T-Bracing for Stability of Compression Webs in Wood Trusses

Abstract

Compression-web members in metal-plate-connected wood trusses may buckle elastically as long slender columns. One common construction technique is to nail a side member on the compression-web member (main member) so that the cross section is “T” shaped, thus increasing the cross-sectional dimensions and the critical buckling load. The bracing efficiency of the T-brace is not known. The objective of this study was to assess the influence of the main-member length, side-member length, and connection method on T-brace efficiency by measuring the first buckling-mode critical load. Finite-element models were developed with simply supported ends and side members that varied from full length to one-quarter length. The models had side members attached with either 16d common nails or phenol-resorcinol adhesive. The results of a finite-element investigation were used to determine which T-brace geometries to test. The testing program included full-length, half-length, and quarter-length side members nailed or glued to 1,830 and 3,050-mm main members. The testing results paralleled the finite-element results. The T-brace assemblies exhibited a linear minor-axis bending deflection from the onset of loading until an instability load was reached; the deflection then increased without bound. Some slipping between the main and side members was observed in the nailed assemblies, and twisting was evident in the deformed shape of the adhesive-attached assemblies. T-bracing was most efficient for very long members; its effectiveness increased with increasing side-member length. Adhesive-attached side members were more effective than nail attached. The effects of end geometry and out-of-plane metal plate bending were not investigated but could have a substantial effect.