I have a 3D macro scale model of half pressure vessel. I am applying pressure on the top surface in y direction while fixing the bottom surface in y direction. The simulation result indicate that the pressure vessel's centre has displaced along 45 degree line in x-z plane. What is causing this error? Is it the material orientation or some other issue that is causing this?
Difficult to provide a comment without a sketch of the model/loads/constraints...
Difficult to provide a comment without a sketch of the model/loads/constraints...
As Dr. Denis Benasciutti mentioned, it is very difficult to comment W/O having a sketch, However, I think that your model is free to move along the X and Z directions and for that reason you see the displacement along 45 degree in XZ plane.
Ideally, if you have perfect symmetric geometry with respect to XY, XZ and YZ plane and apply a load only in Y direction, you should not get displacement in X and Z directions. If FE model deviated a bit from what I explained, certainly, there is going to be a problem. I have seen this situation a lot and sometimes solver was not able to converge even.
I suggest that you review the BCs (specifically displacement) in your model and let us know if you get any improvement in your results.
Denis Benasciutti Hamidreza Ahmadi Moghaddam Claudio Pedrazzi
Thank you for your recommendation and responses. I have shared the model. I request you to kindly have a look at it and share your suggestions based on the model. Thanks.
Claudio Pedrazzi
There was some error in uploading. Sorry for the inconvenience. By model I meant the screenshot which is live now.
Thank you Claudio Pedrazzi
for prompt responses. I am using Mechanical APDL for simulation. I will try to implement weak springs. Also, the elements along the y-axis in this model have Aspect Ratio of 55 (recommended is 20 as per ANSYS) so can poorly shaped elements lead to motion in XZ plane?You are right. I am using this simulation just to validate the model (compare with 2D axisymmetric model for same load case). I actually need to perform impact loading finally.
And fun fact, I created this model from a 2D axisymmetric model because I couldn't perform Impact Loading simulation on a 2D axisymmetric model.
axisymmetric model definitively, although a small thickness value would also make the use of shell elements a possible alternative
Claudio Pedrazzi
Thank you so much for your valuable suggestions. I was able to validate my model. As per your suggestions, I will use a cut portion of the model for impact loading case. Also, the movement in XZ plane is due to nodes (half of them) at the bottom surfaces at not exactly at y=0. I want to shift them to y=0. Any idea how can I shift these nodes to y=0?- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Computer Graphics and Animation
- 3D