Get stress on beam modal analysis

Hello everyone,

I'm actually working on a sub structuring modal analysis. My code works pretty well, but I can't get the stress on my beam structure. My calculations gave my the displacement over the structure for each mode. Theoretically, I have all the informations I need to get the stresses by post treatment (using CALC_CHAM).This command is used to create the assembled model of all the macro elements I have, but actually the model (MoASS) generated can't match with the type of model asked by the command CALC_CHAM.
Is there any command that I can use that would transform my generalized model into a usable type of concept ?

Hello,
doesn't page 8 of the training material regarding dynamic sub-structuring help?
I think the important operators whose documentation you should read DEFI_SQUELETTE, REST_SOUS_STRUC and REST_GENE_PHYS.

Best regards,
Richard

Hi Richard,

Thanks for your participation again !
I have spent my whole day reading papers and papers about ASTER, and of course I read the ones you advised. I have tried like thousand of things, but now my question is more general. I wonder if that post is a complete non sense!!! I'm actually trying to get stress on a simple modal analysis, in which the only thing I can get at the end are eigenvectors that are not real displacement. Eigenvectors can only represent relative nodal displacements (if node 1 move by x at Eigenvector W frequency then node 2 move by f(x), and etc...). Without any excitation force or displacement, getting the stress is absolutely impossible right ? If I'm right, then this post is totally useless .

Not really,
you still get a qualitative view of the stresses. The regions with the highest stresses will be correctly predicted, only the scale is unknown.

Best,
Richard

Thank's Richard, I know have clear ideas about displaying the stresses.

I have succeeded displaying stresses using only one substructure, but I can't do anything with my skeleton. Aster constantly says that the model inside SQUELETTE does not exist. More over, when I put manualy the generalized model into the CALC_CHAMP, I get an error : <modele_gene> can't be used, CALC_CHAMP needs a <modele_sdaster> as an input. Is there anyway to convert a generalized model into a model CALC_CHAMP can use ?

Hello,
I aven't used substructuring so i don't exactly know how to use CALC _CHAMP correctly there,
but I think you do not have to explicitly state the model, you can just state the result and CALC_CHAMP
will get the correct model from the result implicitly.
Of course it should not be a result in generalized coordinates, that's why you need for example REST_GENE_PHYS.
Not sure if this works, but you could try it.

Best regards,
Richard