Seismic Racking Test Evaluation of Silicone Used in a Four-Sided Structural Sealant Glazed Curtain Wall System

ABSTRACT: This paper presents the results of a study in which four-sided structural sealant glazing (SSG) insulating glass curtain wall units were sub­jected to cyclic racking test methods in accordance with AAMA 501.6 testing protocols. The test configuration included three side-by-side primary units and a corner unit. High-resolution cameras were utilized to capture instanta­neous images during the racking test so that the displacement within the sealant joint could be captured and isolated from the displacement of the entire glazed unit. Drift capacity of the system in terms of glass attachment and sealant performance is reported in detail for different levels of racking displacements and boundary conditions. The overall behavior of the system is characterized, and specifically the sealant performance at a corner condi­tion during racking drift is discussed. Additionally, expected strains in the sealant were calculated using a linear-elastic finite element model and were then compared with the strains the actual structural sealant joint underwent during system testing. Silicone sealant damage was evaluated using visual observation before and after cyclic racking. The paper discusses proposed acceptable sealant stress levels for seismic design and the durability of sili­cone used in the SSG system as compared to dry-glazed systems based on glass performance.

KEYWORDS: racking test, seismic evaluation, structural sealant glazing, curtain walls, silicone, Finite Element Modeling

Introduction

There has been an ongoing effort at Pennsylvania State University to develop test data and behavior prediction of different types of glass attachment methods under seismic in-plane racking conditions. A thorough compilation of data and prediction approaches has been published encompassing mechanically cap­tured two-sided and four-sided structural sealant glazing (SSG) designs. An industry need has been identified in particular to further this work to encom­pass the behavior of four-sided SSG curtain wall systems under seismic in­plane racking conditions. Four-sided SSG is a method of attaching glass to the aluminum mullions in curtain wall systems using only silicone sealants and no mechanically captured glazing pockets. This construction method has been used in the United States since 1971 with monolithic glass and since 1978 using insulating glass units [1]. Many buildings in high seismic zones of California were built employing four-sided SSG in the 1980s, with a proven history of suc­cess. Publications documenting glass damage (e. g., glass pieces falling out) in four-sided SSG construction after the Northridge earthquake in 1994 are not available, although one publication [2] mentions sealant damage in one build­ing. In fact, according to EERI [3], most of the glass damage occurred in store­fronts with large glass panes and that glazing systems with silicone sealant on one or more edges generally performed better than dry-glazed systems Despite lack of evidence for poor seismic performance of four-sided SSG, building code officials in some areas, including the cities of Los Angeles and Irvine, noted that four-sided SSG was not explicitly recognized in the California Building Code or the International Building Code, and consequently the system has not been uni­formly accepted in high seismic zones, in particular for healthcare facilities. Some recent research has addressed the topic of two-sided SSG [4,5], making it an easier system to present to building owners and building departments. Much of the existing data regarding four-sided SSG has been limited to a study in the 1990s by Zarghamee et al. [2], a recent study by Memari et al. [6], and many field history cases. There is still a significant need for better understanding of the seismic behavior of four-sided SSG systems. The need for further filling this gap in laboratory research and proving the performance of four-sided SSG has been felt for years. The research work in this paper intends to help fill this gap.

This paper will present actual seismic racking results on a full-scale curtain wall system mock-up that is of a design that would be expected to be installed in a real world project. The system was built by the California glazing sub­contractor, Bagatelos Architectural Glass Systems, in cooperation with Dow Corning and tested at the Pennsylvania State University, using their dynamic racking test facility. The curtain wall system mock-up performance has been evaluated in accordance with AAMA 501.6 test protocols [7]. Additionally, this mock-up included a corner condition to address concerns about differential movement or interaction of adjacent glazing units to either side of a building corner. These results will be compared with the analytical results from a finite

Copyright by ASTM Int’l (all rights reserved); Tue May 6 12:07:08 EDT 2014

Downloaded/printed by

Rochester Institute Of Technology pursuant to License Agreement. No further reproductions authorized.

element (FE) Model. Finally, this paper presents sealant test results, which show the modulus stability and durability of silicone sealants, making it one of the ideal choices of materials for the flexible anchorage system required to attach glass and perform satisfactorily in a seismic event.