[Andrew22]

I'm still wrestling with some issues with Fusion360 tool paths and UCCNC. I've enabled smoothing set to .005" or higher and played with CV settings in UCCNC with some success, however one common thread seems to be that UCCNC doesn't want to keep accelerating through multiple lines of G code despite the path looking smooth and straight enough for it to do so. This is all related to 3D surfacing BTW, I haven't had the same problems with 2D yet. I think there may be issues both with post processing and UCCNC, and could really use some help to understand this more. Despite high smoothing values, I don't usually get arc segments in my tool paths, typically its just lots of straight segments. In fusion I've been inspecting the tool paths and number of points in them, and even when they look good there, it's no guarantee it will run smooth on the machine. One strange case I tested was running a spiral morph finish pass and a scallop finish pass made from the same geometry. The spiral morph had more simultaneous 3 axis movement, yet ran fairly smooth @ 6500mm/min. The scallop that had less 3 axis movement, yet it couldn't accelerate past 3000mm/min despite the same feed rate as the spiral morph. Neither one had much stuttering, but the more complex path could accelerate to the set feed rate while the simpler one couldn't. The only culprit I can think of is that the spiral morph path had arcs in it while the scallop didn't. Considering the shape, the scallop path could have easily fit lots of nice arc segments but it didn't. Any ideas why a post processor would ignore arcs?

[cygnus x-1]

I'm still wrestling with some issues with Fusion360 tool paths and UCCNC. I've enabled smoothing set to .005" or higher and played with CV settings in UCCNC with some success, however one common thread seems to be that UCCNC doesn't want to keep accelerating through multiple lines of G code despite the path looking smooth and straight enough for it to do so. This is all related to 3D surfacing BTW, I haven't had the same problems with 2D yet. I think there may be issues both with post processing and UCCNC, and could really use some help to understand this more. Despite high smoothing values, I don't usually get arc segments in my tool paths, typically its just lots of straight segments. In fusion I've been inspecting the tool paths and number of points in them, and even when they look good there, it's no guarantee it will run smooth on the machine. One strange case I tested was running a spiral morph finish pass and a scallop finish pass made from the same geometry. The spiral morph had more simultaneous 3 axis movement, yet ran fairly smooth @ 6500mm/min. The scallop that had less 3 axis movement, yet it couldn't accelerate past 3000mm/min despite the same feed rate as the spiral morph. Neither one had much stuttering, but the more complex path could accelerate to the set feed rate while the simpler one couldn't. The only culprit I can think of is that the spiral morph path had arcs in it while the scallop didn't. Considering the shape, the scallop path could have easily fit lots of nice arc segments but it didn't. Any ideas why a post processor would ignore arcs?

With 3D surfacing tool paths you won't get arc fitting unless the paths are in one of the major planes (XY, XZ, YZ). It's not a Fusion problem, it's a gcode/machine control problem. Higher end controls have very fast block processing speed, which means that paths with lots of tiny line moves are not a problem. They just get smoothed out in the control. It's only on lower end machines (including older industrial machines) where it becomes an issue.

So in your particular case it could just be that the control can't process the code fast enough and therefore never reaches programmed feed rate. But it could also be that the path created by the line segments is not as smooth in one tool path vs another. Specifically I've found that scallop tends to make more jagged paths than other strategies. There is a way to possibly improve it though. Can you post the file with the spiral and scallop paths in it?


C|

[OlfCNC]

An arc is a contant change of movement direction so it could be that your acceleration parameters limit your feedrate.

[daniellyall]

One thing the HSM dev guys recommend is to use I and J and never use R on anything more than 90.

Posting your file always helps there are a few gotchas in fusion that can make a path go a bit wrong