Optimization of Muffler Acoustics Performance using DFSS Approach

Noise pollution is a major concern for global automotive industries which propels engineers to evolve new methods to meet passenger comfort and regulatory requirements. The main purpose of an exhaust system in an automotive vehicle is to allow the passage of non-hazardous gases to the atmosphere and reduce the noise generated due to the engine pulsations. The objective of this paper is to propose a Design for Six Sigma (DFSS) approach followed to optimize the muffler for better acoustic performance without compromising on back pressure. Conventionally, muffler design has been an iterative process. It involves repetitive testing to arrive at an optimum design. Muffler has to be designed for better acoustics performance and reduced back pressure which complicates the design process even more. A hybrid type muffler is the most commonly used muffler in automotive industry and it plays an important role in noise attenuation by using a combination of impedance mismatch and absorption techniques. In this paper a DFSS approach is developed in order to optimize a hybrid muffler design for a passenger car. DFSS approach has an input, output, control factors and the noise factors for the above problem. Exhaust gas mass flow rate at different engine rpm is the input and the tail pipe noise is the output for the analysis. All the design parameters which affects the output is considered as the control factors and the temperature of the exhaust gas is considered as the noise factor since it is not controlled by the design engineer. Commercial 1D simulation software GT-POWER® is used for this analysis. L18 orthogonal array is developed in order to capture the interactions of all controls factors, its levels and noise factor. Simulation is run for the L18 array for different engine rpm and results are plotted between engine rpm versus tail pipe noise. The tail pipe noises for the different design were studied and lowest one across different rpm is selected. Critical design parameter which affects the tail pipe noise is derived from this simulation. DFSS approach adopted in this paper has provided a better correlation of simulation results with test data. The optimized design shows better acoustic and back pressure performance than the original design. Deployment of advanced software and experimental methods leads to First Time Right product development by effectively reducing valuable design cycle time and can be further used in the field of research for future vehicle programs.