Profile modeler inconsistent toolpath

[Jeremy Schiffer]

Playing around with using profile modeler to miter edges, and we're getting different results using the same profile on different edges of the same part.

The part is an end panel with a toekick cut in it, and we are mitering the front edge of the end panel and the toekick cutout. The front edge of the end panel does perfectly, but the toekick miter has several oddities:

As you can see, for some reason the toekick portion gets cut way deeper than the part thickness - I did cut it, and it did go into the wasteboard by that amount.

Also, it cut into the face of the toekick miter, you can see the toolpath is buried in it.

Can any of the Thermwood crew (or anyone with more knowledge of profile modeler than me) shed some light on this?

[Clint Buechlein]

Jeremy,

Can you send me an export from Control Nesting to cabinets@thermwood.com so I can take a look? You can post it here if you want but tends to be a few extra steps. Just make sure that .twd is loaded in so it is in for testing. I may need more files but that will be a start.

Some initial questions I would ask:
- Did you use the same profile tool for both edges?
- Did you plunge to same depth for both profiles? By chance fat finger and add an extra digit?
- Did you sweep the same direction for both edges? Top to bottom or bottom to top for the edges.

A random guess I have is that your 45 degree tool is taller than your material and fat fingered putting in the value for plunge depth. This can be checked by choosing custom tool for the profile instead of modeling in Control Nesting, the Z output for that sweep should be the depth you swept at. But that's spitballing, export should get me started at least.

-Clint-

[Jeremy Schiffer]

Hey Clint, just emailed you the export. To answer your questions:

1 - Yes, the same profile tool for both edges
2 - Yes, we checked for a fat-finger mistake and the depths are indeed the same
3 - Yes, both profiles are swept in the same direction.

The "bit" is 1" tall, and we're plunging it 3/4", which it seems to be obeying on the front edge cut. But on the toekick cut, it seems to be plunging to the full 1", ignoring the depth setting.

Even though the file you have is actually a flip, the flip doesn't affect it, we have already tried it without the flip, same result. Still fiddling with it...

[Jeremy Schiffer]

We're making progress. It is the eCabs-generated toekick notch that is part of the problem. We got rid of that, and replaced it with a part editor cut, and the toolpath was correct. We then added back the cut on the front edge of the part, and now suddenly the part-editor-cut toekick notch is not doing right. It decided not to follow the angle of the miter. Here's a picture showing it:

[Jeremy Schiffer]

We remade the tool and changed the height of it to exactly the material thickness, and plunged it exactly the material thickness amount, and it shows correctly now. But that means we can never change material thicknesses, or would have to make a new profile tool for every material thickness.

[Clint Buechlein]

Jeremy,

Can you post or email me the .tol for your original tool, your plunge depth, and the cabinet you are working with? I don't get the same results as you working with the below pictured tool on the Std Base Frameless cabinet with toe kick notch. I plunged mine at -.75", the thickness of 3/4 Generic. Picture from Profile Modeler in Control Nesting is using your export with the material added to Group 1.

-Clint-

[Kerry Fullington]

Jeremy,

Make sure your chamfer tool in eCabinets is symmetrical, and that you have it cutting on the line, No comps or offsets. Plunge it the depth of your material. You also won't need to do the toe cut in the part editor.

this is what I get in Profile Modeler

[Clint Buechlein]

Kerry,

I tried a tool similar to yours, but it did profiling for the entire tool in the toe kick notch area. I suspect this is because it was seeing the top of the notch area as having material the profile was "touching" so it created the path for the entire tool. Jeremy is on latest version of Control Nesting, and I assume looking at this in the Configure Specific Regions section of Control Nesting. If it is being looked it in specifically Profile Modeler (such as what it looks like in your screen grab), that is another story as it runs slightly different code than what Control Nesting does. With release of version 7, you can now view your profiles in Control Nesting as well as make tool group selections based off material groups.

-Clint-

[Kerry Fullington]

Clint,

We are using Control Nesting 7.2 and this is the view of the profile modeler from inside Control Nesting

I will try cutting the part tomorrow.

[Clint Buechlein]

Kerry,

Might be my eyes, but I think I see red lines like it is profiling open space around the part? You should be able to see for sure if you rotate around and zoom in. If the red lines are there in the screen the motions will be in the code for them.

-Clint-

[Kerry Fullington]

Clint,
Here is a shot of what I got after cutting.
My cut on the toe was shallow about 1/8" and it cut into the top of the part the same amount which is half the diameter of the tool I used.
It also cut the back side of the tool as you guessed.

[Jeremy Schiffer]

Clint, we've got it working now, using an asymmetrical tool such as you have drawn. It is possible to change the material thinner, and it will still work. But making the material thicker will create problems.

Using a symmetrical tool gets exactly what Kerry has shown - which is what you'd expect given the tool won't fit into the bottom of the V. It's more of a mystery as to why the asymmetrical tool works, because it is sharp on the bottom as well, but the tool DOES go all the way into the point.

Still scratching my head, but we got it working well enough for this job. We did a LOT of experimenting here and discovered several things, too much to try and type out right now. But if you want to investigate/troubleshoot further, feel free to give me a call (706-677-2165) and we can go over it faster and in more detail that way.

[Kerry Fullington]

What is the reason for the cuts being different on the front edge and the toe cut, when they were created with the same tool and applied the same?
The Part Editor - Profile Modeler relationship has always been very difficult. I wish I could learn more about it.

[Clint Buechlein]

What is the reason for the cuts being different on the front edge and the toe cut, when they were created with the same tool and applied the same?
The Part Editor - Profile Modeler relationship has always been very difficult. I wish I could learn more about it.

Kerry,

In my speculation (software engineer I need to ask is in Europe), it is because with the toe kick, the profile tool is touching the shaded area in picture 2. Not removing anything, but the tool is riding along that surface so it is picked up as an area the tool is contacting. In picture 1, I used Contour Mode to create geometry for the tool to sweep on, .060" away from the top of the toe kick, and swept along it. Same tool, same depth.

-Clint-

[Mark McCallum]

What is the reason for the cuts being different on the front edge and the toe cut, when they were created with the same tool and applied the same?
The Part Editor - Profile Modeler relationship has always been very difficult. I wish I could learn more about it.

Yes, Kerry, I also wish I could learn more.
A couple of thoughts I had. Thermwood sets up zoom class and we all pay a ticket price
or they upload a tutorial to the ecabs vimeo training sitehttps://vimeo.com/ondemand/thermwood

Or we all just pick a time and try and pool our bits of knoweldge in our own zoom meeting.
I think a poll can be set up in a new topic if anyone is good at organizing things like that.

Perhaps Clint could give us a heads up?