3D HST "Facing" options

[TFarrell9]

Howdy, 

I'm looking for some advice when it comes to finish facing surfaces that have shoulders. I generally choose not to finish the faces of pads or bearing rails with optirough (even though this is a very simple way) purely for surface finish appearance. I often use 3D Area Rough for finishing these surfaces previously roughed with optirough and it pretty much always comes out the way I want.

However, sometimes it will ramp in the middle of surfaces (reminiscent of Horizontal Area), even with "from outside" control with "outside" compensation. It also seems to completely ignore my entry commands, even with steep/shallow min/max at the same depth, it will enter from 1/4"+ in the Z axis. This is particularly annoying when using a tool not capable of ramping. 

Attached is a stripped sample file. I've noted two pads where it decides to ramp....seemingly completely random. I've tried expanding boundaries on these pads by a large margin and it still wants to ramp here. I also tried giving large "skip pockets" values and "apply leads" with no luck. 

Anyone know of a good way to do this without having to make a bunch of extra geometry or a bunch of different paths? It's so easy to rough out, and surprisingly frustrating to finish some extremely basic surfaces. 

Sample.mcam

[Jake L]

3 hours ago, TFarrell9 said:

However, sometimes it will ramp in the middle of surfaces (reminiscent of Horizontal Area), even with "from outside" control with "outside" compensation. It also seems to completely ignore my entry commands, even with steep/shallow min/max at the same depth, it will enter from 1/4"+ in the Z axis. This is particularly annoying when using a tool not capable of ramping. 

That pretty much sums up why I don't like to use Area Mill or Area Rough. I'll try them to see if I get a good toolpath (no ramps), but often I'll just spend the time to make the extra geometry. Sometimes changing the min/max stepover can get rid of the ramps. But if the toolpath ends up with a blind "pocket" of material to take out I've never been able to stop it from ramping it.

In your sample it looks like the toolpath is morphing kind of weird because it has to go around the .200 high "walls". What I'd do is split that path into two or three paths. Maybe make all the square pads one path, then all the longer rails another path, and the rest a third? That should at least make the toolpath on the pads cut straight along one of the axis.

I don't know of a good way to fix it with a single toolpath. You could use toolpath editor but that comes with it's own pros and cons. And considering it looks like it took almost 3 minutes to regenerate the toolpath , I don't have the time to play with it too much. 

You could try 2D Area Mill and see if that gives you better results but I imagine those two toolpaths use very similar logic to generate motion.

Off topic, but what on earth machine is this going in? The table and part are beyond huge compared to the work I typically see.

[TFarrell9]

28 minutes ago, Jake L said:

That pretty much sums up why I don't like to use Area Mill or Area Rough. I'll try them to see if I get a good toolpath (no ramps), but often I'll just spend the time to make the extra geometry. Sometimes changing the min/max stepover can get rid of the ramps. But if the toolpath ends up with a blind "pocket" of material to take out I've never been able to stop it from ramping it.

In your sample it looks like the toolpath is morphing kind of weird because it has to go around the .200 high "walls". What I'd do is split that path into two or three paths. Maybe make all the square pads one path, then all the longer rails another path, and the rest a third? That should at least make the toolpath on the pads cut straight along one of the axis.

I don't know of a good way to fix it with a single toolpath. You could use toolpath editor but that comes with it's own pros and cons. And considering it looks like it took almost 3 minutes to regenerate the toolpath , I don't have the time to play with it too much. 

You could try 2D Area Mill and see if that gives you better results but I imagine those two toolpaths use very similar logic to generate motion.

Off topic, but what on earth machine is this going in? The table and part are beyond huge compared to the work I typically see.

Same, I only use this toolpath for finishing faces/floors when I've previously roughed with optirough, since I only have to change a few settings that way. Otherwise, pretty undesirable to use.

I'll try breaking it up to see if chooses to not ramp the couple pads. As far as the diamond-shaped morphing though, it's done that before, just facing a rectangle with some avoidance in the corners, so I'm not sure what really dictates that shape, particularly on a surface with nothing to avoid. 

OT: This is going in a Mitsubishi MVR 35. It recently replaced our old cnc-retrofit planer that had a 8'x40' table. The sacrifice in X travel was greatly compensated for by literally everything else.....the old one was only capable of face/drill/tap and mechanically it was pretty shot, and only 500rpm spindle lol We sent it to the scrap yard.

[bird2010]

if you don't want to change your ways
You can try to add entities (do not add boundaries) to make the software change the algorithm to eliminate ramp

But...it's more correct to separate multiple operations

[Jake L]

17 hours ago, TFarrell9 said:

I'll try breaking it up to see if chooses to not ramp the couple pads. As far as the diamond-shaped morphing though, it's done that before, just facing a rectangle with some avoidance in the corners, so I'm not sure what really dictates that shape, particularly on a surface with nothing to avoid. 

I think the way area rough works is it starts at the "walls" or "avoidance" geometry and takes the max stepover away from that wall until it covers the whole area. If it bumps into itself or other walls then there's more calculations to be done. But I think on your part the toolpath is trying to do this:

Which would explain the diamond shape on the pads. If there's nothing to avoid I think the toolpath just cuts straight along the x axis. Someone please correct me if I'm wrong on any of this.

As for that machine, that thing is awesome. Not sure I'd want to program or run it, but super cool nonetheless, thanks for sharing

 

[TFarrell9]

2 hours ago, bird2010 said:

if you don't want to change your ways
You can try to add entities (do not add boundaries) to make the software change the algorithm to eliminate ramp

But...it's more correct to separate multiple operations

I've created avoidance bodies a fair bit with optirough to drive the shape I want, but I hadn't thought about applying it in this situation to see the effect. Thanks! I'll try breaking it up into a few paths and see if I can get what I want without creating the extra geometry first. Of course the less extra entities I have to create and move around, the better. 

4 minutes ago, Jake L said:

Which would explain the diamond shape on the pads. If there's nothing to avoid I think the toolpath just cuts straight along the x axis. Someone please correct me if I'm wrong on any of this

 

You're right, even some other pads that are in proximity to where its working around corners, it echoes the radius to neighboring pads. Meanwhile, other pads are left with linear, single axis cuts. 

[TFarrell9]

Ended up getting away with breaking it up into two toolpaths and adding some avoidance blocks like Bird did. The pads close in proximity seem to be the where the issue lies, so I put some blocks around them and it came out okay, it's not ramping in anymore at least. 

Sample.mcam