Evaluation of Metal Building System Seismic Response Modification Coefficients
Embargo until
Date
2019-04
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
A seismic evaluation was conducted of common metal building configurations within the probabilistic framework defined in FEMA P695 “Quantification of Building Seismic Performance Parameters” with the goal of evaluating the applicability of current seismic design procedures in the ASCE 7-10 “Minimum Design Loads for Buildings and Other Structures”. The evaluation began with the definition of a performance group of index archetypes which was guided by industry steering group-led design studies that explored the influence of seismic load combinations and geographical location on the primary frame design. It was observed that taller, shorter span buildings in the western U.S. were most sensitive to seismic demands. This led to the definition of a performance group covering a range of natural periods and seismic weights designed to ASCE Seismic Category D and with the seismic response modification factor of R=3.5.
The seismic parameters calculated for the performance group, designed by industry with ASCE Equivalent Lateral Force procedures, were the system overstrength, ductility, and probability of system collapse when exposed to the Maximum Considered Earthquake (MCE) which has a 2% probability of exceedance in 50 years. The system overstrength and ductility for each index archetype were calculated with simulated experiments using thin shell high fidelity simulation, where all metal building components were modeled including the built-up primary frames, the girts and purlins, the exterior metal facade and screw down roof, the rod bracing, and the primary frame flange braces that are important for controlling lateral-torsional buckling. The high fidelity simulation protocol was extensively validated with research between 2006 and 2013 that included monotonic and cyclic primary frame subassembly tests and shake table tests at the University of California, San Diego.
The simulated pushover experiments revealed significant system overstrength in the index archetypes and a post-peak ductile response that was sensitive to the controlling limit state. When primary frame lateral-torsional buckling was controlled by the intermediate flange braces, post-peak deformation was available out to large drifts. Panel zone buckling at the knee of the column/rafter resulted in steady post-peak strength degradation. The heavy wall buildings had a higher pushover strength than the light wall buildings because the seismic design load combinations were more influential on the heavy wall primary frame design.
The same high fidelity models were used to characterize the quasi-static cyclic response for each index archetype using an accepted American Institute for Steel Construction (AISC) industry loading protocol. The cyclic response, including strength and stiffness degradation from local and global buckling and column-rafter knee panel zone tension field yielding, was fit to a nonlinear Single-Degree-of Freedom (SDOF) material model used for incremental dynamic analysis (IDA). The IDA performances from 44 far-field ground motions required by FEMA P695 led to a cumulative distribution function of spectral intensity of the far-field record set which could be used to calculate the median collapse probability for each index archetype.
4
Uncontrolled collapse was never observed for these light buildings in the simulations or the shake table experiments, however fracture was, in the knee-rafter panel zone from shear buckling and in the rafter taper joints after lateral-torsional buckling. Both drift and fracture studies were conducted to settle on a drift-based collapse limit of 4.5% for the performance group.
The collapse margin ratio, defined as the spectral acceleration at median collapse probability to the spectral acceleration from the Maximum Considered Earthquake (MCE) was on average across the metal building performance group higher than the acceptable collapse margin ratio corresponding to a 10% collapse probability, with no outliers greater than 20%. This confirms the viability of the existing ASCE 7 equivalent lateral force seismic design procedures for metal buildings in the performance group considered.
With the seismic evaluation process now established and validated, the metal building industry can now investigate other performance groups with potentially large commercial impact - for example, heavier roof buildings that are outside the limits of current ASCE 7 procedures. A modular metal building seismic performance group also becomes available for study with the verified high fidelity modeling protocol used to perform simulated pushover experiments and to quantify cyclic performance.
Description
NBM Report