MrRogers1987
Aerospace
- Feb 20, 2014
- 45
I have a model which has a significant number of high frequency modes >300Hz. For example between 300Hz and 500Hz there are almost 160 modes alone. My excitation environments run out to 2000Hz, however. The runs are taking an inordinate amount of time as a result unless I only include a subset of eigenvalues for the modal response.
A majority of these high frequency modes have very low mass participation. They appear to just be panel model of some sheet metal between riveted locations, and are grouped at very similar frequencies (i.e. within 1-2Hz of each other). I would like to include the high frequency modes which are significant, but exclude these that are not. Is there a way to pair down which modes are actually used from the modal response into a smaller subset? Perhaps by excluding certain frequency ranges from the modal solution (e.g. include modes from 0-300Hz, 450-470Hz, 500-510 Hz, 700-725 Hz, etc...)? Alternatively, is there a way to force Nastran to combine modes that are very similar in frequency into a single mode?
A majority of these high frequency modes have very low mass participation. They appear to just be panel model of some sheet metal between riveted locations, and are grouped at very similar frequencies (i.e. within 1-2Hz of each other). I would like to include the high frequency modes which are significant, but exclude these that are not. Is there a way to pair down which modes are actually used from the modal response into a smaller subset? Perhaps by excluding certain frequency ranges from the modal solution (e.g. include modes from 0-300Hz, 450-470Hz, 500-510 Hz, 700-725 Hz, etc...)? Alternatively, is there a way to force Nastran to combine modes that are very similar in frequency into a single mode?
