Geometrical errors in 0.70b with Slic3r 0.9.1

... when slicing with support and higher resolution I'll get some geometrical errors in the viewer, that had no representation in the G-code, but shows lines starting and ending way outside the printing area:


Edited 2 time(s). Last edit at 08/16/2012 10:17AM by VDX.

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

I had similar problems in the past, where filament made a 180 degree turn causing numerical problems computing the intersection of the filaments. Could be something similar. If you could send me the slic3r settings exported and the stl or the g-code, I will check where the numeric problem comes from and I hope to see them. In the past I learned that some machines (like mine) often tend to remove such objects while older xp computers showed them. I guess they have different handling of NaN values. But in any case, thats something I like to have removed.

---

Repetier-Software - the home of Repetier-Host (Windows, Linux and Mac OS X) and Repetier-Firmware.
Repetier-Server - the solution to control your printer from everywhere.
Visit us on Facebook and Twitter!

... attached the Slic3r-configs - the STL and G-Code file are nearly 10MB zipped, no chance to upload

The G-Code-lines with this errors had the same coords and values as the previous and following.

When at home, I'll try again and make a screen copy of the lines with this errors ...

*** EDIT: - this machine is a DualCore notebook with Win7 Ultimate, at home I'll test with a DualCore desktop and Win7 Home Premium ...

Edited 1 time(s). Last edit at 08/16/2012 11:01AM by VDX.

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Ok, tried some objects and got one with the same error. It was my old friend, the 180 degree turn. Looks like slic3r still inserts it sometimes. The next release will limit the error to 10 times the diameter. The result is still a bit undetermined but that's a fact of an impossible print. Just for the interest, here the code causing the trouble:

G92 E0
G1 X100.468 Y113.622 F7800.000
G1 X101.314 Y111.536 F3600.000 E2.03519
G1 X101.203 Y111.808 E2.03919

---

Repetier-Software - the home of Repetier-Host (Windows, Linux and Mac OS X) and Repetier-Firmware.
Repetier-Server - the solution to control your printer from everywhere.
Visit us on Facebook and Twitter!

... here two connected lines provoking this error (the two yellow highlighted lines):

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Hard to say, if it is the problem without seeing the code, but it looks like the same problem I found. It was also a connected line, where the second line goes back on the same path it came from.

---

Repetier-Software - the home of Repetier-Host (Windows, Linux and Mac OS X) and Repetier-Firmware.
Repetier-Server - the solution to control your printer from everywhere.
Visit us on Facebook and Twitter!

... I think it's only the graphical representation of a nested connected tube, what's causing this 'endless' lines.

When you draw connected tube segments with a really small enclosing angle, then the outer connection will be somewhere ... far ... far ... away ...

When printing, this won't cause big problems, as it seems to be only very short line segments, that are overlapped this way - it's only in the simulation view, where it causes this ugly effect ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Exactly. If you switch to simple line representation it is even viewed correct. It's only the visualization of the filament that gets real problems. The solution for the next release is to virtually limit the angle, so that only 10*filament width is possible in the edges.

---

Repetier-Software - the home of Repetier-Host (Windows, Linux and Mac OS X) and Repetier-Firmware.
Repetier-Server - the solution to control your printer from everywhere.
Visit us on Facebook and Twitter!