2019 CSCE Annual Conference - Laval (Greater Montreal)

2019 CSCE Annual Conference - Laval (Greater Montreal) Conference


Title
FRP Materials for Rehabilitation of Buried Pipes

Return to Session

Author(s)
Ms. Anita Shiny Kanagaraj , Dalhousie University (Presenter)
Dr. Pedram Sadeghian, Dalhousie University
Abstract

Pipes are majorly used as culverts, storm water drains, sewers, oil and gas lines, and water conduits. Over time, these deteriorate and necessitate rehabilitation, in order to increase the service life of structure. Bonding a fiber-reinforced polymer (FRP) liner inside the pipe, is an effective rehabilitation technique to increase the strength and stiffness of deteriorated pipes. To do so, it is very important to know the structural behavior and mechanical properties of the FRP liner subjected to loading. This paper discusses the results of solid wall FRP liners with four layers of FRP, subjected to compressive transverse loading. A customized compression testing machine with string pots to measure the diametrical change and a load cell, was set up to test the liners under parallel-plate loading test method. Four specimens of glass FRP (GFRP) and two specimens of carbon FRP (CFRP) liners, having an average internal diameter of 333 mm were tested to find the diametrical change, stiffness factor (SF) and pipe stiffness (PS) at 5% and 10% diametrical change. Both GFRP and CFRP liners started cracking at springline. Also, the CFRP pipes lifted at the invert with crackling sound of the individual fibers breaking continuously until failure. GFRP liners reached an average peak load of 11.5 kN, whereas the CFRP liners reached 13.0 kN. An analytical model for diametrical deflection both in vertical and horizontal direction and their corresponding strains at spring line and crown/invert were developed to substantiate the geometrical non-linearity theoretically. The model is in good agreement with the experimental data, where the SF at 5% diametrical deflection matched with the model SF for GFRP and CFRP liners with an accuracy of 99.9% and 92.2%, respectively.