Unit cell stress analysis

[antera]

Hello,

I'm new to Pro/M and this forum so please point out any mistakes I'm making.

The problem I'm having is in using Pro/M to analyze stress on a unit cell of a plate heat exchanger. My boss suggested imposing flat rigid symmetry boundaries on 6 faces of the unit cell then moving each boundary around until the stress on each face is equal to the average stress environment that the actual part would be in. Correct me if I'm wrong but Pro/M takes a set of constraints, impose a set of load then calculate the stress. It starts out by calculating the strain from the loads, then the stress from the strain. What I have to do is impose constraints and the amount of strain on one face. Does Pro/M allow me to do that?

In case I confused everyone, what I'm basically doing is I take a unit cell, put rigid flat faces over all 6 sides, and compress the rigid faces to simulate the ambient pressure of the actual part.

Anybody have experience with this or knows how, or if it can be done in Pro/M?

 

[mloew]

antera,

Pro/MECHANICA is fundamentally no different from other analysis tool utilizing the Finite Element Method, so your statement on calculating strains from loads (by inverting the stiffness matrix) is correct. Stresses are indeed calculated from the calculated strains. I hope you are familiar with FEM theory, especially as you are in the nuclear industry.

I am confused about the symmetry constrain on 6 faces of the cube, however. This makes no sense to me. Why not analyze the actual geometry, taking care to properly represent the boundary conditions? It is good practice to only apply the constraints that eliminate rigid body motion (and those that provide symmetry constraints) and that the loads keep the structure in static equilibrium.

If you are still confused as to some of the basics of finite element modeling, I suggest making a post in the Finite Element Analysis engineering (forum727). Use this forum for questions specific to Pro/MECHANCIA.

Good luck!


Best regards,

Matthew Ian Loew
"I don't grow up. In me is the small child of my early days" -- M.C. Escher

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[antera]

Hey Matthew,

Thanks for answering!

The model is actually not a cube but a unit cell of a much larger model. The actual model is much too large to analyze in Pro/M, so I had to create that cell. The reason why I apply 6 symmetry boundaries was because the cell is symmetric in all 6 directions. My boss told me that by compressing any one face you would get a change in stress on that face of the cell, and use the change in stress to calculate an effective elastic modulus for the unit cell in that direction. But to do that I need to know how to find the stress on the face (possibly by integrateing the stresses over that face? I didn't find any function in Pro/M for that.)

The second thing I was trying to achieve is to find the actual stress state the unit cell would be in if the all the faces have an average stress equal to the ambient pressure the part will operate in. Which I can do if I could integrate the stress over the face.

I guess this problem is very confusing. I'll try out the FEM forum like you suggested. But if any one knows how to extract stress data from a plane in a model please let me know.

And thanks again for answering.