Learning Forum

[poldsnake]

Hello, I was doing an excercise given at lesson at university but I'm going crazy: I have followed every single step but whenever I try to solve, it warns there's a likely lability problem in the upper structure, I think it's due to the coupling of some link/beam elements but since I'm still learning, I really ca\n't figure out how to solve this problem.
The green elements are link180, the violet elements are beam188, there's only gravity force applied. I have meshed the link dividing by one. if you have any question just tell me

[Erik Kostson]

Does it actually run (you mention warning only which is not always critical), and if it runs can you post process?

If it does not run/solve, then copy paste/attach the error message inline in your next post.

Also do you use K and L commands for creating key-points and lines and the LMESH to mesh those?

----
Finally I would recommend using Workbench mechanical as it is easier to use - one can also go to apdl if needed from workbench (Workbench mechanical can be thought of as a user friendlier pre and post processor of mapdl).

Erik

[poldsnake]

Thank you for your answer.
i started litterally a week ago so I don't use the command line, I only use the menu on the left to create keypoints: Preprocessor->Modelling->create->Keypoints->on active CS/ On working plan. To create the first frame I used the grid and created keypoints on the working plan, then I copied the initial frame 4 times with an offset of 5000. The lines connecting the 4 frames are generated starting by keypoints created "On active CS"
Whenever I try to solve it, I get 2 error warnings on this random point of the upper part of the structure, as show in the next 2 pics. I tried to block all DOFs of this point but there's always another point of the upper part that makes the structure instable, no matter how many points i block. In order to solve this I tried to do the exercise a 2nd time checking the stability on each single step, I found out the exact moment when the instability begins.
In the next 2 pics there's my second trial in the step where the structure goes unstable. It's due to the upper line that connects all 4 frames

[Erik Kostson]

Hi

What elements are you using, think there are only LINK180 - so redo this (that is why you should use commands so they can be reused when modelling again) and define only beam188 and see if that runs.

Again I would strongly recommend to learn the commands since many commands are not accessible in the GUI.

Finally one should use try and use Mechanical for these type of analysis it is much easier.

[poldsnake]

Ok, to make everything clear, I made sure the upper element was a BEAM188 with the correct section, then I made the problem even easier by mesh EVERYTHING with a NDIV =1 so every element is composed of just 1 mesh, I stil get errors. Also, I used the command NUMMRG, KP.
Check this
However, I am bound to use MAPDL because my professor uses it and all of the notes and books that I have use it, otherwise I would have definitely changed it

[poldsnake]

P.S.: The central frames have 1 more keypoint on their side cause I have to build the 2 windshields in the middle

[Erik Kostson]

The only thing is to learn apdl so find tutorial and example and practice take it step by step and learn the commands and functionality.

Finally to overcome the above and to see what parts are not constrained ok, or have the proper stiffness (that is what this pivot error indicates) use a modal analysis and calculate the first 20 modes and look at them (mode shapes) to see which parts have issue.

Best of luck.

Erik

[poldsnake]

Update:
I think I've solved the problem: basically when I created the central line, I attached it to the middle of a frame beam with a ficticious keypoint, to solve the problem i used Modelling->Operate->Boolean->divide->line w/ option, then I divided the line by 2 the central part of each upper frame line creating a natural keypoint, I eventually used those keypoints to create the lines connecting the 4 frames