How to get force (not stress) through node/element/surface/whatever?

[zozomama]

Hi,

I have a loading condition similar to something shown in the picture. Different materials are shown in different colors.

Some magnitude of force (say 100N) is evenly applied across the top surface of the yellow and green materials. The bottom of the blue material is fixed. Since the yellow and green have different material properties, 100N of force will not split evenly between them as it carries through the materials. I need to know how much force each material is taking.

Reaction forces are only given at the region where boundary conditions are applied (blue material). The value comes out correctly as 100N in the other direction, but this isn't very helpful because it tells me the force that the blue material is taking.

I do not mind if I have to look at many nodal/elemental forces and sum them up to get the total force through the material, as I already wrote a script to do that. Is there a field output I am forgetting to include, or is there a trick to it?

The reason I want to do this is that I want the yellow material to carry the majority of the load. I can change the material being used to be more/less stiff to control that, but if what I have now is ok, then there is no need to do so. I don't want to tell the folks at the shop to switch out materials when it's unnecessary.

Thanks in advance.
-z

 

[johnhors]

I write element forces (at nodes) to an ascii fil file using *EL FILE followed by NFORC on the next line. You then have the forces acting on every element, by summing these forces across a set of element faces that passes through the block of material you can get the total force acting on the material block.

Though in your case the green and yellow materials share a common boundary right up to the blue material, the load path in this structure will be complex and changing with respect to position. There is no obvious or natural place for you to make a summation of forces in this particular scenario.

http://www.Roshaz.com