There is currently no way, other than directly adding rigid body members with the necessary releases, to simulate an an axial angle or tee member (brace) under eccentric axial load. This causes a high amount of manual calculations to determine the true utilization of these cross-sections, as it will experience the axial load and an induced eccentric moment due to connecting these sections to one of their flanges and away from the section's elastic centroid.
Can a new specification be assigned (say "Eccentric Axial Specification", with options of either Mz (geometric) for tees/angles or My (geometric) for angles only, with a field to add in the connection thickness) to allow one of these geometric axes get a P*(e + eccentricity to connection material's elastic centroid) load applied in the load matrix and subsequently designed?
If too difficult above, can a design parameter be added in the code check to add P*e to the "required" moment demand (applied in the geometric axis, and converted to principle-axis moments)? User would specify ez (geometric) or ey (geometric) to be put under consideration (= e + eccentricity to connection material's elastic centroid).
Either of these options would allow eccentric tees and angles be designed within STAAD without manual calculation. The first would be best, but I understand is more complex. Second option is a great alternative.
Third time the charm (sorry)
See correct screenshot; there were some ambiguities
Hi Carlos,
This link shows a similar connection: https://structuraldetailer.com/wp-content/uploads/2021/06/Vertical-Bracing-Connections_2.png
Single angles and tees are typically connected eccentrically from a column or beam centerline. There are two avenues that can be taken: the connection and column/beams resolve the eccentricity, or the brace is designed for the axial load + eccentric moments. A lot of the time, especially for braced steel structures, we would like to have the brace be designed to take this eccentric moment instead of the connection/columns/beams.
It is correct that we can use an offset command in STAAD, however this does not allow the brace to take the eccentric load, as end releases are applied to the member attaching to the fictitious offset member. To make this into the analysis file, there would need to be a way to assign the releases to the fictitious offset member at the node, and leaving the brace member-to-fictitious offset line as fix-fix. I attached an image to visually describe (a) a centroid case (base case, no offsets and concentric loads only), (b) current staad capability for using offset commands and implications when trying to model an eccentric brace, and (c) the desired analytical configuration to allow a brace to have eccentric moments applied to it.
For (c), the only way currently is to model the offset as a physical member, assign it very stiff properties like a rigid body, and apply the releases; this overly-complicates analysis models, difficult to check, and subject to issues.
I believe trying to do this type of analysis would be difficult to implement in the processing step; likely it is much easier to handle as a design parameter to tell STAAD to add an eccentric moment of P*e during code checks, but in the analysis file only the "P" load is shown (if that makes sense)
Thank you for posting this idea. I am trying to better understand the situation you are attempting to replicate. Is it say a brace formed from an angle profile that is fixed to the flange of a column formed from a wide flange such that you expect that the axial force in the angle would induce a torsion in the column? Therefore if that is concern would you not define an offset on the end of the braces (angles) to have their centroid align with the flange of the column (wide flange)?