The joint may be considered a weak point along the pipe and can have a major impact on pipe performance. However, little research has been conducted in regard to joint design. To improve current structural design criteria, this thesis presents the findings and conclusions of experimental and computational studies of the effects of longitudinal bending on joints in rigid (reinforced concrete) and flexible (corrugated steel and thermoplastic) pipes. Solutions for expected shear force, longitudinal bending moment and rotation across the joint for rigid and flexible pipes are formulated for use in structural design of pipe joints. Exact algebraic solutions for rigid pipes are derived using the beam-on-elastic-spring approximation. The formulations for flexible pipes are developed assuming that the two pipes are very long, and that the response is not affected by either the location or characteristics of those other joints. The flexible pipe design equations are developed using various closed form solutions for beams on elastic foundations developed by Hetényi (1948). Parametric studies are then presented where the key factors controlling the behaviour are examined, and the comparisons to recent experimental measurements show that the joint rotation calculated using the equations and a value of soil stiffness proposed for use in design are generally reasonable and conservative compared with the laboratory tests. To measure the capacity of the joint to accommodate the demands, a pipe joint testing frame has been designed to facilitate joint characterization experiments. Shear tests and articulation (rotation) tests have been conducted using this testing frame to examine the shear force capacity, longitudinal bending moment capacity of moment-transfer joints, and rotational characteristics of joints. The shear test results show that the joints of PVC pipe and corrugated steel pipe have similar shear stiffness while the reinforced concrete pipe joint is significantly stiffer. The results of the articulation testing indicate that the rotational capacities of the three joint systems are similar in general. Design of rotational capacity of these joints is likely dominated by considerations of assembly in the field, rather than the rotational capacity that is needed once the pipes are installed. / Thesis (Master, Civil Engineering) -- Queen's University, 2013-01-31 15:33:41.058

No description available.

Current fire induced burnthrough models of aluminum typically rely solely on temperature thresholds and cannot accurately capture either the occurrence or the time to burnthrough. This research experimentally explores the fire induced burnthrough phenomenon of AA6061-T651 plates under multiple sized exposures and introduces a new burnthrough model based on the near melting creep rupture properties of the material. Fire experiments to induce burnthrough on aluminum plates were conducted using localized exposure from a propane jet burner and broader exposure from a propane sand burner. A material melting mechanism was observed for all localized exposures while a material rupture mechanism was observed for horizontally oriented plates exposed to the broader heat flux. Numerical burnthrough models were developed for each of the observed burnthrough mechanisms. Material melting was captured using a temperature threshold model of 633 deg C. Material rupture was captured using a Larson-Miller based creep rupture model. To implement the material rupture model, a characterization of the creep rupture properties was conducted at temperatures between 500 and 590 deg C. The Larson-Miller curve was subsequently developed to capture rupture behavior. Additionally, the secondary and tertiary creep behavior of the material was modeled using a modified Kachanov-Rabotnov creep model. Thermal finite element model accuracy was increased by adapting a methodology for using infrared thermography to measure spatially and temporally varying full-field heat flux maps. Once validated and implemented, thermal models of the aluminum burnthrough experiments were accurate to 20 deg C in the transient and 10 deg C in the steady state regions. Using thermo-mechanical finite element analyses, the burnthrough models were benchmarked against experimental data. Utilizing the melting and rupture mechanism models, burnthrough occurrence was accurately modeled for over 90% of experiments and modeled burnthrough times were within 20% for the melting mechanism and 50% for the rupture mechanism. Simplified burnthrough equations were also developed to facilitate the use of the burnthrough models in a design setting. Equations were benchmarked against models of flat and stiffened plates and the burnthrough experiments. Melting mechanism burnthrough time results were within 25% of benchmark values suggesting accurate capture of the mechanism. Rupture mechanism burnthrough results were within 60% of benchmark values. / Ph. D.