gcode generation: corrupt walls with 'Spiral vase' enabled #2761

klpeders opened this Issue Mar 28, 2015 · 5 comments


None yet

2 participants


The generated gcode showing strange cross instead of a thin wall.

Slic3r version

Slic3r 1.2.6 Prebuild binary for Linux 64bit

Operating system type + version

Fedora 21, Intel 64 bit

Steps to reproduce the issue, including:

  • Generate STL in OpenSCAD:
    $fs = 0.1;
    $fa = 5;
    union() {
    linear_extrude(height=8) scale(7)
    text("6", font="serif:style=Heavy");
    linear_extrude(height=5) scale(2.8) translate([17.5,2.2,0])
    text("号", font="Source Han Sans CN:style=Heavy");
  • Expected result
    slic3r_1 2 6
  • Actual result

Slic3r configuration file

I had to add .gif to be able to upload the config and stl files:
slic3r_thin-wall-extrude-config txt

no6 stl xz

alexrj commented Mar 28, 2015

I was able to download your config, but the STL file is corrupt. I guess GitHub processes it as an image file. Can you host it somewhere and put the link here?

@alexrj alexrj added this to the 1.2.7 milestone Mar 28, 2015

It works for me:

$ wget https://cloud.githubusercontent.com/assets/11691683/6880344/dbd738f6-d566-11e4-87a4-f058eaa721e3.gif -O no6.stl.xz
$ unxz no6.stl.xz
$ ./bin/slic3r no6.stl.stl 
=> Processing triangulated mesh
=> Exporting G-code to ./no6.gcode
Done. Process took 0 minutes and 1.245 seconds
alexrj commented Mar 29, 2015

You didn't tell me it was XZ compressed...

alexrj commented Mar 29, 2015

Okay, thank you, I can reproduce this.

@alexrj alexrj added a commit that referenced this issue Apr 12, 2015
@alexrj Fixed regression introduced by the recent PerimeterGenerator refactor…
…ing causing spiral vase not to be correctly skipped on multi-loop layers. Includes regression test. #2761
alexrj commented Apr 12, 2015

Fixed. This model is not suitable for spiral vase, and Slic3r wasn't detecting this correctly (anymore).

@alexrj alexrj closed this Apr 12, 2015
@alexrj alexrj added the Fixed label Apr 12, 2015
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment