Search

The paper describes load and resistance factor design (LRFD) calibration for the resistance factor used in the Stiffness Method internal stability soil failure limit state for geogrid mechanically stabilized earth (MSE) walls. The Stiffness Method was recently adopted in the current American Association of State Highway and Transportation Officials LRFD Bridge Design Specifications in the US, and will appear in the next edition of the Canadian Highway Bridge Design Code. The paper describes the details of the calibration of the soil failure limit state which is unique to the Stiffness Method. Calibration outcomes include consideration of the concept of level of understanding in the selection of nominal load and resistance values which is unique to LRFD foundation engineering practice in Canada. A practical conclusion from these calculations is that if product line-specific creep test data are available to estimate the reinforcement secant creep stiffness used for design, then a resistance factor of 1.0 is reasonable for US practice. If only minimum average roll value tensile strength data are available, then a value of 0.95 is recommended for US practice. For Canadian practice, the corresponding values for typical level of understanding are 0.90 and 0.85, respectively. [ABSTRACT FROM AUTHOR]

The next edition of the Canadian Highway Bridge Design Code will contain a table of geotechnical resistance factors to be used for seismic design. This paper will estimate the geotechnical resistance factors for shallow foundations required to achieve various target maximum acceptable failure probabilities, which in turn may depend on the assumed design earthquake return period. The investigation will include consideration of design lifetime, uncertainty in the magnitude of the maximum lifetime earthquake event, and the uncertainty in ground properties. The results suggest resistance factors that are lower than commonly used at the moment in Canada and that the failure probability is not greatly dependent on the return period of the design earthquake. The paper will present recommendations on geotechnical resistance factors for seismic design that can be used to guide and calibrate future editions of civil design codes in Canada. [ABSTRACT FROM AUTHOR]

This paper describes the development of a unique rolling load simulator (ROLLS) for testing bridge superstructure with a footprint up to 4 m ×17 m, and its first application to test a full-scale 1220 mm ×900 mm ×16000 mm B900 prestressed concrete box girder. This facility at Queen's University in Kingston, Ontario, is the first of its kind in Canada. ROLLS can apply cyclic loading in a controlled laboratory environment, under realistic highway scale 'rolling wheel loads', in lieu of the conventional 'pulsating stationary loads'. It has two half-axles of a large tandem, each comprising a dual 1140 mm diameter air-inflated tires spaced at either 1.2 or 2.4 m. Each half-axle can apply up to 125 kN, representing the heaviest half-axle load of the CL-625 design truck of the Canadian Highway Bridge Design Code (CHBDC). The maximum travel range and speed are 14.9 m and 6 m/s, respectively. A case study involving analysis of a bridge with eight adjacent B900 box girders of 27.6 m span was carried out prior to experimentally testing one of the girders using ROLLS. Load distribution analyses were conducted using both (i) a finite element model of the full bridge under various CL-625 truck loading configurations and (ii) the CHBDC load distribution method, and both agreed well. Scaling analysis of the girder load share was then conducted to account for shortening it to 16 m to fit in the laboratory, resulting in two-115 kN ROLLS design loads, 1.2 m apart. Multiple passes were conducted at various loads of 40%–100% of the design load, at speeds of 1–5 m/s to examine the machine and girder behaviours. It was found that the applied load fluctuates by less than 10% of full capacity and a 0.13 s/cycle time lag occurs. The measured girder deflection and elastic strains were 11%–20% lower than predicted theoretically. With the two half-axles assembly spaced at 1.2 m, the apparatus has the ability to complete three million cycles in approximately 4.5 months if ran continuously at 5 m/s. [ABSTRACT FROM AUTHOR]