Incorrect Surface Normal Stress in Harmonic Analysis with Structural Damping - ANSYS 2024R2

KristianHansen

Incorrect Surface Normal Stress in Harmonic Analysis with Structural Damping - ANSYS 2024R2

Problem Description

I have identified what appears to be an error in ANSYS 2024R2's implementation of harmonic analysis using the "full" solution method. The surface normal stress at a boundary does not match the applied pressure load when structural damping is present, which, I believe, violates fundamental equilibrium requirements.

Observed Behavior

1. When applying a pressure of 1 Pa normal to a surface (lets say z-direction):
   • With zero structural damping: sigma_z amplitude = 1 Pa (correct)
   • With non-zero structural damping: sigma_z amplitude > 1 Pa (incorrect)

Expected Behavior

The surface normal stress should equal the applied pressure (1 Pa) at the loaded boundary regardless of damping settings. This is a fundamental requirement for equilibrium that should be satisfied independent of:

• Material properties
• Damping characteristics
• Solution method
• Dynamic effects

Justification

Surface tractions must be balanced by internal stresses at boundaries. This is a basic principle of continuum mechanics that holds true even in dynamic and damped systems. The current behavior suggests an incorrect implementation of either:
1. The boundary condition enforcement
2. The damping terms in the "full" solution method
3. The stress recovery procedure

Simple Test Case Steps

1. Create a simple block geometry
2. Apply fixed support on bottom face
3. Apply 1 Pa pressure normal to top face
4. Set up harmonic analysis using "full" solution method
5. Run two cases:
   • Case 1: Set constant structural damping coefficient = 0
   • Case 2: Set constant structural damping coefficient ≠ 0 (e.g., 0.1)
6. Compare amplitude surface normal stress at the loaded surface

Additional Information

• ANSYS Version: 2024R2
• Analysis Type: Harmonic
• Solution Method: Full
• Element Types: Standard structural elements (can provide specific details if needed)

Would appreciate confirmation if others can reproduce this behavior or clarification if this is an error on my part! If it is in fact an error, I would appreciate feedback on a possible workaround to obtain the correct surface stresses.

Best regards,
Kristian

Erik Kostson

Hi. This forum is for questions on mechanical/spaceclaim/fluent scripting and coding. If you are a current Ansys customer, I suggest you contact your Ansys Account Manager or contact your Tech Support provider for help on this question. If you are not a current customer, I would recommend posting your question on the Ansys Learning Forum: https://forum.ansys.com/