PIPE INTERACTION WITH THE BACKFILL ENVELOPE

Abstract

This report summarizes a study of installation practices for buried (culvert) pipes. Current practice was reviewed through a literature search and a survey of users, manufacturers, and others involved in the use of buried pipes. Typical backfill materials were characterized through standard and variable effort compaction tests, CBR tests, penetration tests and one-dimensional compression tests. Standard classification systems were compared and standard groups of backfill materials were evaluated. The soil properties that were used to develop the AASHTO SIDD designs are proposed for use as standard properties for application to the installation of all types of pipes. The constrained modulus is proposed as the standard measure of soil stiffness replacing the empirical modulus of soil reaction. Laboratory soil box tests and full-scale field tests were conducted to investigate soil behavior during installation. Variables include pipe type (concrete, steel, plastic) and size [900 and 1,500 mm (36 and 60 in.)], in-situ soil condition, trench width, backfill type, compactive effort, haunching effort, and bedding condition. The tests showed that all of the backfill-related test variables have a significant effect on pipe behavior. Tests with controlled low-strength materials showed this to be an excellent type of backfill. Computer modelling demonstrated that the finite element analysis can effectively model installation effects as well as effects of the fill over the pipe. The elastic solution for behavior of buried pipe, developed by Burns and Richard, shows promise as a basis for a simplified design method. Recommendations for future practice include the use of soft bedding under the bottom of the pipe and of uncompacted fill over the top. Selection of trench width must consider the ability to place and compact backfill in the haunch zone and at the sides of the pipe. Hand tampers, sized differently for different backfills, were shown to be useful for providing haunching effort. It was shown that relatively small changes in backfill density can have significant effects on backfill stiffness. The project shows that pipe performance is controlled by installation practices. Implementation of a design process that realistically assesses how a project will be built, and construction that understands and implements the design, is imperative for successful long-term culvert performance.