Predicting performance of buried conduits

TLDR: The capabilities of finite element computer codes FINLIN, CANDE, SSTIP and NLSSIP for predicting performance of buried conduits were evaluated in this paper, where the authors concluded that the Duncan-Chang hyperbolic model is a satisfactory representation of nonlinear soil behavior for routine studies of conduit response.

Abstract: The capabilities of finite element computer codes FINLIN, CANDE, SSTIP and NLSSIP were evaluated CANDE was judged to be the best over-all code for predicting performance of buried conduits A number of improvements to this code were made Example solutions are given to illustrate the effects of conduit stiffness, interface slippage, and soil properties on conduit performance It was found that conventional concepts of soil "arching" are misleading; that slippage at the soil-conduit interface strongly affect the response; and that soil behavior plays the most crucial role in conntrolling performance It was concluded that the Duncan-Chang hyperbolic model is a satisfactory representation of nonlinear soil behavior for routine studies of conduit response Further use of equivalent elastic, overburden dependent, or default values in the extended Hardin soil models is not recommended For prediction purposes, especially to investigate the effects of soil compaction and of localized shear failures, a plasticity model of soil behavior is needed Duncan's (1979) SCI procedure for design of long-span metal culverts with shallow cover was also investigated The procedure provided good estimations of maximum thrust for the problems investigated in this study; however, agreement in both the magnitude of the bending moment and in the form of the bending moment equation was found to be poor, and the proposed safety factor to guard against plastic hinging in the conduit wall is considered to be overly conservative (FHWA)