Simple 2D analysis

DaveStupple

I feel like this should be really simple, but I'm struggling. I want to model a 'simple' 2D square with two orthogonal shear forces and two orthogonal normal forces.

1. How do I constrain this? I have tried and failed to find a clever way of rearranging the forces and displacements. I can see how the rotation is unconstrained, but I don't get how the deformed result is so huge. Is it because the constraints are infinitely strong and the shape must stretch to obey them? I would have thought the stiffness of the material would limit this.

2. Using Static 2D, in materials properties I cannot leave geometry as 'none' and, with 'shall/membrane' selected I cannot leave thickness at zero. Do I just need to put in a dummy thickness, or am I missing something?

Looking forward to feeling stupid when you put me right! Thanks for reading.

Victor

Hi Dave

1. It turns out the large displacement is actually a rotation because of the unconstrained rotation you identified. But the point displacement constraints are going to cause point reaction loads too which I guess isn't what you want. I'm not quite sure how you want to constrain it but if you're going for something like the textbook description of general stress on an infinitesimal element, then perhaps model the loads on all 4 sides and use some point constraints just to protect against possible rigid body motion. I've done that in the attachment with some of the forces changed to make them uniform and balanced. You don't actually need the displacement constraints but it might wander away a bit without them.

2. The thickness is needed for some loads to make sense. For example, a given force on something with an unknown thickness will cause an unknown displacement. What you did by setting it to 1m is a good idea to make it behave like arbitrary thickness if you prefer to think of the forces as force per unit thickness.

DaveStupple

A neat trick, I hadn't thought of that. Yes, I'm just visualizing basic textbook concepts. Would I be right in saying that in a simple system like this with uniform loads the principal planes (if I'm using the right term) would be in the same orientation throughout, but in a more complex system the principal planes will vary from point to point?

Victor

I think that's right. The entire stress state should be uniform throughout, including the principal directions.

DaveStupple

Thanks Victor.