Problems in soft tissue indentation simulation using ABAQUS Standard

[ksayew]

I am having problems getting ABAQUS Standard 6.2 to solve an indentation simulation. The description of my model are as follows:

-Axisymmetric modelling of soft tissue indentation (rigid indentor indenting a rectangular block)

-Rigid body for hemispherical ended indentor using a reference node to define

-Tissue material properties are E=0.7 MPa, Poisson's ratio=0.49, linear elastic, CAX4 elements

-Boundary condition are xsymm along axis of symmetry for tissue. Last row of nodes at the bottom of tissue block is not allowed to move. Indentation is simulated by displacement of the reference node. Reference node allowed to move in vertical direction only (no rotation as well)

-multiple step loading. Total 3mm indentation (1mm/step)

Using ABAQUS Standard, I was able to get results for up to step 2, but at the last step (3mm indentation), I exprienced convergence problem. I have looked through the .msg file and it looks like this at the last iteration:


INCREMENT 3 STARTS. ATTEMPT NUMBER 1, TIME INCREMENT 1.000E-06

***ERROR: TIME INCREMENT REQUIRED IS TOO SMALL. ANALYSIS TERMINATED.

Appreciate if anyone can provide any ideas to get around this problem to solve in ABAQUS Standard. I have tried hyperelastic elements and it works. I have also tried adaptive meshing with ABAQUS Explicit, while it works, its computational time is longer (about 45 mins). Is this a case of element locking due to the incompressibility?

 

[harry123456]

In increment three look at the message file. Is abaqus cutting back because it cannot resolve contact? maybe severe discontinuity iterations? or it can resolve contact but cant achieve equilibirum? you could try and put some damping in step three either for the contact or just for the step and see if it works. If you are working with explicit you will need to be careful as to what sort of problem you are solving, static or dynamic.
hope this helps
harry

 

[ksayew]

I have looked at the .msg file again. ABAQUS was able to resolve the contact but unable to achieve equlibirium. In this case, how should I resolve this?

If I have to introduce damping to the contact, how should I go about doing it? BTW, thanks for the advice on the explicit bit, I will keep that in mind.

 

[harry123456]

Contact is going through. Equilibrium is a more difficult issue. Did you try much more refined mesh? Also, if contact is going through dont use Contact Damping instead use *Static, Stabilize. It sometimes might help achieve equilibrium. Yes, look into the quasistatic analysis section in explicit. That will help you understand that part.

 

[cmgv]

Have you opened the output database to look at the first two increments? Doing this and using the job diagnostics tools can help you determine where the problem is (by highlighting the elements that have the largest residuals, are experiencing distortion, etc.). You may also want to look at the stresses developed so far. With a stiffness of 0.7 MPa, you might be completely crushing the material.