The overall objectives of this thesis are to determine if Indirect and Direct Design methods currently used for reinforced concrete pipe are able to accurately predict the capacity of the pipe, to identify discrepancies between the two methods, and to provide potential modifications to the methods to reduce inconsistencies. As part of this investigation, two 0.6 m pipes (nominal strength classes 100-D and a 140-D) and two 1.2 m pipes (a 65-D with Wall B and a 65-D with Wall C) were tested under single wheel pair loading at burial depths of 1.2, 0.6 and 0.3 m. The test pipes did not crack at the applied service load of 110 kN and did not pass the crack width limit until between 2.5 and 4 times the service load. A 0.6 m 100-D pipe was also tested under simulated deep burial and it was found that the calculated test D-Load is 1.9 times greater than the designated D-Load of the test pipe. It was found that both methods were conservative and that the Direct Design method should be modified to more closely align with the Indirect Design. An investigation of the Direct Design parameters found that by considering thick ring theory and the Modified Compression Field Theory with two layers of reinforcement, the required amount of steel from Direct Design could be made to align very closely with the Indirect Design. An additional test was completed to further assess the Direct Design method on a 0.6 m 140-D pipe to measure the pressure around the circumference of the pipe and compare this measured pressure to the commonly used pressure distribution for Direct Design. The results show that at the minimum cover (0.3 m) the test pressure is higher than predicted at the crown, lower than predicted at the invert, and nearly zero at the shoulder, springline, and haunch, which is inconsistent with most of the predicted results at these locations. / Thesis (Master, Civil Engineering) -- Queen's University, 2014-06-20 16:29:39.037

The goals of this research are to develop and validate the use of distributed fibre optic sensors for use in strain monitoring of buried culverts, and to use full-scale experiments to evaluate the performance of both deteriorated steel and reinforced concrete culverts rehabilitated with grouted slipliners subjected to surface loading. Bench scale experiments were conducted to evaluate the use of fibre optic sensors against conventional strain sensors. Then, fibre optic sensors were attached to a full-scale culvert that was tested in a buried state as a proof of concept. Finally, fibre optic sensors were used in two large scale buried pipe tests to explore the performance of rehabilitated flexible and rigid culverts. A deteriorated steel culvert was tested in a buried state under surface loading, then rehabilitated with a grouted high density polyethylene (HDPE) slipliner while still in a buried state and tested under surface loading at 0.9 m and 0.6 m burial depths. The rehabilitated steel pipe was tested under service loading, and up to 1250 kN of applied load. The results suggested that the grouted annulus stiffened the overall structure, and increased the capacity of the system to over 3 times the fully factored design load. A deteriorated reinforced concrete culvert was tested and rehabilitated in a similar fashion. The grout in the annulus penetrated the cracks at the crown, invert and joint of the concrete pipeline. The lined concrete pipe was tested to 1200 kN under single axle loading, and to 800 kN under single wheel loading. The results suggested that while the concrete pipe was stiffened by the grout, it remained the primary contributor to structural capacity, with the liner contributing little to the capacity. Repair reduced the diameter change by an average of 90%, with the capacity reaching approximately 3.3 and 4.2 times the design loads for single axle and single wheel pair loading, respectively. The maximum response was under single axle loading over the barrels of the concrete pipe. In no instance did the structures reach an ultimate limit state, and the tests were stopped after bearing failure of the soil occurred. / Thesis (Master, Civil Engineering) -- Queen's University, 2013-11-28 17:24:50.815

O procedimento usual no Brasil para o projeto de tubos de concreto enterrados é o procedimento de Marston-Spangler. Este trabalho avalia o comportamento deste procedimento com o procedimento padrão recomendado pela ASCE (American Society of Civil Engineer), denominado de SIDD (Standard Practice for Direct Design of Buried Precast Concrete Pipe Using Standard Installations). A ferramenta de análise dos dois procedimentos é o método dos elementos finitos, o programa SSCOMPPC. São feitas várias simulações numéricas das instalações submetidas às mesmas situações para que assim se possa avaliar o comportamento de cada umas delas. O trabalho conclui que a falta de parâmetros para a execução do procedimento da Marston-Spangler pode comprometer o comportamento desta. Quando comparado as classes de instalação de A a D de Marston-Spangler com as instalações padronizadas SIDD tipo 1 a 4, respectivamente, as instalações SIDD tipo 3 e 4 possuem uma melhor distribuição de esforços do que as instalações classe C e D. Quando comparado a forma de projeto habitual no Brasil, o método indireto do procedimento de Marston-Spangler, com o método direto, tem-se que o método direto acarreta numa grande economia de armadura. Esta economia para o procedimento padronizado SIDD vai de 81,1% a 97,1% de armadura, e em relação com o método direto para o procedimento SIDD a economia vai de 54,4% a 93,1%, para o caso de aterro de 3 m de altura e tubo de 1200 mm de diâmetro interno. Com esta economia e com os recursos computacionais disponíveis hoje, o cálculo pelo método direto é vantajoso. Na comparação dos procedimentos de Marston-Spangler com o procedimento padronizado SIDD, pelo método direto, o procedimento de Marston-Spangler leva vantagem na classe B, sendo que a instalação tipo 2 do SIDD acarreta num consumo em média de 72,5% a mais de armadura. Porém quando a classe C e classe D, que são as mais executadas, são comparadas com as instalações tipo 3 e tipo 4, as instalações tipo 3 e tipo 4 consomem menos armadura, em média consomem 43,8% e 55,6% menos armadura do que as instalações classe C e classe D, respectivamente. / The most popular procedure in Brazil for the buried concrete pipe design is the Marston-Spangler\'s procedure. This work compares the performance of the Marston-Spangler\'s procedure and SIDD (Standard Practice for Direct Design of Buried Precast Concrete Pipe Using Standard Installations), recommended by ASCE (American Society of Civil Engineer). This work uses the finite element method, using the software SSCOMPPC, to analyse the two procedures. It is made many diferent simulations using the finite element method to analyse the behaviour of each procedure. The work concludes that Marston-Spangler\'s procedure needs more paramether to define each class of instalation. When compared the classes A to D of the Marston-Spangler\'s procedure with the type 1 to type 4 of the SIDD procedure, the work concludes the type 3 and 4 have a bether performance compared to the classes C and D. When compared the most popular procedure in Brazil, the Marston-Spangler\'s procedure using the indirect method, with the direct method, the direct method has a better performance. The economy of reinforcemet goes from 81,6% to 97,1% using the SIDD procedure and goes from 54,4% to 93,1%, using the Marston-Spangler\'s procedure (direct design). With the computational resources avaliable today, the direct design is a great advantage. Comparing the procedures of Marston-Spangler and SIDD, using the direct method in both procedures, the class B of Marston-Spangler\'s procedure has a better performance compared to the type 2 of SIDD procedure. The economy of reinforcement for this case is 27,5%. But the type 3 and 4 have a better performance compared to the class C and D. On type 3 and 4 installations, the economy of reinforcement is 43,8% and 55,6% compared to classes C and D, respectively.

