Negative sliding energy
Ramdas Chennamsetti
James Kennedy
Dear Ramdas,
Negative sliding energy in an explicit dynamics simulation (such as with LS-DYNA) is an unphysical result and indicates problems with the simulation setup, typically related to the contact definition or mesh quality. It often results from slave nodes penetrating the master segments without the contact algorithm properly detecting and resolving the penetration, essentially the contact is "giving" energy back to the system instead of dissipating it.
Causes
Initial Penetrations: The most common cause is undetected initial penetrations between the projectile and the panel. Even small overlaps can lead to significant issues, especially with node-to-surface contact.
Node Sliding Between Segments: In node-to-surface contact, a node might slide from one master segment to an adjacent, but numerically unconnected, segment. If a penetration is immediately detected on the new segment, this can result in an erroneous negative energy contribution.
Poor Mesh Quality: Severely distorted or poor-quality elements can lead to instabilities and negative energy, including negative volumes.
Time Step Issues: While less direct, an overly large time step scale factor (TSSFAC) can contribute to instabilities, which might manifest as negative energy.
Redundant Contacts: Defining more than one contact interface for the same parts can cause conflicting calculations and energy errors.
Contact Stiffness: The default penalty-based contact force is proportional to the penetration depth. If the stiffness is not appropriate or the penetration is large, it can lead to non-physical energy results.
Solutions
To address the negative sliding energy, consider the following troubleshooting steps:
Eliminate Initial Penetrations:
Carefully check the initial geometry and shell offsets for any overlaps between the projectile and the panel using pre-processing tools (like LS-PrePost).
Use contact parameters like IGNORE (on *CONTROL_CONTACT or the contact definition card) set to 1 or 2 to allow the solver to ignore initial penetrations, though manually fixing the geometry is preferred for accuracy.
Improve Contact Definition:
Ensure only one contact is defined between the relevant parts to avoid redundancy.
Consider switching to a segment-based contact (e.g., *CONTACT_AUTOMATIC_SURFACE_TO_SURFACE with SOFT=2 in LS-DYNA) if possible, as it often provides better stability and energy conservation, especially for sharp edges or complex geometries.
If staying with node-to-surface contact, ensure all optional contact control parameters are set appropriately, perhaps by using default settings and only modifying key variables like SOFT=1.
Refine Mesh Quality:
Improve the mesh quality in the contact area, especially around sharp edges or high-stress zones, to prevent element distortion during impact.
Adjust Simulation Controls:
Reduce the time step scale factor (TSSFAC in *CONTROL_TIMESTEP, e.g., from 0.9 to 0.66) to enforce smaller, more stable time increments.
Monitor the total energy balance in the GLSTAT or SLEOUT files. If the negative contact energy mirrors a positive increase in internal energy locally, the problem might be localized, but still requires investigation.
Ensure that hourglass energy is a small percentage of the internal energy, as excessive hourglassing can also indicate numerical issues.
By addressing these issues, you can improve the physical accuracy and stability of your simulation results.
Sincerely,
James M. Kennedy
KBS2 Inc.
November 17, 2025
--
You received this message because you are subscribed to the Google Groups "LS-DYNA2" group.
To unsubscribe from this group and stop receiving emails from it, send an email to
ls-dyna2+u...@googlegroups.com.
To view this discussion visit
https://groups.google.com/d/msgid/ls-dyna2/CAGHiZ7dD2TjqYtn1mEzTeT%3D-%3DwHk4Em10LeBrHxC4UhxjdrsDg%40mail.gmail.com.