I'm trying to perform a punching test in a steel wire mesh (MT1500). In order to simplify the model, I'm using shell elements. I'd like to use a material with a growing Young Modulus, corresponding to two phases: the initial behavior in wich wires get straight, and the later tension until rupture, much stiffer.
Does anybody know how to do it?
From a practical standpoint you can use a multilinear isotropic hardening material model. It allows you to input plastic strain vs stress directly, so you can make it do anything you want. Just remember it's PLASTIC strain, so there will be a small component of elastic strain up to 0.2% strain that you will have to manipulate by using the elastic young's modulus. Also since it is a multilinear curve you can input more data points to get a smoother curve that should help with numerical stability.
That was my first try, but unsuccesful. ANSYS abort with an error, and I think it was because of the growing slope. There's no problem with second (plastic) slope lower than the first (elastic) one, but it's not my case.
What was the error? Something along the lines that it could not converge? In that case you should add more timesteps to the solution so that the solver can get over the instability of the sudden change in stiffness. With things like this I like to add several loadsteps (i.e. step 1 from 0 to 1 second, step 2 1 to 2 seconds etc.). Then when if you tweak timesteps etc you can just restart the last loadstep instead of solving from the start again.
If you can get more data on the transition between young's moduli or you can make an assumption that helps to smooth out your curve it will likely also help a lot.
One way around the problem would be to use two materials, each with different elastic modulus. The first material would be the one with the lower slope and would be the default material of your model. You could then change the material in these elements that present strain higher than the limit at which the slope changes. The selection and recalculation procedure should be repeated a few times until you are sure that the deformation overall does not change. This is relatively easier to accomplish with APDL.
- Badges
- Science topic