LinkDaily and spreadsheet parameters

[wsteffe]

The annexed file is a small test which extrudes a sketch.
The extrude started to fail after I have redefined a few sketch dimension with parameters taken from a spreadsheet.
Have I done something wrong or is it a bug ?

[Shalmeneser]

In PartDesign, a body need to be a continous connected form.
With 2 forms, your body is illegal.

[wsteffe]

In PartDesign, a body need to be a continous connected form.

It is not required in the RT Linkdaily branch which I am using.
In fact the discontinuous body was created (and is still visible) before introducing the parametric dimensions.

[Shalmeneser]

Code: Select all

 Origin "CMP_SFILT_6_4#Origin001" doesn't contain feature with role "XY_Plane"

Copy everything in a new PART with correct axis or plane.

[chrisb]

Moved to the right forum where this program is discussed. We ask for the FreeCAD infos not just for fun, but rather to avoid such misunderstandings.

[wsteffe]

This issue was posted also at https://github.com/realthunder/FreeCAD/issues/174

RT has concluded that it should be a sketch issue in the upstream. So, following his suggestion, I have simplified the test case and removed all upstream incompatibilities. Now the sketch should contain a single closed curve. The Container Part was removed due to the unsupported delayed origin.

Here annexed the simplified test case.

Chrisb, may you please move again this thread to a more appropriate forum section.
I do not think that Assembly is the write place to discuss a sketch issue.

[GeneFC]

The sketch is broken. The upper-left joint between the arcs is not closed. Probably because Point-on-Line constraint was used instead of Tangent.

When I cleaned up the sketch it padded fine.

Gene

[wsteffe]

The upper-left joint between the arcs is not closed. Probably because Point-on-Line constraint was used instead of Tangent.

A tangent contraint is already present in the upper left corner but this is not the problem. What is missing is a point-point concidence constraint.
I think that several point-point concidence constraints were automatically inserted when I selected the strarting point of an arc in coincidence with the end point of the previous arc. But probably this kind of constraint was not set at the last point (which should be the upper left corner).

Sorry for having wronlgly thought about a possible bug.
I was confused by the fact that sometimes the sketch resulted closed and other times not depending on the values of the parametric dimensions.

[chrisb]

A tangent contraint is already present in the upper left corner but this is not the problem. What is missing is a point-point concidence constraint.

You should not use line-to-line tangency plus coincidence. Use the proper tangency variant instead.

[wsteffe]

Use the proper tangency variant instead.

Yes thanks. In fact when I tried to add a coincidence FC automatically replaced it (and simple tangency) with proper tangency variant.