Learning Forum

[AsarAJ]

Hello, 

I am having difficulties converging my nonlinear analysis. I have used a large substep number, checked my contacts, but to no success. 

the analysis includes fracture contact debonding, and I would like to see how the non linear force-displacement curve looks like. however, it never gets to converge at the point where full debonding ( failure happens) any suggestions would be highly appreciated.

[Wenlong]

Hi,

At the fully debonding point, you probably need some damping to make it converge. Please try adding a nonlinear stabilization and see if it helps.

Regards,

Wenlong

 

[AsarAJ]

Hello Wenlog 

Thank you for your reply, 

that didn't work well.. 

I am trying to find the effects of geometry at material( crack zone material model) on the denbonding fracture contacts. 

I have defined 3 crack zone materials (same critical fracture Energy for Normal separation/ different maximum normal contact stress)  and I have also defined three geometries ( I am using a unit cell from a hex lattice). the goal is to see with FEM how these different models compare ( through generating a load-displacement graph) - so it would be nice to see the nonlinear relationship fully go down to zero after breaking the contact ( also to compare the areas under the curve

some models converge easily others I have tried relentlessly to even get to its nonlinear behaviour ( load-displacement curve). 

also one model requires me to define the elasticity of the crack zone material, which is a little strange (??) 

if you have any feedback or suggestion, please let me know!  

[Wenlong]

Hi,

I would also try transient structural analysis to take care of the dynamic effect at the breaking point, since there will be a sudden release of energy. Below is an image of the single lapped joint specimen delamination simulation. I am using a contact rebounding with a small nonlinear stabilization (energy factor=1e-4). 

If you are using the bilinear cohesive model, I think the elastic modulus is used for the initial slope. 

Can you please share an image of the hard-to-converge model? After knowing the geometry and the boundary conditions, maybe I can give more suggestions.

Regards,

Wenlong

 

[AsarAJ]

I would like to not have a transient analysis actually, and also I am trying to see the effects of different parameters on the failure; so I don't want to consider damping

 

and the other ''top branch'' has the same conditions (symmetric -locked in normal direction and rotation) 

 

[Wenlong]

Hi,

Sorry for my late reply. Can you try displacement control instead of remote displacement? That may make it easier to converge. Even if you don't include any nonlinear stabilization damping, a tiny bit of artificial damping in the separation-distance based debonding may be necessary to converge. For example, the first image I showed uses artificial damping=0 and unable to converge, while the second one only uses artificial damping coefficient in CZM as 0.0001 and it helps tremendously in convergence. 

By the way, I commented wrongly on Young's modulus in the previous reply. It doesn't directly influence the initial slope of the bilinear cohesive law, but indirectly influence initial stiffness by influencing the normal contact stiffness. Please refer to this website for more details: https://ansyshelp.ansys.com/account/Secured?returnurl=/Views/Secured/corp/v194/ans_thry/thy_mat11.html

Regards,

Wenlong

================ Note ====================

If you have trouble opening the links I attached, please see the first useful link below

 

 

---

Useful Links

---

• How to access Ansys Online Help Document


• How to show full resolution image


• How to use Google to search within Ansys Student Community