Artefacts when using a linear pattern

[mfraz74]

Code: Select all

OS: Ubuntu 22.04.1 LTS (KDE/plasma)
Word size of FreeCAD: 64-bit
Version: 0.21.0.31709 (Git) AppImage
Build type: Release
Branch: master
Hash: e188802ca6997d2564e7570ab648462e6a059f87
Python 3.10.8, Qt 5.15.6, Coin 4.0.0, Vtk 9.1.0, OCC 7.6.3
Locale: English/United Kingdom (en_GB)
Installed mods: 
  * sheetmetal 0.2.61
  * freecad.gears 1.0.0
  * Manipulator 1.5.0
  * lattice2 1.0.0
  * fcgear 1.0.0
  * Lithophane
  * Curves 0.6.5
  * Assembly4 0.12.4
  * InventorLoader 1.3.0
  * fasteners 0.4.54
  * A2plus 0.4.60i
  * QuickMeasure 2022.10.28
  * ThreadProfile 1.84.0
  * Assembly3 0.11.4
  * VendorParts

I've created a sketch and revolved it in the Part Design WB. I've then added a pocket and used a linear pattern to create multiple pockets, but as soon as I create the linear pattern artefacts are visible in the model and when exported.

Screenshot_20230131_113744.png (136.22 KiB) Viewed 1165 times

[domad]

....

Hi mfraz74, hello to the Community!
I confirm the anomaly, I tried the export in *.stl and opened in Blender and Meshlab and the models have less evident anomalies in the same place but different in the conformation, however, leaving the angular deflection at 28° and bringing the deviation to 0, 02 anomalies also disappear in exports.
I enclose the modeling performed in "Part" using the original sketches interfaced with the "DynamicData" Wb to perform some dynamic changes in real-time, I have not encountered any problems.
I am also attaching the original file with the diversion setting reset as previously indicated which does not show any anomalies.

[papyblaise]

there is coplanar: the radius of the sketch are the same as those of the 1st body, increase them by 0.05mm

[mfraz74]

there is coplanar: the radius of the sketch are the same as those of the 1st body, increase them by 0.05mm

I've removed that sketch and just created a padded circle, it's still showing artefacts.

Screenshot_20230131_223917.png (74.04 KiB) Viewed 981 times

[papyblaise]

I modified the sketch of the windows and its position, and that of the teeth, reversed the order of the pockets and the linearPatern, the result is good, but I did not identify which disturbs the site

[mfraz74]

I modified the sketch of the windows and its position, and that of the teeth, reversed the order of the pockets and the linearPatern, the result is good, but I did not identify which disturbs the site

Still seeing artefacts

Screenshot_20230201_111300.png (80.16 KiB) Viewed 828 times

Do I report this as a bug?

[papyblaise]

I don't know what to say anymore, I don't see any at home and yet my version is older
someone else would have to confirm (Schlm en dit-il plus )

[Shalmeneser]

Body > Property View > Deviation = 0.5 for better quality ?

[FCuser2019]

Hello, unfortunately the tessellation algorithm has serious problems in certain cases, this is an example.
Using the Draw Style "Hidden line" you have the following situation:

Original_dev020.png (100.11 KiB) Viewed 752 times

By decreasing the deviation to 0.1, the defect apparently disappears:

AsIs_dev010.png (58.45 KiB) Viewed 752 times

but in reality it is still present:

HL_dev010.png (67.81 KiB) Viewed 752 times

decreasing the deviation to 0.05, in the hope of obtaining correct tessellation throughout the body, gives the following horrible result:

HL_dev005.png (71.59 KiB) Viewed 752 times

HL_dev005_b.png (64.43 KiB) Viewed 752 times

Unfortunately, there are cases where no deviation value can correct the incorrect tessellation!

[papyblaise]

the mystery deepens: if I do the 1st revolution at 179° the artefact increases, if I do it at 181° all is well