[njitnjau]

Hi all

This is a letter i sent to Onecnc support describing the problem. Maybe someone can help us here on the forum how I can sole this. I am not convinced that Onecnc is the best CAD CAM software out there. Are you?
I also attach the xfa for all you onecnc experts out there!

Cheers,

--
Letter to onecnc support

Hello

We have bought Onecnc professional XR2. We are trying to make inlays on a guitar´s fretboard. The inlays are curved in the bottom (variable z), and also they sit on different z heights, because the fretboard is banana shaped lengthwise. Therefore we have created a surface that we use as a z-plane to machine against. And sidewise we use a boundry. We cut with 2 different tools, one 3.175 mm (1/8 inch) mill for doing the roughing, and one 0.794mm (1/32 inch) mill for finishing the corners.

The strategy we have tried is to do a first run using the 3.175 mm end mill and do an SMT Planar Finish path to get most material cut out of the inlays. Because we do not want the small end mill to travel where the big mill already has cut, we have tried to make an offset inside the boundry that equals the radius of the 0.794 mm mill. The we do a Stock Cut chain variable z to follow this path.

First problem is that when we do the cuts using SMT planar finish, in two similar sized inlay pockets, the program wants to run the mill up and down in zigzag pattern. However the number of cuts are different between the two pockets of example. We can not understand why the program choses to run different number of cuts for two similar jobs on the same workpiece.

Secondly, while in the first run when we are using the 3.175 mm mill, the code skips some areas that it will go back to do as the last operation. For instance we have 9 cavities to do, and when the code is doing nr 6 the code skips cutting some material. After hole nr 9 is finished, the machine moves back to finish off the number 6 hole, and then stops. Why does it not finish every cavity before it moves on to the next one? This seems to be a completely wrong order to do things.

As said we have tried many ways of working out how ONECNC shall machine the cavities. However it doesn´t seem we have the appropriate CAM system for our needs. Of course there could be alternative ways of doing this part, but we are unable to figure out how. If you have any suggestion how we should approach this particular problem it would be most valuble for us to receive this information from you.

In this email i attach the XFA file we use and a screendump of the erratic (?) toolpath in jpg format. The toolpath can be found in NC manager - Fretboard - Inlays 3175mm by doing a Preview toolpath. Cavities number 4 and 5 are same size but are cut differently and turns out having different sizes. The program goes back to finish off inlay number 6 in the end.
- Layer Fretboard -1.3 is the plane we use as a bottom, it lies 1.3 mm under the fretboard surface.
- Layer Inlays offset is the inlays having a smaller perimeter than the actual nlay size.
- Layer Inlays outline are the outlines for the full sized inlay boundry.

---

[wms]

Looks to me like you are trying to use a flat end mill to planar a curved surface...You are doing 3D with a flat end mill. That is not normal practice for sure..

Use a ball nosed mill to do this..see below..works fine..

[Kiwi]

>First problem
>However the number of cuts are different between the two pockets of example. We can not understand why the program choses to run >different number of cuts for two similar jobs on the same workpiece.

I believe the spacing is a multiple of the tool diameter from the start line and because the second pocket is not exactly the correct spacing (relative to cutter dia) the first cut in the second pocket is not exactly the same as the first.
Suggest you create the tool path for each pocket separately.

[Kiwi]

Correction
I believe I should have said Step-Over not Tool Diameter.
Hope I've got it right this time!

[njitnjau]

Well thanks for the feedback. I don't think using a ball end mill would change the behavour of the paths, or would it?? looks you have difference between 6 and 7 instead in your pics. I use the flat mill because i dont need the high accuracy in the bottom of the cavity, rather the edges i need to get sharp all way down. i glue in the inlays and the glue fills up the rough bottom that the end mill might create. besides the raduis is very big compared to the mill diameter.

I will try to do the pockets separately and see if that does the trick...

[HuFlungDung]

Using a ball mill will have a slight effect. OneCNC will prevent the flat ended mill from descending to full depth everywhere except when it is exactly tangent to the apex of the curve. But, that is not going to really affect the problem you have with the stepover amount not being exactly divisible by the allowed width of the pockets.

If I understand your problem here correctly, I would advise that you use a ball mill, and use an angled setting for your planar toolpathed direction, like 45° instead of 90 or 0 degrees. This will shuffle any unequal cut (due to the stepover amount not being equally divisible into the allotted width of the pocket) to one of the corners, which will be cleaned up later when you profile cut the pocket wall.

[njitnjau]

Well i tried to do each cavity separately. Same result, in similar sized cavities i have different numbers of cuts. Also in two of the cavities ONEcnc jumps over a path and goes back to finish off in the end. So this is not the correct approach, apparently.

>First problem
>However the number of cuts are different between the two pockets of example. We can not understand why the program choses to run >different number of cuts for two similar jobs on the same workpiece.

I believe the spacing is a multiple of the tool diameter from the start line and because the second pocket is not exactly the correct spacing (relative to cutter dia) the first cut in the second pocket is not exactly the same as the first.
Suggest you create the tool path for each pocket separately.

[Kiwi]

Looks like ONEcnc uses the spacing increments from the same point regardless whether generated separately or not.
Your solution will be to cut a profile which will clean the edges, as Hu suggested.