Strengthening T-Joints of Rectangular Hollow Steel Sections Using Through-Wall Bolts and Externally Bonded FRP Plates
Aguilera, Jose Jr
MetadataShow full item record
T-joints are common in beam-column connections of steel frames, vierendeel girders and at mid-span of N-trusses. Strengthening the members of these structures increases the demand on the joints, which may require joint strengthening. This thesis examines different strengthening techniques of T-joints of RHS members. In Phase I, the effectiveness of through-wall steel bolts is examined. This is accomplished by controlling the web outward buckling of the chord under the brace axial load. The study examined the effect of the number and pattern of bolts, as well as the web height-to-wall thickness (h/t) ratio of the chord, on strengthening effectiveness. Rectangular 203x76x(3.09, 4.5, and 5.92) mm chord members were tested. The 8 mm diameter steel bolts varied from a single bolt to 15 bolts of various distributions. The joint strength increased by 3.1%, 6.2%, and 29% for chords with (h/t) of 34, 45, and 65, respectively. The number and distribution of bolts had little effect on their effectiveness. In Phase II, similar T-joint specimens were strengthened using adhesively bonded GFRP plates, 9.5 mm thick, of different configurations, and 2 mm thick high-modulus CFRP plates of equivalent stiffness. It was shown that strength gain increases significantly, from 9% to 38%, as (h/t) ratio of the HSS chord increases from 34 to 65. In thin-walled HSS (h/t = 65), retrofitting provided significant gains in strength but not in ductility. In thick-walled HSS (h/t = 34), retrofitting provided little strength gain, but enhanced ductility, especially with properly bonded plates extending on the brace. Generally, plates fractured under local bending or delaminated within plate layers while bond was fully intact. In Phase III, selected configurations of the two retrofitting methods were used in additional T-joints with chord (h/t) ratio of 65, to study their effectiveness in presence of axial compression load in the chord. Two sustained load levels were induced in the chord, representing 45% and 80% of its full axial capacity. The transverse brace load was then gradually increased to failure. The through-wall steel bolts increased the joint capacity by 13% to 25%, depending on the chord’s axial load level, while the bonded GFRP plate increased the capacity by 38 to 46%.