Solid Layer & Anomalies in Hole of Object when .20mm Layer Height [1.41.1-beta]

[wsmagee]

Version

1.41.1-beta-win32

Operating system type + version

Windows 10 - 32bit

Behavior

When I slice a spool shaped object it appears to slice fine if .15mm or less layer height, but when on the .20mm layer height anamolies appear including a solid layer in the middle of the object.
I reverted to 1.41.0 and problem went away. So this is new in the beta release.

Expected results would be this:

.20mm layer height looks like this:
It has the full layer that should not be there.

And it has a couple funny anomalies(the yellow parts)

.15mm layer looks correct, like this:

STL/Config (.ZIP) where problem occurs

STL-And-Config.zip

[MarcoZ76]

The same behavior i have with mac os Mojave. Please solve it ASAP

[bubnikv]

@MarcoZ76 Would you please provide your data? Thanks.

[bubnikv]

@wsmagee Your model has flipped normals. The model slices correctly after fixing with the netfabb service (for Win10 users, the netfabb service is integrated into Slic3r).

Looking at the incorrectly sliced model, it seems there are some zero thickness protrusions inside the model, which are not shown in the model view? I somehow cannot imagine how Slic3r could hallucinate these protrusions. Most likely they are there in the model.

[wsmagee]

@bubnikv . Ok, good to know, I will reload 1.41.1-beta and attempt to fix with the netfabb service.
I am however wondering what has changed from previous versions of slic3r to this one that would make these anomalies appear. I have used that same .stl model in multiple past versions of slic3r and this is the first time I have had it show up. Was something auto-fixing the errors in past versions different than this beta?

[bubnikv]

I was trying to fix exactly this kind of anomalies. I am now testing yet another logic, which is much simpler than the previous attemps, but which should be more robust.

…

On Mon, Oct 1, 2018 at 4:59 PM Wayland Magee ***@***.***> wrote: @bubnikv <https://github.com/bubnikv> . Ok, good to know, I will reload 1.41.1-beta and attempt to fix with the netfabb service. I am however wondering what has changed from previous versions of slic3r to this one that would make these anomalies appear. I have used that same .stl model in multiple past versions of slic3r and this is the first time I have had it show up. Was something auto-fixing the errors in past versions different than this beta? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#1265 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AFj5I4y-y_oopTCdEffcUYDkOEq6GjR8ks5ugi3bgaJpZM4XAIqK> .

[MarcoZ76]

@MarcoZ76 Would you please provide your data? Thanks.

The stl file come from Gregsaun github (Grégoire Saunier) , is the new X axis for Prusa MK3.
X axis.zip

My operation system is Mac Os 10.14 Mojave
Version Slic3R PE 1.41.1 beta.

I tried to slice with Slic3r PE 1.41.0 and all is sliced correctly but with this version many stl i sliced comes bad.

[bubnikv]

This is a regression issue, which will be closed in 1.41.1-rc.
Closing.