Objective - This project will use experimental data being gathered in ongoing testing in a parallel extramural project to identify unanticipated failure modes, develop accurate behavioral models and corresponding design rules for deep, slender wide-flange steel beam-columns in earthquake-resistant construction and for the base plate connections that attach them to supporting foundations. The models developed will assist designers in characterizing the earthquake behavior of these elements, to support both advanced Performance Based Seismic Engineering (PBSE) design methodologies and provide recommended refinements in current seismic provisions in US model building codes and design standards.
What is the new technical idea? In modern earthquake-resistant steel frame design, lateral force resistance is often concentrated in a few frames or bays in frames. Columns in these frames are subject both to significant axial force (particularly in lower stories of buildings) and to bending forces. Accordingly, these members are referred to as ‘beam-columns’. Increasingly, deeper and more slender wide-flange sections are being selected to increase the in-plane stiffness of a frame in the most economical manner. This increased stiffness reduces earthquake motion-induced drift, but member cross-section characteristics can lead to member instabilities as demands increase.
A closely related technical issue is that of column base plate design. The connection of a column to its foundation is critical in developing the full capacity of a beam-column under significant earthquake demands; yet, this is an area that has not received significant attention. Current methods for designing base plates and foundation anchorage for steel beam-columns are not comprehensive, and failures of base plates, plate-to-column connections, or plate anchorages can compromise the development of the yield mechanism assumed during design. Anchor bolt damage was observed after the 1994 Northridge and 1995 Kobe (Japan) earthquakes (to list two recent examples); however, design provisions for column-to-base connections have not changed in any significant manner as a result of post-earthquake observations. Therefore, NIST will identify unanticipated failure modes associated with deep beam-columns and develop design strategies that can eliminate these modes for strong earthquake motions.
Available relevant test data on these types of deep, slender beam-columns and their base plate connections subjected to cyclic-type loadings are limited, as identified in NIST GCR 11-917-13 (NIST 2011). The NIST BSSC R&D Roadmap (NIST 2013) also identified improving base plate design as a high priority research need. Expanding the available test data supports enhancing the accuracy of building codes and design standards, developing improved design and assessment provisions, and assisting designers in identifying behavior as required for nonlinear PBSE—ASCE/SEI 41 (ASCE 2014) and FEMA P-58 (FEMA 2012)—analysis.
What is the research plan? This research involves a coordinated experimental and analytical study that is outlined in NIST GCR 11-917-13. A NIST-funded FY 2012 task order to the NEHRP Consultants Joint Venture (NCJV) supported a comprehensive experimental research program to characterize the behavior of 25 deep, slender beam-column members that are considered stability-sensitive at large deformations. These 25 tests are grouped as “Phase 1” in a comprehensive extramural testing series; this experimental work was concluded in June 2015. Testing of an additional number of specimens (Phase 2) is planned for contracting in late FY 2015. This second phase of tests will provide supplementary data points to facilitate targeted changes to seismic design provisions in U.S. building codes and standards. Several specimens in Phase 2 will be identical to those previously tested either in Phase 1 or Phase 2. These tests will provide critical data for examining the influence of experimental uncertainty on developing appropriate new seismic provisions. The University of California at San Diego (UCSD) conducted the Phase 1 tests under a subcontract from NCJV because of the nationally unique testing facility and qualified staff. It is anticipated that UCSD will also perform Phase 2 tests under a subcontract from the Applied Technology Council (present NEHRP IDIQ Contractor), to ensure continuity and consistency with the Phase 1 tests. The unique testing equipment and corresponding loading protocol for testing the beam-column elements ensures that plastic hinges form, enabling analysis of beam-column inelastic behavior under different combinations of axial load and bending moments, and providing interaction information in the inelastic regime; such data are currently unavailable.
Work performed by NIST will complement the extramural experiments with analytical modeling of the beam-columns to develop validated design relationships (including axial load-bending moment interaction modeling equations for model building code provisions) and to develop modeling guidelines for beam-column analysis. Numerical modeling will employ advanced nonlinear finite element analysis to evaluate the experiments and to assess the larger question of inelastic stability of simple frames.
Following the work on beam-columns, advanced nonlinear finite element analysis will also be used to assess how base plate configuration interacts with demands from the tested beam-columns, to evaluate demands on the anchorage, and to determine problematic areas with current design provisions. Test data will be evaluated and analyzed to identify parametric relationships between base plates and columns that control performance. The analytical work will be used to expand the range of applicability of design guidance beyond the limited data obtained by tests. Once completed, the observations from analysis and testing will be utilized to provide new, robust design rules that better characterize beam-column and base plate design for implementation in model building codes.