Controlling contact point in deburr toolpaths

[JB7280]

Is it possible to control the contact point in a Multiaxis Deburr (being used in 3 axis) toolpath?  The reason I ask, is I'm having some trouble.  About a year ago, I made this post;

Where I learned that sometimes adding extra length of cut will get you better results in a deburr toolpath.  However, I have a lollipop mill that looks like this;

The only way for it to be able to calculate a toolpath is to extend the length of cut, but then it wants to try to cut in that internal radius area.   I know I could just do it the old way, with a contour, but I'm just wondering if there's a way to make this work properly.  

 

[Aaron Eberhard]

Can you post a file?

[JB7280]

51 minutes ago, Aaron Eberhard said:

Can you post a file?

Not of this part.   But I'll make something similar later to recreate it.   I've left the shop, but Ill try to post it when I get home.   

[JB7280]

On 2/23/2024 at 3:59 PM, Aaron Eberhard said:

Can you post a file?

I made one toolpath with the cut length set to .16", per Harvey's website.  Then another one with an extra .01" added on, so it would generate a toolpath.

 

https://drive.google.com/file/d/1v2mQA36a01resceaPnPiBnz0HMhcrjHL/view?usp=sharing

[Aaron Eberhard]

20 minutes ago, JB7280 said:

I made one toolpath with the cut length set to .16", per Harvey's website.  Then another one with an extra .01" added on, so it would generate a toolpath.

 

https://drive.google.com/file/d/1v2mQA36a01resceaPnPiBnz0HMhcrjHL/view?usp=sharing

I couldn't get it to work, either.  I'm guessing that it's something to do with the 3 axis engine driving it?  I'd send it in to [email protected]. 

What you can do to make it work is to switch to 3+2.  With this part, you still get 3 axis only output, but it will generate.. However, it's still cutting on the "no go" area behind the ball, just like it is when you set the cutting length longer. 

 

To be honest, though, with this cutter, I'm not sure what the desired results are in 3 axis.

I mean, I know we want a perfect .0125" edge break cut like this:

However, that requires the center of the ø.188 cutter to be perfectly 45° from the corner, tangent to the radius with that .0125 width.

If we bring your tool profile over that, you get exactly what you see when you allow it to cut .17 length of cut, instead of .16 (remember that the toolpath will always calculate against a true ball, your custom profile is only for display/verification purposes):

If you try to slide your tool down so it's not cutting with the back part of it, then it becomes impossible to maintain a constant width:

 

I would expect it to work, though, with a constant depth (you'll have an uneven width on the two sides, of course), but that doesn't work either.

 

I'd guess that this is an enhancement, that the algorithm team at ModuleWorks would have to handle specifically.

I think you'll need to drive this one manually and report it for future improvements.

[JB7280]

1 hour ago, Aaron Eberhard said:

I couldn't get it to work, either.  I'm guessing that it's something to do with the 3 axis engine driving it?  I'd send it in to [email protected]. 

What you can do to make it work is to switch to 3+2.  With this part, you still get 3 axis only output, but it will generate.. However, it's still cutting on the "no go" area behind the ball, just like it is when you set the cutting length longer. 

 

To be honest, though, with this cutter, I'm not sure what the desired results are in 3 axis.

I mean, I know we want a perfect .0125" edge break cut like this:

However, that requires the center of the ø.188 cutter to be perfectly 45° from the corner, tangent to the radius with that .0125 width.

If we bring your tool profile over that, you get exactly what you see when you allow it to cut .17 length of cut, instead of .16 (remember that the toolpath will always calculate against a true ball, your custom profile is only for display/verification purposes):

If you try to slide your tool down so it's not cutting with the back part of it, then it becomes impossible to maintain a constant width:

 

I would expect it to work, though, with a constant depth (you'll have an uneven width on the two sides, of course), but that doesn't work either.

 

I'd guess that this is an enhancement, that the algorithm team at ModuleWorks would have to handle specifically.

I think you'll need to drive this one manually and report it for future improvements.

That makes sense.  So it's trying to make it a perfect 45, but the tool profile doesn't allow for it.   

 

I can just do a normal contour, since I'm not really looking for a specific chamfer.  Just a deburr edge break.  

 

Also, I was not aware that Mastercam sees it as a true ball, despite the profile, so I appreciate that bit of knowledge.   

[Aaron Eberhard]

1 hour ago, JB7280 said:

That makes sense.  So it's trying to make it a perfect 45, but the tool profile doesn't allow for it.   

 

I can just do a normal contour, since I'm not really looking for a specific chamfer.  Just a deburr edge break.  

 

Also, I was not aware that Mastercam sees it as a true ball, despite the profile, so I appreciate that bit of knowledge.   

That's what I'd do... 

And yeah, in any toolpath (in pretty much every general-use CAM system I'm aware of), it's only going to be using the mathematical model of the tool selected, and then only apply the custom profile after the fact for stock models/verification/etc.     Otherwise the math driving the toolpaths would get REALLY complex really quickly!