*ERROR in bounadd: parameter FIXED cannot be used for rotations

1

I got strange behaviour using the *Boundary keyword and the Fixed parameter. I applied the Fixed parameter to a single node in the one finite element model attached at the end of the message. I started with translations, and everything worked fine. Then I added the rotational DOF around the x-axis, the DOF number 4, and still, everything worked fine (I know the solid nodes do not have the DOFs 4, 5 and 6). But when I add the DOFs 5 or 6, I get an error message: " *ERROR in bounadd: parameter FIXED cannot be used for rotations".

So my question is, why does it work for DOF number 4? Am I missing something?

*Node
1, 0.00000000E+000, 0.00000000E+000, 1.00000000E+002
2, 0.00000000E+000, 0.00000000E+000, 0.00000000E+000
3, 0.00000000E+000, 1.00000000E+002, 1.00000000E+002
4, 0.00000000E+000, 1.00000000E+002, 0.00000000E+000
5, 1.00000000E+002, 0.00000000E+000, 1.00000000E+002
6, 1.00000000E+002, 0.00000000E+000, 0.00000000E+000
7, 1.00000000E+002, 1.00000000E+002, 1.00000000E+002
8, 1.00000000E+002, 1.00000000E+002, 0.00000000E+000
**
** Elements ++++++++++++++++++++++++++++++++++++++++++++++++
**
*Element, Type=C3D8, Elset=Solid_part-1
19, 1, 2, 4, 3, 5, 6, 8, 7
*Material, Name=S235
*Elastic
210000, 0.28
*Solid section, Elset=Solid_part-1, Material=S235
*Step
*Static
*Boundary, Fixed
7, 1, 1
7, 2, 2
7, 3, 3
7, 4, 4
*Node file
RF, U
*El file
S, E
*End step

2

I’ve encountered this error message some time ago when working with shells but I didn’t know that it doesn’t apply to DOF 4: Load Case Creation - #17 by FEAnalyst - General Questions - PrePoMax

The source code is set to show this error when DOF >= 5:

        if((fixed).and.(i.lt.5)) then
           val=vold(i,nodetrue)
        elseif(fixed) then
           write(*,*) '*ERROR in bounadd: parameter FIXED cannot'
           write(*,*) '       be used for rotations'
           call exit(201)
        endif

Perhaps it’s just a mistake. We could report this as an issue or even submit a pull request here: GitHub - Dhondtguido/CalculiX: This repository contains the source files of CalculiX, a three-dimensional Finite Element Program (www.calculix.de).

1 Like
3

I’m not 100% sure, but something in my memory from an ancient ansys theory manual tells me, that the linear elements normally only have 3 degree of freedoms per nodes, and in that case it shouldn’t be allowed to fix any of the rotations.
Since the linear element can only maintain a linear stress distribution between 2 opposite nodes and imagine the thought experiment to impose opposite angular rotations to opposite nodes, then it will violate the solution as the element can’t flex in the middle between the nodes

4

Linear or not, solid elements don’t have rotational DOFs at their nodes. However, here the issue is that for some reason DOF 4 (rotation about 1st axis) works with the FIXED parameter while DOFs 5 and 6 (rotation about 2nd and 3rd axis) don’t.

5

You are right. This is probably a mistake. But the general question is whether the FIXED parameter is allowed for rotational DOFs. The documentation has no information about it. There is a case when a normal node is used as a rotational node, for instance, for the definition of a rigid body and its rotations. In such a case, the rotations are not defined by the DOFs 4, 5 and 6 but by the DOFs 1, 2 and 3. So, in this case, the error checking should be accordingly modified.

How is it in Abaqus?

1 Like
6

I don’t know specific which type of solid elements implemented in Calculix but solid elements with rotational degree of freedoms has actual been formulated.

Solid elements with rotational degrees of freedom: Part 1—hexahedron elements - Yunus - 1991 - International Journal for Numerical Methods in Engineering - Wiley Online Library

7

That’s right, their formulation can be found in research papers but I’ve never seen them being used in practice in any FEA software.

1 Like
8

In Abaqus, this parameter is ignored if it’s used in the first step of an analysis but it can be used for rotational DOFs as well.

2 Likes
9

it was found by well known researcher and book writer used by many academic and practical around the world.

LS-DYNA one of FE most industry used has implemented these element type, seems it more effective compared to fully quadratic.

however CalculiX till the current version use standard common solid element without rotational DOF’s. could be useful when someday have such as element, especially in contact problems.

1 Like
10

I do not know what is the best way to report such problems but I would suggest adding Guido to the conversation.

11

Sure, let’s notify him.

@dhondt Could you take a look at this ?