MES Analysis on a complex femur

[jdoggk40]

Hi everyone,

I'm attempting to peform analysis upon the head of the femur by simulating impact on a hard surface. I am using an IGES model of the femur and a skin model that I created around the femur head. I attempted to pin the femur (at a node) in an upright position at the knee end, and apply a force near the top so that it falls in the direction of the fixed surface below. I set up surfaces for contact (the outside of the flesh and the floor), all of the parts meshed successfully, gravity was enabled, and all of the material properties were defined. When I ran the analysis, it didnt even make it to step 1 before it stopped. I tried a few other analyses including attempts with the inside surfaces of the skin and the outside surfaces of the bone set as contact surfaces, but had the same problems. Either way, the bone would not fall towards the floor. Are there any other settings that need to be adjusted so that the femur will fall and make contact with the floor?

If anyone has any ideas, or experience with impact analysis, that would be a great help.

Thanks a lot,
Jesse

 

[CRG]

Have you tried the folks at ALGOR?

 

[jdoggk40]

Yeah. The license I was using is expired, and I don't think it is going to be renewed, so I'm out of luck on that end of things.

-Jesse

 

[GBor]

You say you "attempted to pin the femur". When you say "pin", are you suggesting that you placed an x, y, AND z translational constraint? If you did, you need to redefine the constraint to just the two direction perpendicular to your impact plane. I also wouldn't use surface-to-surface contact for this, but would set up an impact plane. I would avoid too many contact surfaces (such as the bone to flesh that you talked about).

I done quite a bit of impact analysis. Yours really doesn't sound too difficult. Let me know any of this doesn't make sense.

Garland

Garland E. Borowski, PE

 

[jdoggk40]

Yes, when I pinned it, I constrained translation in all directions. I just want the head of the femur to rotate around the pin at the knee end. It seems like the head will still move around as long as the rotational constraints are off.

I'm not sure what you mean by constraining the two translational directions perpendicular to the impact plane. There is only one direction (x) perpendicular to the impact plane that I set up on the surface of the floor.

 

[jdoggk40]

Should all elements be defined as "brick" elements? Choosing 3d kinematic only allows for the density of the material to be entered.

Thanks,
Jesse

 

[GBor]

OK, I reread your post. I'm assuming that you have only 3-D elements in your model. Here's the problem: Brick elements have 3 degrees of freedom (DOF), meaning that rotations are not transferred thorugh the model. This means that your model will not swing the way you had hoped.

Two possible solutions: plates form a hinge with brick elements, so you can model a fictitious, infinitely stiff plate, attach it your bricks at the knee and fix the opposite edge of the plate (doing this will cause the bricks that make up your femur and skin to rotate about the edge of your fictitious plate) OR, you can connect every node of your knee area to a common node some distance away, define the lines drawn from the end nodes to the common node as beams, give the beams infinite stiffness (modulus of elasticity of 1e9psi, area of 1, moment of inertia of 1e6, etc.), and then pin the single node to which your model is connected.

You may also want to restrict the node to which you applied your force so that it stays in the plane of rotation. For instance, you mentioned that the x direction is the only one perpendicular to your impact plane. to keep the femur rotating in the x-y plane, do not allow your load application point to move in the z-direction. This should speed things up a little.

Feel free to call me at 251-232-3723 with questions. I'm in the central time zone in the US.

Garland



Garland E. Borowski, PE