Contribution of Frictional Resistance to Restrain Unbalanced Thrust in Buried Pipelines

Current restrained joint design theories for different pipe materials depend on two primary forms of resistance mechanisms to resist the unbalanced thrust forces that arise at a horizontal bend, valve, tee or other similar sources of thrust in a buried pressure pipeline. The first and most commonly recognized form of resistance is from friction and adhesion resistance at the pipe-soil interface; and the second form is from passive and/or bearing resistance of the pipe against the soil surrounding the pipe. Considering these resistance mechanisms, the various American Water Works Association (AWWA) design manuals appear to have been based on certain simplifying assumptions in arriving at different, yet similar, simple (M11, M23, and M41) as well as somewhat complex (M9 2008) design formulae. These simplifying assumptions have not been adequately documented from basic principles. For friction and adhesion resistance, simplifying assumptions have been made with regard to direction of pipe movement along the pipe legs; direction of the resultant friction (and/or adhesion) force generated; coefficient of friction at the pipe soil-interface; and most importantly a simplified approach to estimate frictional resistance without explicitly considering the normal pressure distribution at the pipe-soil interface. Questions regarding the validity of these assumptions have been raised from time to time, quoting the inconsistencies amongst the various design formulae and the appearance of non- compliance or violation of simple Coulomb frictional analysis principles. It is not clear how the frictional resistance estimated by this simplified approach compares to an estimate using the normal pressure distribution and the frictional characteristics of the pipe-soil interface. This paper critically examines the simplifying assumptions utilized in the current design equations to estimate friction (and adhesion) contribution to thrust restraint design problems and the validity of these assumptions. A similar examination for passive and/or bearing resistance contribution to thrust restraint design problems is planned to be undertaken by the task committee in a subsequent paper.