Background:
Using HyperMesh and the Radioss user profile it is possible to create a rigid body where the mass is dependent on the defined mass density of the material assigned to the rigid part and, also, to add mass using the RBODY card (thus, if an initial velocity is applied to the rigid part, the total kinetic energy is determined using 'mass + rho*volume').
I am wondering what the equivalent keyword(s) would be for an LS-Dyna .k file. In HM using the LS-Dyna user profile, I understand defining the mass density in the MAT_RIGID keyword, but adding mass to a node (1D > mass > create > 'select node' > create) does not yield the same results as simply adding mass using the RBODY card for Radioss. I have confirmed this difference by first ensuring that the added mass is part of the rigid component and calculating the mass (Tool > mass calc > 'select component' > calculate). The expected total mass is provided (mass + rho*volume). Yet, when I run the .k file, the initial kinetic energy provided only takes 'rho*volume' into account and not also the added mass.
Actual Question:
Is there any way to add mass to a rigid component - or a node that is a part of the rigid component - without modifying the mass density of the material assigned to the rigid component? (I want to be able to modify the overall mass of the rigid body using either/both a node with assigned mass and mass density of the assigned material.)
Additional information:
- I have attached a sample .k file which has mass + rho*volume = 0.5kg + 0.5kg = 1kg, but the initial kinetic energy listed during initialization is only 20.25 J. The initial velocity is 9 m/s, so I expect 0.5*1kg*(9m/s)^2 = 40.5 J.
- Having done some digging, I feel like the answer lies with the *CONSTRAINED_NODAL_RIGID_BODY_INERTIA keyword, but my testing with it seems to suggest I can't also be using MAT_RIGID (I receive "Error 20393 (STR+393) Nodal rigid body ID ## contains node ID ## already used in a rigid body/nodal rigid body definition.").
Please let me know if you require any clarification of my question.
Thank you in advance for your time and any suggestions.
Try using Keyword>Part>Inertia instead.
This allows you to input the translational mass along with the inertia tensor and some other parameters which should overwrite the calculated mass from the material density. I am certain that this method will work with the Rigid material card as well.
Hopefully this helps you!
Hi @rybask1,
Thank you for your suggestion!
*PART_INERTIA does allow me to provide a translational mass and even an initial velocity. Unfortunately, as you mentioned, it overwrites the mass from the material density, instead of taking into account both (which is possible in Radioss). However, it does work with *MAT_RIGID, so I think I will go ahead with this solution for the time being.
Thanks again!
And as an additional note for anyone following this same route of using *PART_INERTIA and defining initial velocities of the rigid body within the keyword: the solver doesn't seem to calculate it as a part of the kinetic energy of the system when it does its initialization (for me, kinetic energy = 0). However, everything seems to solve properly and plotting kinetic energy over time gives me the expected kinetic energy at time = 0sec.
Interesting that it is excluded from the system's kinetic energy, but I'm glad it worked out for you!
I suppose that the 'initial velocity' for the rigid body is not applied at t=0, rather instantaneously afterwards in the d3plot.