Quick air job, which only confirms that the jagged portions along the toolpath end up slowing down the movement
(this was during a random portion of the toolpath, while air-v-carving the left side of the “B” in “Braves”)
Quick air job, which only confirms that the jagged portions along the toolpath end up slowing down the movement
(this was during a random portion of the toolpath, while air-v-carving the left side of the “B” in “Braves”)
Here is the downloadable link…
Well there is your problem. You are making a Braves sign and not an Astros sign.
The nodes in the file seem reasonable:
so this would seem to be a problem in how Carbide Create calculates the toolpaths which @robgrz will have to speak to.
The output from CC has moves that are as small as the precision of the output, 0.01mm. This is less than a step. I’m not sure how GRBL handles that (I’m sure there’s a mechanism as I think I remember reading it somewhere), but there’s definitely something up with those Advanced V-Carve toolpaths.
This topic was automatically closed after 30 days. New replies are no longer allowed.
Neil: would you suggest that a post processor should do basically rounding to whole steps?
(and eliminate no-move moves)
that’s an interesting idea and likely not all that hard to implement
I think that makes sense. Less for the tiny MCU to deal with.
The only issue I see would be with the varieties of steps/mm values used with the same controller. There is such a variety of machines using the same controller…would you make this a part of a post-post processor?
at least CC has a separate post proc between shapoko / nomad / generic
I’ll note that if you do that for a curve approximated using straight line movements you’re likely to get a more jagged result, depending on how rounding errors fall out.