Concentric infill

I built a simple pen plotter at home, and I’m finetuning it. It uses GRBL on an Arduino Uno.
Laserweb seems to work quite well, when using the laser mode. Tool path is much more intelligent than LaserGRBL for instance, and the software isn’t as complicated and cumbersome as GRBL-plotter.

Anyway, I’m more familiar with 3D printing. I see I can laser cut a path, and do infill, and can lasercut on the inside or outside of the path. Very nice. Infill by horizontal scanning seems quite slow, and it’s not that great if the Z-axis goes up and down like a yo-yo all the time. Isn’t there an option to do the infill in concentric patterns, like you see on a 3D printer?
I can iterate the “laser cut inside” tool several times, changing the margin every time, but that’s very cumbersome.

I believe what you’re referring to is a pretty interesting fill algorithm called Connected Fermat Spirals. See the paper, “Connected Fermat Spirals for Layered Fabrication” - https://homes.cs.washington.edu/~haisen/CFS/index.html Basically what this algorithm does is fill a given space with a fermat spiral in such a way that the start and end positions are looped. This means that any other spacefilling spirals are also loops and you can always just link loops together to create a Eulerian path. This means that these types of spacefilling curves have a bunch of properties that are really good for 3d layered fabrication. Namely, they do not have sharp turns. You aren’t rastering to one side then jerking directly backwards. That is hard with regard to filament, it’s also kind of time consuming as part of the laser, since it triggers and acceleration and deceleration cycle. With regard to embroidery it would allow filling of arbitrary space with a sort of spiral stitch linking up all contiguous regions in fairly interesting ways. Given my background in both laser cutter design and embroidery, I think they are pretty awesome.

However, the short answer is no. Nobody has done anything with these types of curves and lasers.

You might be able to coax a slice out of 3d printer or get some application somewhere to give you either a gcode layer or a svg of the spiral and try running it on a laser and see if it actually has nice properties for lasers too. But, again, no there isn’t an option for such a thing. I don’t think there’s an option for that in any laser cutting software.

1 Like
1 Like

That’s fascinating. Maybe one day someone will code it.

I’m still a little confused by this:

Do you have a 3d object that you are engraving on top of?

Hmm, no. I meant the pen going up and down all the time, not sure if that qualifies as z-axis then.

Oh, my confusion. I missed the “pen plotter” part and was thinking laser. My mistake…

It seems like the main idea behind this could be tested regardless of the code behind this. Just convert a single slice of a slicer into SVG. Then run that code faster than your laser usually runs, especially on an M2 Nano or anything else that’ll run too fast for its own good.

You can then compare that slice to the same thing in a the more typical boustrophedon pattern.

1 Like