.

Yes, I have no arguments that the two step clearing process, suggested in this thread and others, for the recessed area is also a sound and efficient logical progression.

The images presented related to the use case in this thread are intended to demonstrate the improved capabilities for MillFace in the current PR, not purposed to tell the user how to machine their part.

Let's make FreeCAD more robust and capable for multiple logical progressions. ... Yes, I know some use cases do have a single best, and perhaps only, logical progression, but l encourage us to make FreeCAD capable of handling use cases where multiple logical progressions are acceptable for part production. ... Yes, you are correct; I am not a machinist and am in the majority that are not so. ... Yes those of you with more knowledge and experience are teaching me the ways of the machinist jedi, and I am thankful for it.

Russ

[Thorhian]

I'm not sure if I like what I see. Would this mean that I mill the part above the posts first and afterwards the complete pocket down to the bottom?
Although it seems to be so clear what I would do manually, I see that it is by far not an easy task to be done automatically.
For generating the GCode manually I would use a big mill and clear everything down to the post top level. Then I may change the tool and mill the rest of the pocket.

+1

I found to clear the inner room of this keyboard was a predestined job for the new waterline-function.

With them we can do 'mill the room above the posts' first and with a next waterline clear the rest to the bottom.

Example: https://forum.freecadweb.org/viewtopic. ... 81#p403063

The main problem that I have realized with the waterline operation is that it doesn't really give you an opportunity to leave stock radially, whether it be the walls of the inside of the case or on the sides of the posts. This doesn't really give me an opportunity to create a finishing path to make everything nicer in terms of surface finish and tolerances, especially if I don't care if the roughing operations are somewhat sloppy in terms of those two things and I just want to remove a lot of material in as little time as possible (at least according to what my machine will be able to handle anyway). I also would want to use something like the adaptive tool ideally because it will be faster at removing material based on everything I have seen (but I am no expert in machining, this is based on other people's opinions and teachings and even the freesteel blog). The adaptive path operation also makes it easy to leave stock radially (not vertically but freecad's adaptive operation won't even touch the tops of the posts because it seems to be 2.5D only and is not a fully 3D operation). The improved facing operation should allow for finishing the tool posts after using the 2.5D adaptive operation and not have to rely on the waterline or 3D pocket operations exclusively (for the inner recess of the model).

Also, when creating a second waterline operation, it keeps trying to mill out the hole for the USB port lol. Not good. I tried using the boundary adjustment but that also pushes the outward paths more inwards (which would be useful for leaving stock on the walls, not so much on the posts though).

tilt2_cam.FCStd

(322.56 KiB) Downloaded 25 times

Also, thank you Russ4262! Being able to actually face the posts in a separate operation would make it a lot easier to finish the tops of them after some roughing operations.