2
$\begingroup$

enter image description here

I did this simple analysis on RFEM, of a rigid-supported beam loaded with a point moment. Before analysis, I didn't assign any kind of mesh manually. When I turn on the FE Mesh visible on Project Navigator, RFEM highlights the two end nodes of the member. I assume that if no mesh is defined manually, RFEM uses the end nodes of the member as the mesh node points.

I have only basic knowledge of FEM, but I understand that usually results are calculated at the mesh node points and interpolated linearly between nodes. But if I have only two nodes, why am I still getting non-linear results? Shouldn't the results, here the displacement of the beam (the curve shown below the beam) be linear between the node points?

If this question would be better under Engineering SE for example, let me know. I felt this is more of a directly FEM-related question than engineering.

$\endgroup$

2 Answers 2

2
$\begingroup$

I am guessing that RFEM uses Euler- Bernoulli beam theory. In that case, you need elements that have continuous derivatives. This is achieved using Hermite interpolation where you end up with cubic polynomials.

$\endgroup$
0
2
$\begingroup$

I don’t know what FEM method RFEM uses, but many methods use elements which are already interpolatory and so the point values of the solution vector are the approximate solution at that point, as you note. What you do to make a picture at in between points is up to you. Some vis packages may be able to resample on your element type to get more accurate values, but I think that most simply interpolate linearly.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.