Learning Forum

[MubasherAli]

Hello, all respected mentors and Helpers!

I am simulating spring in static structural with non-linear data but the ANSYS does not converge even for a single point and abruptly stop the ANSYS unsolve result. However, I tried to put full the non-linear stress-strain data ( True Stress-Strain Data)  from Yielding to Ultimate Strength but still not result. Please have a look at the Pictures. Also, I have tried with both mentioned below non-linear data but no result.

Boundaries :

1: Force

2:Fixed Support

3: Displacement ( Restrained in X, Z and only free in Y-axis)

[peteroznewman]

Let's look at the big picture here. Springs are supposed to be elastic. That is their basic purpose. You press them down and when you remove the force, they come back to their original length.

Why are you using a plasticity material model on a spring shape?

What is the purpose of the structure that you have created?

[MubasherAli]

Thank Peteroznewman!

Yes indeed that we should not use Non-linear Stress-strain data. But basically my aim is to investigate the property of the geometry that whether the stresses and deflection could effect by geometry or not. So, for this reason, I have put the non-linear data. Secondly, the purpose of the structure I have created, for Designing of midsole by using the springs. In this case, I wanna check the non-linear behave but unfortunately, I am not getting any result.

[peteroznewman]

"my aim is to investigate the property of the geometry that whether the "stresses and deflection could effect by geometry or not"

You can do this with a linear material model of simple isotropic elasticity.  You can run the model with one geometry, for example diameter of "wire" and get the deformation for the force input, then you can change the geometry to have 0.8 times the original diameter, get the deformation, then change the geometry to have 1.2 times the original diameter, get the deformation.  In this way you will have three inputs of diameter and three outputs of deformation. They will not be in a straight line. That is your nonlinear response.  Keep going with more points like 0.6 times and 1.4 times the original diameter.  Now you have five points along a nonlinear curve.

 

[MubasherAli]

Thanks for reply 

"You can run the model with one geometry, for example, the diameter of "wire" and get the deformation for the force input, then you can change the geometry to have 0.8 times the original diameter"

 

Yes, I got your point but basically this geometry is about like a cone not like what you thought about. We can not do what you mentioned because it is gradually decreasing or increase the spring diameter and constant at one end.  More clearly, before I have simulated successfully the same spring with the non-linear data but now I am facing this problem because of the decrease in the Stress-Strain points in the non-linear data.

 

Summing up: My query is why the simulation is not going to succeed when we decrease the Stress-Strain Points?

Please let me know if you need further information regard this issue

[peteroznewman]

A model that has Plasticity as you have made, and also has a Force input can fail to converge before the full Force is applied simply because the material cannot support the full Force.

This is easily proven by changing the load from a Force to a Displacement. Now you can plot the Force Reaction on that displacement and you will see the force increase linearly, then a curve develops and eventually, the curve flattens. The point when the force-displacement curve has a zero slope, that is the failure load.  If that failure load is less that the full Force, that explains why a model with a Force input is going to fail to converge before reaching the full Force.

[MubasherAli]

Thanks again. I got your point.