I have an assembly made up of plate and brick elements with bolted joints simulated with numerous stiff and short beam elements connecting the parts. When setting up a modal analysis, I noticed the toggle for including rotational mass for beam elements and obtained a solution both with and without that box checked. I should note that I don't expect torsional beam modes, but was simply interested in the effect of this option on the results.
The difference in results with the rotational mass box checked vs. unchecked is significant, which surprised me a bit. With the rotational mass accounted for, the first mode of the assembly is 564 Hz, with the mass ignored (unchecked) the first mode is 1,345 Hz. This seems like a very large difference for this effect. Which is the correct setting for the torsional mass in my application, and do you know why the difference is so large?
Solved! Go to Solution.
Is the total mass or weight for all of the short, stiff beam elements correct? If you made them stiff by using a large area, and if the material was chosen from the library, it is likely that the beam elements contribute a significant mass.
Even if a beam is not rotating about its own axis, the entire element may be rotating about another point. I believe the rotational inertia in this case is Icg + m*d^2 (the parallel axis theory), so the extra inertia could lower the frequency.
John Holtz, P.E.
Senior User Experience Designer, Simulation
Current version of Mechanical & Multiphysics: 2013 SP1 (2013.01.00.0012 28-Jun-2012)
Thanks for the quick reply. The beams are sized correctly for the small fasteners (4-40) so the mass is correct, but I2 and I3 were forced to be rather large to increase apparent stiffness. Based on what you say, this must be increasing the overall inertia of the system and lowering the modes.
I reset the MOI's to the default values for the fastener size and am now getting the same results regardless of the setting for rotational mass. It is the same as the original result with the rotational mass ignored. I guess I need to be aware of that setting if I am artificially manipulating MOI's.
From the modling point of view, beam element is one of what called "Engineering " element since it provides an abstraction of a long and slender solid in reality. Comparing to a solid element like brick, it has additional rotational DOFs at each node, and rotational mass is the coressponding mass terms for these rotational DOFs.
By check the box in UI, the program will pick calculated rotational mass for the rotational DOFs instead of setting them to be 0, and their values will be taken into account by the modal(eigenvalue) solver. That is why you see the difference in your results.