[Solved] Broken view with section

[panda1967]

Hi.
A shape seems to broke the projection mechanism. As you can see, the section A-A is broken.

Istantanea_2020-08-06_20-24-12.png (41.03 KiB) Viewed 1763 times

I tried to attach the FreeCAD file but the system rejected it because too big (it is 1 MB). In this file there is only the 3D model and a page with two viewes. I don't know how to reduce the file.
Thank you.
Massimo

This is my system:

OS: Ubuntu 18.04.4 LTS (XFCE/xfce)
Word size of OS: 64-bit
Word size of FreeCAD: 64-bit
Version: 0.19.
Build type: Release
Branch: unknown
Hash: e813c5adfdb48f4b4c27966cb8b04948e97573d2
Python version: 3.6.9
Qt version: 5.9.5
Coin version: 4.0.0a
OCC version: 7.3.0
Locale: Italian/Italy (it_IT)

[Gregory son of Carl]

This seems to happen when the Projection direction and Section Normal vectors are not aligned. I was able to get a similar effect by changing the Section Normal Y value from 1 to -1.

WhereTheLinesAt.png (78.14 KiB) Viewed 1735 times

OS: Debian GNU/Linux 10 (buster) (KDE/default)
Word size of OS: 64-bit
Word size of FreeCAD: 64-bit
Version: 0.19.22039 (Git) AppImage
Build type: Release
Branch: master
Hash: 2bfc6301bc80c0344cbf13dbfe041fbd78cac93d
Python version: 3.8.2
Qt version: 5.12.5
Coin version: 4.0.0
OCC version: 7.4.0
Locale: English/United States (en_US)

[wandererfan]

I tried to attach the FreeCAD file but the system rejected it because too big (it is 1 MB).

You can use Dropbox, GoogleDrive, OneDrive, etc to share larger models.

Direction (for the shape) and SectionNormal (for the section faces) have to match. Changing 1 without the other will lead to the situation you describe.

If the model was created in v0.19 and you did not change Direction or SectionNormal properties, then this should be investigated as a possible bug.

[Gregory son of Carl]

Changing 1 without the other will lead to the situation you describe.

Is there a reason they are kept as separate properties? It sounds like both ought to be merged into 1 property if they need to match in order to get a working section view.

[panda1967]

I have put the FreeCAD file into DropBox. Here's the link:
https://www.dropbox.com/s/qzx6gexumib21 ... FCStd?dl=0

If the model was created in v0.19 and you did not change Direction or SectionNormal properties, then this should be investigated as a possible bug.

In this case the model was created in v0.19 and I did not change Direction or SectionNormal.

Note that the error happens only when the y component of Section Origin is 17 and 21; to be more precise, when the section plane falls into one of the two little holes.
Massimo

[wandererfan]

Note that the error happens only when the y component of Section Origin is 17 and 21; to be more precise, when the section plane falls into one of the two little holes.

Please try this:
Edit>Preferences>TechDraw>Advanced - check Draw Section Edges
(Modifica>Preferenze>TechDraw>Avanzato - Consenzi Sovascrivi Pagina)

Close FreeCAD
Open FreeCAD
Load Test.FCStd

Let me know what happens.

Grazie.

[panda1967]

I've set:

Edit>Preferences>TechDraw>Advanced - check Show Section Edges

01 - Istantanea_2020-08-09_18-27-41.png (124.26 KiB) Viewed 1537 times

What happens is a strange image in which the inner geometry appears somehow, but in general the section view is not as expected.

01 - Istantanea_2020-08-09_18-18-20.png (184.93 KiB) Viewed 1537 times

To get an idea, I attach also the image of the 3D model with a clipping plane perpendicular to the axis y, at y=17:

01 - Istantanea_2020-08-09_18-16-39.png (139.54 KiB) Viewed 1537 times

The following image is about the section at y=18; it seems ok!

01 - Istantanea_2020-08-09_18-48-40.png (187.96 KiB) Viewed 1537 times

Massimo

[wandererfan]

The following image is about the section at y=18; it seems ok!

There is something odd about the Body. If I set the Tip to Mirror020, the section cut works, but if I set the Tip to Hole013 the cut doesn't work. If I cut the body with a Part Cube, the cut operation complains about non-solid input.

I don't see anything obviously wrong in the Body, but the Dependency Graph looks very convoluted to me. Maybe one of the power-users can have a look at the model.

ping

ping

TipAtMirror020.png (30.57 KiB) Viewed 1479 times

TipAtHole013.png (30.56 KiB) Viewed 1479 times

cutWithCube.png (39.36 KiB) Viewed 1479 times

[chrisb]

I will have a look later.

[chrisb]

The file is not really big, just 1 MB; yet it loads already for 3/4 of an hour. Do you experience the same? Is it possible that you have a cyclic dependency? Can you check with the Dependency Graph or DAG view?