Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[KratosCore] Bug in Prism3D15N geometry #9818

Closed
dbaumgaertner opened this issue Apr 1, 2022 · 2 comments · Fixed by #9882
Closed

[KratosCore] Bug in Prism3D15N geometry #9818

dbaumgaertner opened this issue Apr 1, 2022 · 2 comments · Fixed by #9882

Comments

@dbaumgaertner
Copy link
Member

Description
The prism3D15N geometry seems to be buggy. Here (test_Prism3D15N.zip) you can find a minimal test case. It is a structural modal analysis with one SmallDisplacementElement3D15N which uses this prism. The single included element follows the Kratos node ordering specified here. However, if one runs the analysis, it throws an error saying that the element base class is called. From the other output, it seems that something is very wrong with the computation of the Jacobian or the shape function (derivatives).

Now, I checked, if there is a test for this geometry. However, it seems not. In fact, a corresponding test was once committed but then commented out. @loumalouomega from the git history, I noticed that you commented it out saying in the title of the corresponding commit that "Prism3D15 [is](not working yet, the element has some problems to be fixed in the future)".

So my questions is. Did anybody experience this bug before or know a solution for it? It would be great if anyone involved in the geometry implementation could have a look into the element and check if there is something fishy.

Side note: If one changes the prism3D15N to a prism3D6N, the minimal example attached runs again without errors (relevant mdpas are provided). So it just seems to be a bug of the quadratic prism (and not the solver / model setup, etc...).

Scope

  • KratosCore

To Reproduce
Minimal example attached

Expected behavior
The simulation (a modal analysis) should run through and print the eigenvalue of the first mode

Environment

  • OS: Windows
  • Branch: master
  • Python 3.9.7
@loumalouomega
Copy link
Member

The mythical TODO, maybe I wrote that, if git says so. We can check the problem. The easiest way is with a simple patch test, that can be mono-element.

@loumalouomega
Copy link
Member

The mythical TODO, maybe I wrote that, if git says so. We can check the problem. The easiest way is with a simple patch test, that can be mono-element.

From the quoted test a priori looks like an ordering issue, I will check if I can

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging a pull request may close this issue.

2 participants