A New Approach to Determine the Stresses in Buried Pipes Under Surface Loading

Abstract

All buried pipes experience loading from the weight of soil overburden. When pipelines cross railroads, roads, parking lots or construction sites, the pipes also experience live surface loading from vehicles on the ground, including heavy construction equipment in some scenarios. The surface loading results in through-wall bending in pipes, which generates both hoop stress and longitudinal stress. Current standards limit the stresses in buried pipes to maximum values in terms of hoop stress, longitudinal stress and combined biaxial stress. An early approach to estimating stresses and deformations in a pipe subjected to surface loads dates back to Spangler’s work in the 1940s. Many models have been developed since then. API RP 1102 provides guidance for the design of pipeline crossings of railroads and highways following the model developed by Cornell University for the Gas Research Institute (GRI). The Cornell model was developed only based on experiments on bored pipes crossing a railroad or a highway at a near-right angle. The live surface loading distribution is also limited to the wheel-layout typical of railroad cars and highway vehicles. Most other existing models only focus on the hoop stress in the pipe. In this paper, a new approach to determine the stresses in buried pipes under surface loading is introduced. The approach is suitable for assessing pipes beneath any type of vehicle or equipment at any relative position and at any angle to the pipe. First, the pressure on the pipe from surface loading is determined through the Boussinesq theory. Second, both hoop stress and longitudinal stress in the pipe are estimated. The hoop stress is estimated through the modified Spangler stress formula proposed by Warman and his co-workers (2006 and 2009). The longitudinal stress, due to local bending and global bending, is estimated by the theory of beam-on-elastic-foundation. The modulus of foundation can be determined through the soil-spring model developed by ASCE. The hoop stress, longitudinal stress and the resulting combined biaxial stress can then be compared against their respective limits from a pertinent standard to assess the integrity of the pipe and determine the proper remediation approach, if necessary. The performance of the proposed approach is compared in this study with the experimental results in the literature and the predictions from API RP 1102.