FORMULATION OF ADVANCED FRICTION MATERIALS FOR THE AUTOMOTIVE INDUSTRY

Author

Bipin K C, Southern Illinois University CarbondaleFollow

Date of Award

8-1-2024

Degree Name

Master of Science

Department

Mechanical Engineering

First Advisor

Filip, Peter

Abstract

Today, the friction performance of brake materials is defined by the level and stability of the coefficient of friction, wear propensity, and sustainability [Filip, Friction Science and Applications, MAME, SIU Carbondale, 2023]. The paradigm shift in the transportation industry towards hybrid, electric, and autonomous vehicles demand a regenerative braking system in conjunction with a friction braking system that provides certain advantages like lower wear from friction brake, but also numerous challenges including corrosion and related noise as well as friction stability. In addition to wear from vehicles, large amounts of not used friction materials originating from worn pads and shoes are placed on various waste storage facilities/locations. These could be re-used in optimized new formulations. Recycled friction material mixed with new raw materials can be re-used to make a new formulation. This, however, requires that an optimized optimization process is in place. This thesis will try to explore that opportunity by using the Design of Experiments and Artificial Neural Networks in combination with scaled-down performance testing and analysis of friction surfaces of tested brake pads.The research focuses on the development of advanced friction materials for the brake industry using recycled friction materials that should have a high and stable friction coefficient, low wear, and be environmentally friendly. The basic formulation for the brake pad is developed based on the literature review with the addition of available friction modifiers in the friction industry. By using the Design of Experiments Taguchi Method eight samples were molded using the pre-developed manufacturing setup in the lab. The molded sample bulk and impervious density, porosity, and Shore D hardness were measured and found to be comparable with the commercial brake pads available in the market. After running all model DOE samples in the Universal Machine Tester (UMT) against uncoated C30 Waupaca rotors following the scale-down FMVSS 135 testing standard, the results were analyzed in the Minitab along with wear measurement data. Based on statistical analysis of DOE sample results, the optimized DOE sample 9 was molded and tested in UMT, and the result was close to the commercially available brake pads. The developed friction layer in the pads and rotors was analyzed in the Scanning Electron Microscope to understand the layers' surface chemistry and morphology and the impact of the performance of the tested samples. Artificial Neural Networks on different models were prepared to predict the coefficient of friction of the optimized pads based on the inputs from the composition of the materials and testing conditions such as temperature, force, and torque. The final optimized DOE 9 sample showed high and stable friction performance in the heating snubs section of the testing procedure and had less wear as compared to other DOE samples. SEM and EDX analysis showed the developed friction surface in the pads and rotors. Among different tested models Recurrent Neural Network showed the better result on the prediction of coefficient of friction material.