I am looking for the examples of using modal analysis (SOL103) results externally (in matlab, python or any other language), for example for frequency response analysis. It is understood that Nastran can do it (SOL111), however the idea is to try out all steps manually to gain more insight.
I came across the presentation exactly on this topic – Solving Dynamic Problems Outside of NASTRAN femci.gsfc.nasa.gov/.../Gordon_Modal_Methods.pdf where modal effective mass table and the rigid body modes vectors are calculated via DMAP. The presentation is quite old (dated from May 2002), and DMAP does not run in latest versions of MSC Nastran. But, in general, it exactly covers the topic I am interested in – frequency response due to base excitation using eigen modes calculated in Nastran.
I have noticed that in MSC Nastran 2019, SOL103 outputs MODAL_EFFECTIVE_MASS and RIGID_BODY_MASS_MATRIX directly in the HDF5 file (btw, very useful output format! Thank you!) However, I am still missing the RBGL1 data (rigid body vectors). How can I generate this data? Is it only possible via DMAP?
Any recommendations or links on this topic are appreciated.
Let's see if anyone else has some thoughts, but I do think you will need to use an Alter. There are many locations you could generate the rigid body vectors... as a starting point, you can pull the appropriate lines from that Alter and insert them in the MODEFSRS subdmap, as all the necessary pieces are there. This might get you started:
COMPILE MODEFSRS
ALTER 'RETURN' (1,-1) $ insert Alter before Return
$ Generate partitioning vector for G to A set partition
VEC USET/VGRC/'G'/'A'/'COMP' $
$ Partition rigid body modes to A-set size
PARTN RBGL,,VGRC/RBAA,,,/1 $
MATPRT RBGL// $
MATPRT RBAA// $
ENDALTER $
Once you are comfortable this gives you what you want, there was another thread on this forum that discussed putting data blocks on the HDF5 file... you might be able to get that to work, or you could always output4 this information.