Implications of Design Standard Requirements and Assumptions for Local Buckling Capacity Calculation of Pultruded GFRP Members

First-generation structural design standards for pultruded glass fiber–reinforced polymer (GFRP) materials are now available in North America and Europe. The orthotropic nature of pultruded GFRP (pGFRP) makes it susceptible to local buckling, and variation in available material properties adds uncertainty to the calculation of buckling limit states. Additionally, the method of analysis selected by the designer impacts the reliability of the final pGFRP design. In this paper, it is shown that depending on the combinations of these factors, pGFRP design may be very conservative, resulting in inefficient use of material, or may mask expected in situ structural behavior, resulting in undesirable design outcomes. A multiparameter study of cross-section-dominated stability—flange local buckling (FLB) and web local buckling (WLB)—of pultruded GFRP members is presented. In the context of North American and European design standards, two approaches to calculating the local buckling of cross sections assuming either no rotational restraint from adjacent plates (SS) or elastic rotational restraint from adjacent plates (WS) were carried out. Single-web W-, I,- and C-shapes and double-web box sections were considered, each having a significant range of individual plate slenderness. Additionally, the effects of varying material properties were captured by varying transverse (ET/EL) and shear (GLT/EL) modular ratios. It is shown that the ratio of FLB to WLB critical stress calculated using SS assumptions—already a minimum requirement for design—can be used to differentiate behavior and to assist in defining the means by which local buckling limit states can be mitigated. This paper highlights some of the shortcomings of the first versions of design standards for pGFRP. Pultruded GFRP behavior is very sensitive to cross-sectional shape, and reliability calculations—whether material resistance factors or partial factors—need to be reconsidered to include the effects of cross-sectional shape. The predictions of this parametric study were applied to a data set of experimentally obtained local buckling results to make an assessment of the reliability of extant design provisions.