Missing infill in version 1.1.5 #2144

NinjaCross opened this Issue Jul 2, 2014 · 12 comments


None yet

6 participants


I'm having an issue with version 1.1.5.
It seems that in some cases the infill is not generated in small linear parts like the one in the pictures.



Here are the files to reproduce/analyze the issue:

harriv commented Jul 3, 2014

Did you try to print it? AFAIK, Repetier-Host doesn't visualize differences in flow rate.


Yes, I converted the GCODE into X3G and printed it using a Wanhao Duplicator 4, and the infill was missing.
The very strange thing is that, after posting this issue, I regenerated again the GCODE (without making ANY modification... I just forgot to save it on file the first time), and the second time the infill has been generated !


i do confirm. i get same missing parts of bottom/top infill , in actual prints (printing with plain i3)

@alexrj alexrj modified the milestone: 1.1.6, 1.1.7 Jul 23, 2014
alexrj commented Jul 26, 2014

@NinjaCross, what layer is that? I can't recognize the part in the screenshot in the file Guscio_laterale_XS.stl

harriv commented Jul 30, 2014

I can see same problem, I'm missing parts of bottom infill with concentric infill.



Command line from Repetier-Host:
C:\Program Files\Repetier-Host\Slic3r-1.1.6\slic3r.exe --load "C:\Documents and Settings\harriv\Local Settings\Application Data\RepetierHost\slic3r.ini" --print-center 74,126 -o "C:\Documents and Settings\harriv\Local Settings\Application Data\RepetierHost\composition.gcode" "C:\Documents and Settings\harriv\Local Settings\Application Data\RepetierHost\composition.obj"

Files here: https://drive.google.com/folderview?id=0Bw1VY8X1WZhdT0hrcWo3TEJVUzA&usp=sharing


@alexrj As far as I can remember, all the layers where missing the infill, including the top and bottom ones.
I've attached a global image that shows the part where the problem is.


alexrj commented Jul 30, 2014

Then you forgot to say that you rotated the model... I need full steps to reproduce issues.


I see, sorry for the imprecision.
Is it clear now or do you need for more details ?

lvperlo commented Jul 30, 2014

I have a problem with 1.1.5 that could be related to this one.
The vertical walls in my model are 1.5mm thick. I print using a nozzle of 0.5mm. So there should be 3 traces to fill the wall. However, the trace in the middle is left out in many cases, creating hollow walls. I just tried another simple model with walls of 1mm, 1.5mm amd 2mm and that one prints fine. The problem seems to depend on the complexity of the model. Update to 1.1.6 does not make a difference.
Previously with 1.0.1, the same model printed well, with solid walls.
If you think it's not related or need details, let me know.

jeder73 commented Jul 31, 2014

@lvperlo: I dont think you can simplify it that way: "The vertical walls in my model are 1.5mm thick. I print using a nozzle of 0.5mm. So there should be 3 traces to fill the wall."
It depends on the actual configured width for perimeters, if you use gap fill or not, of course the layer height, and so on...I don't even think, it is possible to create an odd number of perimeters...

lvperlo commented Jul 31, 2014

@jeder73: Exactly the same .stl prints fine when gcode generated with Slic3r 1.0.1. A simple model with 1.5mm wall and exactly the same Slic3r parameter settings prints 1.5mm solid walls with 3 traces.

alexrj commented Aug 3, 2014

I was able to reproduce this and fix it. I also added a regression test to cover this issue. Thank you

@alexrj alexrj closed this Aug 3, 2014
@alexrj alexrj added Fixed and removed Needs details labels Aug 3, 2014
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment