spandrell and frame beam produce different results etabs

[Statiker95]

Hello, with etabs, the results I get when I assign spandrell between the curtains seem more conservative than the results I get when I assign the same beams as frames. The periods are lower, the moment and shear forces of the curtains are high. I get 0.15 for bending stiffness modifiers in frame elements. In spandrel, I use stiffness modifiers bending + F11, 0.15. What do you think could be the reason for this different conclusion? I have door openings in the curtain walls. There are 80cm depth beams on them. Their lengths vary between 2m and 3.2m.

 

[CSIESPL INDIA]

1. You indicated using a bending stiffness modifier of 0.15 for both frame elements and spandrels. However, if the effective stiffness of the spandrel is lower due to its connection details or geometry (e.g., door openings), it could lead to increased deflection and thus higher moments and shear forces.
2. The inclusion of F11 (the axial stiffness modifier) along with bending modifiers can further complicate the analysis. If the spandrel’s axial rigidity is significantly different from that of a frame element, this will affect how loads are distributed through the structure.
3. When modeled as frames, loads might be distributed more evenly across multiple members, whereas modeling them as spandrels might concentrate loads at specific points (e.g., where they connect with curtain walls).
4. The presence of door openings can significantly alter load paths and stress distributions within curtain walls. If these openings reduce the effective width or height of your curtain wall panels when modeled with spandrels, it could lead to increased forces being transferred into adjacent structural elements.
5. Frame elements generally provide greater overall stiffness compared to isolated spandrel beams due to their continuous nature and ability to engage more effectively with lateral loads.
6. A lower natural period indicates a stiffer system; hence, if your model with spandrels shows longer periods compared to frames, it suggests that the overall system is less stiff when using spandrels.
7. Confirm that both models use the same analysis type (linear static vs dynamic) and that all parameters such as load combinations are identical.
8. Ensure material properties assigned (like Young’s modulus) are consistent across both configurations.
9. The differences you’re observing between modeling approaches likely stem from variations in structural behavior due to modeling assumptions regarding stiffness modifiers, load paths influenced by door openings, boundary conditions, and dynamic response characteristics inherent in each configuration.