Yes it is possible to change from a coordinate system to another one with Ansys CS command which allows to define local a coordinate system from an active one.
You have also the possibility to change the element type and its corresonding coordinate system.
plane elements have both "plane stress/plain strain" and "axisymmetric" capabilities (in APDL you control them with "KEYOPT", which switches the equations describing the element behaviour, see any FE textbook). However, I agree with Claudio Pedrazzi
that you cannot activate both options in the same analysis: your model can be in either plane stress/strain or axisymmetric condition.
Hi all. This interesting question drove me to try the simualtion out myself. I included both axisymmetric and plane stress elements in Ansys Apdl. It did give me a solution! Of course it might just be mathematical junk but the software does give an output. I guess this is not possible in Ansys Workbench.
I am sure that you have understood the example that I have given. This means that you are wrong to reject my response.
In finite elements there several ways to mesh a structure using ONE element type or TWO elements types. As requested by Manish kumar Dubey, one can use for the simple 2D case Plane13 coupled-field solid with plane strain, axisymmetric and plane stress : this one can be difficult to model a 3D structure because it only needs the 2D plane element and using plane strain for an area plus axisymmetric or plane stress for another area is available without changing the element type. A simple way is to use the Shell181 element for membrane (tensile/compression) effects and / or bending effects with no problems to mesh the axisymmetric part. it is also possible to use Shell41 membrane shell in 3D. But Solsh190 3D-8 node structural solid shell is better. And Shell281 8-node structural shell is more convenient with CS command and better to mesh curved parts.
Hello Claudio Pedrazzi. There were no warnings while running the simulation. I too was surprised because it is physically untenable. I am attaching below screen shots from the problem I had tried. The violet part of the model is axisymmetric.
It is not enough for the geometry to be axisymmetric, the loading should also be axisymmetric which is not possible when the axisymmetric elements are used in conjunction with the plane stress elements. But the software does give an output.
I believe an equivalent 3-D model cannot be constructed because this problem that I have modeled to test the propostion (as shown in the images) lacks physical meaning.
Claudio Pedrazzi, thanks for the answer. It has been a good learning for me. May be I can try the "two models in one" approach some time. I now understand that you have been seeing this from a point of view different from mine. You were probably looking for "two models in one" while I was looking for compatibility between the two parts of the model.
this "hangar" is not axisymmetric, as it is not a solid of revolution. As correctly shown by Dr. Claudio Pedrazzi
, a "solid of revolution, also called an axisymmetric solid, is generated by revolving a plane figure about an axis in the plane" (I quote from the well-known book by Cook "Finite element modeling", Ch. 6, pag. 145). My opinion is that the original question was somehow misleading and thus it led to some misintepretations
The hangar figure is an assembly of two symmetric flat plane parts and an axisymmetric cylindrical part with respect to a parallel axis to OZ. This structure can be meshed with plane shells or plane and curved shells.
To handle a shell element you have to use the Kirchoff's theory which combines the membrane effect with displacements of a quadrilateral element plus the bending effect with the flexure of a plate. After that the stiffness matrix is assembled in the global matrix using transformatio matrices. It is also important to notice that tje plane stress or plane strain gives the specific expressions of the material properties matrix of a membrane element. This example must not be used in the axisymmetric case of a volume structure where a plane element generates the axisymmetric part as it is shown in the example given by Pedrazzi.
But one can combine different element types on a specific application with plane elements and axisymmetric volumes generated from plane areas if the axisymmetric part is a thick volume or using another additional volume to the considered structure.
Mohammed Lamine Moussaoui Denis Benasciutti Mir Aamir Abbas for your reply.
I am bit perplexed with the answer received as of now.
But the question is straightforward. In ANSYS when you choose element type such PLANE XXX it will show three option under element behavior Axisymmetric, Plane Stress and Plane Strain respectively. My doubt when you model Axisymmetric and you had plane stress properties how do you approach to such problem. Will you go back and convert those reduced properties to full one.
Hi Manish. Let me try to guess what you are asking if it is not answered already. Are you asking whether we can model an axisymmetric shell where the shell is represented in the axisymmetric coordinate space by a line rather than an area and the thickness is given as a real constant or section data?
I just wanted to add a comment in addition to your instruction on how to perform axysimmetric analysis in WB. The Geometry that you want to model as axysimmetric should be modeled in "Positive" X and Y coordinate. Otherwise Ansys doesn't accept it as axysimmetric geometry.