Development of a volumetric strain influence ground improvement prediction model with special reference to impact compaction

Abstract

Aubrey Berrangé, a South African roads engineer, invented the impact compactor in 1949 with the intention of achieving improved compaction to greater depths than possible with conventional equipment available at the time. The aim of this dissertation is to present a simple prediction model for the profile of improvement in the ground, using surface settlement as the main input parameter. The model is based on the information reviewed, observation of field data and a static numerical analysis. For simplicity sake, no attempt is made to predict the energy requirement to achieve the input value of settlement. The model is then verified on fifteen impact compaction profiles at six different sites. A 2 ton-meter dropping mass compactor was also used in the verification process with reasonable success. In addition, the model was tested against comprehensive testing performed at a dynamic compaction site with very promising results. The method is also shown to give acceptable results for prediction of density increase during a vibratory compaction trial. It is concluded that the improvement in the ground can be estimated with reasonable success if the surface settlement is monitored, providing strains are taken into account.