Hi everyone,

I am working on a highly nonlinear Abaqus/Explicit simulation of a cutting process, where a deformable workpiece interacts with rigid cutting tools.

The deformable workpiece undergoes combined motion, including:

• horizontal translation, and
• simultaneous rotation with a prescribed angular velocity.

To model material separation, I am using ductile damage with element deletion.

However, I am encountering an issue where the workpiece exhibits premature, nonphysical deterioration before and during the cutting stage. Instead of observing localized damage at the cutting interface, the material fragments into multiple segments along its length, including regions that are not in direct contact with the cutter.

Contact is defined using general contact with a friction coefficient of 0.3.

The mesh of the workpiece is Sweep Hex, element type, and material properties are summarized below:

The workpiece appears to be fragmenting into small segments before or during the cut, and the model is removing elements too early, not only at the intended cutting zone as shown below:


 

My goal is to achieve localized damage at the cutting interface and not other points. I have some questions that need your technical insights:
 

1. What could be the main reason for the observed premature deterioration and fragmentation of the workpiece?
2. Is using a single stress triaxiality value (0.33) in the ductile damage model sufficient for a cutting simulation, or should a range of triaxiality values be defined to capture different stress states?
3. Would switching from displacement-based damage evolution to energy-based evolution improve stability and reduce mesh sensitivity in this type of problem?
4. Could the observed behavior be related to:
   • overly aggressive damage evolution parameters,
   • inaccurate representation of the stress state, or
   • mesh dependency of the ductile damage model?
5. Could the combined translational and rotational motion of the workpiece introduce additional stress states that contribute to premature damage initiation?
6. For metal cutting simulations in Abaqus/Explicit, is it generally more appropriate to use:
   • the current ductile damage model, or
   • Johnson–Cook plasticity with Johnson–Cook damage?

Thank you,