Who uses triangular mesh tool paths for 3 or 4 axis parts?

Started by neurosis, January 08, 2021, 09:04 AM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

Matthew Hajicek

I used Metacut Utilities for a while.  I believe support was dropped many years ago.

Vericut has filtering capabilities, but they're a bit pricey.

CNCAppsJames

I used to use MetaCut BITD when I was an AE at Mori Seiki in the mid-90's. It was the only utility available that would take point to point G-Code and convert it to NURBS g-code (G06.2). At the time IIRC UG, SurfCAM, and like 1 or 2 (max) other CAM systems could produce that code. NURBS ouperformed all the other code flavors/options in terms of accuracy, surface quality cycle time, etc.... It required a RISC processor in the machine, as well as HPCC (G05P10000), and a Dataserver was strongly reccommended because RS-232 is dog shit for DNC, and the 16i was not yet on the market so poor man's Dataserver wasn't an option yet either.

Today AICC-II and the handful of other options that usually get combined with it outperform what NURBS was capable of BITD.

As far as the "verification" of MetaCut... I would say it was vastly superior to Mastercam's verify. Both in terms of speed and image quality. Not sure how it would compare today. NorthWoods is no longer around and it's been out of development for a very long time.

I don't use the ModuleWorks 5-Axis toolpaths. Mastercam's implementation does not include Cutter Compensation so they are useless to me. That's why I continue to use the Old-School 5-Axis toolpaths (Flow, Curve, and Swarf mostly). All of our 5-Axis machines have 3D Tool Comp (G41.2/G42.2) and I use it... on just about every 5-Axis path.

That was the last feature request I submitted to Mastercam. That was like 5 years ago. Maybe even longer now. I thought it might get some traction since one of the guys on the Dev. Team reached out to me like last year. He asked why I stopped submitting enhancement requests. I told him it was a complete and total waste of my time. Even the low hanging fruit CNC refuses to add. FWIW, the Cutter Comp "enhancement" is nothing more than CNC software flipping a switch to enable it. It's already built into ModuleWorks, CNC just chooses to not enable it. So... no more requests. I use what I use and leave the enhancement requests to the n00bs who they like to listen to more than us old Salty AF programmers anyway.
"That bill for your 80's experience...yeah, it's coming due. Soon." Author Unknown

Inventor Pro 2026 - CAD
CAMplete TruePath 2026 - CAV and Post Processing
Fusion360 and Mastercam 2026 - CAM

crazy^millman

Quote from: CNCAppsJames post_id=4604 time=1611536356 user_id=62I used to use MetaCut BITD when I was an AE at Mori Seiki in the mid-90's. It was the only utility available that would take point to point G-Code and convert it to NURBS g-code (G06.2). At the time IIRC UG, SurfCAM, and like 1 or 2 (max) other CAM systems could produce that code. NURBS ouperformed all the other code flavors/options in terms of accuracy, surface quality cycle time, etc.... It required a RISC processor in the machine, as well as HPCC (G05P10000), and a Dataserver was strongly reccommended because RS-232 is dog shit for DNC, and the 16i was not yet on the market so poor man's Dataserver wasn't an option yet either.

Today AICC-II and the handful of other options that usually get combined with it outperform what NURBS was capable of BITD.

As far as the "verification" of MetaCut... I would say it was vastly superior to Mastercam's verify. Both in terms of speed and image quality. Not sure how it would compare today. NorthWoods is no longer around and it's been out of development for a very long time.

I don't use the ModuleWorks 5-Axis toolpaths. Mastercam's implementation does not include Cutter Compensation so they are useless to me. That's why I continue to use the Old-School 5-Axis toolpaths (Flow, Curve, and Swarf mostly). All of our 5-Axis machines have 3D Tool Comp (G41.2/G42.2) and I use it... on just about every 5-Axis path.

That was the last feature request I submitted to Mastercam. That was like 5 years ago. Maybe even longer now. I thought it might get some traction since one of the guys on the Dev. Team reached out to me like last year. He asked why I stopped submitting enhancement requests. I told him it was a complete and total waste of my time. Even the low hanging fruit CNC refuses to add. FWIW, the Cutter Comp "enhancement" is nothing more than CNC software flipping a switch to enable it. It's already built into ModuleWorks, CNC just chooses to not enable it. So... no more requests. I use what I use and leave the enhancement requests to the n00bs who they like to listen to more than us old Salty AF programmers anyway.


James with CAMPLETE that ability could be added pretty easily I would think. I wouldn't even think about trying to machine some of what I am machining today on 5 Axis machines without the MW toolpaths. I could get it done, but my 1100 hour project would be close to 3000 if I were doing it the way your going about it. The 4 Axis roughing we did for that one customer was MW Multiaxis rough and that turned out sweet.

Agree implementation of basic things seems to be slipping hopefully change will come and the bugs get attention I have been hearing they will get for almost a decade now. Not a noob and I still send my enhancement requests someone has to keep trying it what I am thinking.

CNCAppsJames

If only CNC would pick some low hanging fruit every once in a while. It would just make things SO much simpler for EVERYONE involved.
"That bill for your 80's experience...yeah, it's coming due. Soon." Author Unknown

Inventor Pro 2026 - CAD
CAMplete TruePath 2026 - CAV and Post Processing
Fusion360 and Mastercam 2026 - CAM

TylerBeer

The cutter comp in multiaxis thing was frustrating to me as well. Here I am, agreeing with James!

the moduleworks paths sometimes crunched mesh paths better than the MCAM toolpaths so I HAD to use them locked down to 3ax. It's been awhile but I remember a 2gb scan that would just crash unless I used the 5ax paths.