substructuring and frequency clustering
Altair EmployeeTo whom it may concern,
I am writing my bachelor thesis at STADLER Pankow (a client of yours).
The goal is to conduct a Harmonic analysis (direct frequency response) on an attachement part of a train wagon.
A modal analysis was already conducted, therefore possible frequencies of interest are already known.
I have two issues at the moment:
1. the card freq3 offers the possibility to achieve a frequency clustering when approaching a eigenfrequency, the problem is that the documentation of optistruct states that it is only allowed in a modal analysis and not in direct frequency response. Is there no possibility to do a clustering in a direct frequency response? if yes I would be grateful if a someone could explain that to me.
2. the rail car body is vital in order to achieve a realistic stiffness, the problem is that it increase dramatically the calculation time. A valid method to solve this issue is to apply SUBSTRUCTURING. This would statically compress all the nodes of the rail car body, holding the stiffness but reducing the calculation time. I have consulted all the documentation of Optistruct, there is no tutorail on how to achieve that. I am in need of a step-by-step guide on how to carry out this procedure.
I would really appreciate a promt answer and I am confident you can help me out.
best regards
Andrea Torcianti
1. No. As stated in the documentation, this is only available for modal FRF analysis, which by the way should be much more efficient for solving large problems (I suggest you to use EIGRA for using AMSES for eigenvalue solving, as it is muuuch faster). Is there any particular reason why you want to avoid Modal FRF and you're using direct?
2. I believe you mentioning superelement approach. In the documentation I recall seeing some description of the Craig Bampton and other methods for superelement creation. I'm pretty sure there's something here in the forum as well. Have you take a look?