Error occurs when Sweep remove is used on this model.

[ZapKS]

I am trying to use the sweep remove feature to make a groove, but when select the oath and click ok, it says "No Spine linked." I tried it with several different path and different profiles but nothing is working, pls help me out.

[jmaustpc]

Welcome to FreeCAD.


Please post your version data in the manner explained in the forum rules link above this forum.

[chrisb]

That is an intermediate message in 0.18. For a PartDesign AdditivePipe you select the (first) section, invoke the tool and then you have to select the path. The message says that you haven't yet done the latter.

[edwilliams16]

I noticed a few problems with your file.

Part|CheckGeometry on your additive pipe gave numerous errors. This was based on two sketches, neither of which was fully constrained, likely causing the problems.
Your oval sketch, IMO, would have been better constrained using symmetry constraints. Anyhow, when I redrew those sketches, they did the additive pipe fine.
Cutting the groove in the top doesn't require sweeping along a circular path as the simpler groove tool does this more simply by rotating the sketch around the z-axis. Also, cutting with a circular sketch instead of the D-shape automatically avoids the coplanarity problem on the bottom surface of your part - you don't have to kludge this by offsetting the D by a few microns as you did.

I enclose a version which at least passes the check geometry tests.

[chrisb]

I extracted from your model a sketch which may be interesting for Abdullah.

ping

Here is another example of an overconstrained sketch. I can add it to the issue #4538, but it may be something different, as this one uses symmetry and shows never an overconstrained message.

Steps to reproduce:
- Open the file overconstrained.FCStd
- Constraint11 can be dropped,
-> the sketch is still fully constrained

[chrisb]

As the previous post was rather off topic, I enclose a version of the path how I would have it constrained using only one symmetry and do the rest with vertical, horizontal and equalities.

[edwilliams16]

Chrisb

How did you figure my oval sketch was over-constrained? As you suggested, I never was warned during construction, and it worked...

[chrisb]

How did you figure my oval sketch was over-constrained? As you suggested, I never was warned during construction, and it worked...

I didn't supect it to be overconstrained. I had seen the symmetries, and wanted to check if they could be replaced by "cheaper" constraints.

[abdullah]

I enclose a version

I extracted from your model a sketch which may be interesting for Abdullah.

ping

Here is another example of an overconstrained sketch. I can add it to the issue #4538, but it may be something different, as this one uses symmetry and shows never an overconstrained message.

Steps to reproduce:
- Open the file overconstrained.FCStd
- Constraint11 can be dropped,
-> the sketch is still fully constrained

This one is indeed different, because the solver is very much aware of the partial redundancy, but it is being tolerant with the user and letting it through.

Code: Select all

14:31:21  EigenSparseQR, Threads: 1, Vectorization: On, Pivot Threshold: 1e-13, Params: 61, Constr: 63, Rank: 61
14:31:21  Analysing groups of constraints of special interest:
[5 4 6 0 0 0 0 11 11 6 5 2 10 10 2 1 1 8 8 9 9 ]
[13 5 7 3 4 0 13 0 0 0 0 3 0 0 11 11 6 7 5 4 2 10 10 1 1 8 9 13 0 ]
14:31:21  DL: tolg: 1e-80, tolx: 1e-80, tolf: 1e-10, convergence: 1e-10, dogLegGaussStep: FullPivLU, xsize: 61, csize: 62, maxIter: 100
14:31:21  DL: stopcode: 1, Success
14:31:21  Sketcher::RedundantSolving-DogLeg-
14:31:21  Sketcher Redundant solving: 1 redundants
14:31:21  (Partially) Redundant, Group 1, index 14, Tag: 11
14:31:21  (Partially) Redundant, Group 0, index 7, Tag: 11

This one would go well in the ticket about the "popularity contest". Which is something I would like to check during 0.20 too.

Let's say that this one I know why it is happening. The other one is much more dangerous because I do not have a clear understanding of why it could be happening (without having looked into the matrices, ranks, dependent rows,...).

[chrisb]

This one would go well in the ticket about the "popularity contest". Which is something I would like to check during 0.20 too.

Added to issue #4387.