O procedimento usual no Brasil para o projeto de tubos de concreto enterrados é o procedimento de Marston-Spangler. Este trabalho avalia o comportamento deste procedimento com o procedimento padrão recomendado pela ASCE (American Society of Civil Engineer), denominado de SIDD (Standard Practice for Direct Design of Buried Precast Concrete Pipe Using Standard Installations). A ferramenta de análise dos dois procedimentos é o método dos elementos finitos, o programa SSCOMPPC. São feitas várias simulações numéricas das instalações submetidas às mesmas situações para que assim se possa avaliar o comportamento de cada umas delas. O trabalho conclui que a falta de parâmetros para a execução do procedimento da Marston-Spangler pode comprometer o comportamento desta. Quando comparado as classes de instalação de A a D de Marston-Spangler com as instalações padronizadas SIDD tipo 1 a 4, respectivamente, as instalações SIDD tipo 3 e 4 possuem uma melhor distribuição de esforços do que as instalações classe C e D. Quando comparado a forma de projeto habitual no Brasil, o método indireto do procedimento de Marston-Spangler, com o método direto, tem-se que o método direto acarreta numa grande economia de armadura. Esta economia para o procedimento padronizado SIDD vai de 81,1% a 97,1% de armadura, e em relação com o método direto para o procedimento SIDD a economia vai de 54,4% a 93,1%, para o caso de aterro de 3 m de altura e tubo de 1200 mm de diâmetro interno. Com esta economia e com os recursos computacionais disponíveis hoje, o cálculo pelo método direto é vantajoso. Na comparação dos procedimentos de Marston-Spangler com o procedimento padronizado SIDD, pelo método direto, o procedimento de Marston-Spangler leva vantagem na classe B, sendo que a instalação tipo 2 do SIDD acarreta num consumo em média de 72,5% a mais de armadura. Porém quando a classe C e classe D, que são as mais executadas, são comparadas com as instalações tipo 3 e tipo 4, as instalações tipo 3 e tipo 4 consomem menos armadura, em média consomem 43,8% e 55,6% menos armadura do que as instalações classe C e classe D, respectivamente. / The most popular procedure in Brazil for the buried concrete pipe design is the Marston-Spangler\'s procedure. This work compares the performance of the Marston-Spangler\'s procedure and SIDD (Standard Practice for Direct Design of Buried Precast Concrete Pipe Using Standard Installations), recommended by ASCE (American Society of Civil Engineer). This work uses the finite element method, using the software SSCOMPPC, to analyse the two procedures. It is made many diferent simulations using the finite element method to analyse the behaviour of each procedure. The work concludes that Marston-Spangler\'s procedure needs more paramether to define each class of instalation. When compared the classes A to D of the Marston-Spangler\'s procedure with the type 1 to type 4 of the SIDD procedure, the work concludes the type 3 and 4 have a bether performance compared to the classes C and D. When compared the most popular procedure in Brazil, the Marston-Spangler\'s procedure using the indirect method, with the direct method, the direct method has a better performance. The economy of reinforcemet goes from 81,6% to 97,1% using the SIDD procedure and goes from 54,4% to 93,1%, using the Marston-Spangler\'s procedure (direct design). With the computational resources avaliable today, the direct design is a great advantage. Comparing the procedures of Marston-Spangler and SIDD, using the direct method in both procedures, the class B of Marston-Spangler\'s procedure has a better performance compared to the type 2 of SIDD procedure. The economy of reinforcement for this case is 27,5%. But the type 3 and 4 have a better performance compared to the class C and D. On type 3 and 4 installations, the economy of reinforcement is 43,8% and 55,6% compared to classes C and D, respectively.