prusa3d / PrusaSlicer

G-code generator for 3D printers (RepRap, Makerbot, Ultimaker etc.)
https://www.prusa3d.com/prusaslicer/
GNU Affero General Public License v3.0
7.73k stars 1.93k forks source link

Solid Infill odd geometry #10677

Open The-Bishop82 opened 1 year ago

The-Bishop82 commented 1 year ago

Description of the bug

Slicing a part at .2 layer height, the solid infill layers display odd geometry. Instead of a rectilinear pattern, it's half rectilinear and half concentric, in an alternating pattern. Visible at layer 184, 183 and so on. Seems to be in all of the smaller solid infill areas.

Project file & How to reproduce

GiviSlider-LayerOdd.zip Screenshot 2023-05-27 084517 Screenshot 2023-05-27 084532

Checklist of files included above

Version of PrusaSlicer

PrusaSlicer-2.6.0-beta3+win64-202305261352

Operating system

Windows 11 22H2

Printer model

Highly Modified Ender 3V2, using Ender 3V2 profile

The-Bishop82 commented 1 year ago

Just to add in, Prusaslicer 2.5.2 does not exhibit this behavior.

Godrak commented 1 year ago

Yes, this is intentional. We use Rectilinear, except for regions where a bunch of subsequent lines would be very short, there we use concentric. This reduces vibrations during internal solid printing, which is usually faster. And especially avoids the zigzag pattern on overhangs, which caused melting, curling and patterns on the surface.

Did you encounter any issue, apart from the fact that it may look slightly odd in the preview? :)

The-Bishop82 commented 1 year ago

No problems with the print, just thought the solid infill layers looked odd as they were different than previous behavior.

The way you explained it makes complete sense, thank you!

The-Bishop82 commented 1 year ago

Learned that this is expected behavior.

5axes commented 1 year ago

Yes, this is intentional. We use Rectilinear, except for regions where a bunch of subsequent lines would be very short, there we use concentric. This reduces vibrations during internal solid printing, which is usually faster. And especially avoids the zigzag pattern on overhangs, which caused melting, curling and patterns on the surface.

Did you encounter any issue, apart from the fact that it may look slightly odd in the preview? :)

Any possibility to activate/deactivate this new behaviour via an option ?

image

secv commented 1 year ago

In my case, the concentric section overlaps too closely causing a big increase in nozzle pressure, which then spits a huge blobs after the concentric section is complete. There's basically 1 or 2 more perimeters at center than necessary.

RickM777 commented 1 year ago

We do have issues other than the different infill pattern. In particular we often see gaps in smaller embedded objects like letters. First layer will be fine, layers 2,3,4 will have the gap. It is primarily an issue for us if we're using translucent filaments for the letters where the gap leads to inconsistent coloring.

We have noticed an improvement in speed in 2.6.0 and appreciate the reduction in vibration just hope the gaps get filled. Images show first layer (perfect) and 2nd layer (with gaps). See the "S" and "2". 1st Layer 2nd layer

Jan-Soustruznik commented 1 year ago

Hello, @RickM777 ,

Could you share a project file(3mf), please?

thank you, have a nice day

RickM777 commented 1 year ago

Hi Jan, I didn't save that project file but I created another one to show the same issue. I tried to remember which letters cause the most problem but it is a combo of size and how much I have "squished" the letters horizontally and at this size most of the letters look ok. I'll attach 2 more similar photos and the project file.

Letter Gap Layer 1 Letter Gap Layer 2

Letter Gaps.zip

As you can see, same issue on layer 2 with the "S" and the "2". The issue presents in the green parts between letters as well though this is not as much of an issue for us as there will eventually be solid green over the gaps so less color discrepancies.

Interestingly if I switch to a standard 0.2 quality preset print settings the issue is less, though the gap remains in the "2" AND it does show one anomaly at layer 4 in the top "S" which is unusual.

Let me know anything else I can do. Cheers.