Comments

• sofien_73

  October 15

  I Mishal I'd seen the same problem on steel frame analysis for a complex one so I've modeled the simpliest cantilever beam with a concentred mass and I had the same kind of result between Mecway internal and ROBOT and RSTAB. At the time I thought that the rotational mass are filtred on the other commercial software it's the main reason which explain this kind of results. I hope Code Aster solver will be avaiable soon for the next version of Mecway
• sofien_73

  October 15

  I can send the cantilever model if it's valuable for you.
• Mishal

  October 15

  @sofien thanks, that would be nice...
  
  I am inclined to think that there may be an issue with the torionsal modes on the structural engineering packages. I have managed to get similar from Code Aster through the plugin created for Mecway.
• disla

  October 15 edited October 15

  Modes are not sorted for proper comparison. isn't it?
  The fact that modes 3 and 4 are almost the same in three software’s is very suspicious (Against Mecway).
  
  EDITED: One of your Young modulus is wrong (20GPa)
  Check how beams are orientated. Specially the L70.
• Mishal

  October 15 edited October 15

  A serious moment of wekaness on my end!!! I was rushing whne doing this. Fixing the modulus brings the first 3 modes close to each other
• Mishal

  October 15

  Interestingly the same model (with the modulus fixed) gives different results between Code Aster and Mecway.
  
  
  
  
• disla

  October 15 edited October 15

  Orientations are not taken into consideration and the structure is much less stiff. Code Aster result is the one without the orientation cards in which beams are oriented to their default values.
  
  I have report on github
• Mishal

  October 15

  Thanks guys!!! I have rectified the issue with the oriantations. Please check the other thread regarding the Mecway and Code Aster bridge.
• sofien_73

  October 16

  I Mishal, I'm going back for my cantilever natural frequencies analysis. Before I send you the results I'd checked once again and guess what I've made a mistake by inversing the web and flange thickness (EUROPEAN HEB300) so now the parameters are correct I can say that MECWAY is the most accurate for the fundamental modes (the first one for example is closer to the analytic solution that ROBOT...) But There's some torsionnal modes with no masses that remains on MECWAY and could make some damage for the design of steelframe by increasing a lot internal forces. On the other hand theses torsionnal modes can be filtered with dedicated steel structure code such as ROBOT. Here's the model and Excel sheet with the results. BR

  analyse modale.liml 7M

  analyse_modale comparative.xlsx 430.8K
• Victor

  October 16

  This problem of torsional modes has come up before. Do you mind explaining what the problem is? Mecway has axial rotational inertia on its beam element nodes which enables torsional modes. Are you saying those modes are wrong or you just want to exclude them? Perhaps an option to give them tiny rotational inertia?
  
• sofien_73

  October 17

  There's no problem with MW but on other commercial software such as RSA, you can select lumped without rotation on the mass matrix option with this method you can filter local rotational modes for small torsional inertia element. I don't know how it works but it have some impact on the dynamic response.
• Mishal

  October 19

  @sofien_73 thanks for your model and data.
  
  @Vicotr, as Sofien mentioned there isn't a problem with Mecway. After I fixed some of the issues I managed to get the same results on Mecway and Code Aster.
  
  The issue sits with structural software. All 3 structural packages give different results while "mechanical" software generally always correlates with each other.