Modeling, Design and Validation of an Exhaust Muffler for a Commercial Telehandler 2009-01-2047

This paper describes the design, development and validation of a muffler for reducing exhaust noise from a commercial tele-handler. It also describes the procedure for modeling and optimizing the exhaust muffler along with experimental measurement for correlating the sound transmission loss (STL). The design and tuning of the tele-handler muffler was based on several factors including overall performance, cost, weight, available space, and ease of manufacturing. The analysis for predicting the STL was conducted using the commercial software LMS Virtual Lab (LMS-VL), while the experimental validation was carried out in the laboratory using the two load setup. First, in order to gain confidence in the applicability of LMS-VL, the STL of some simple expansion mufflers with and without extended inlet/outlet and perforations was considered. The STL of these mufflers were predicted using the traditional plane wave transfer matrix approach. The geometry of these mufflers was then created, meshed in Hypermesh and imported into LMS-VL. Transmission loss data were computed using the boundary element method. The results of both methods were compared to show correlation between the computed values and the values obtained using the analytical transfer matrix approach.

A 3-D model of the designed tele-handler muffler was created and analyzed in LMS-VL for predicting its performance and for fine tuning the muffler for optimum design. Geometry of this muffler was created and meshed using Hypermesh and since this muffler consisted of complicated internal components like resonators and perforated tubes, the STL was computed using the finite element module in LMS-VL. A prototype of the final design was fabricated and tested in the laboratory to measure STL using the two load setup. The engine noise was used as input through the inlet side of the muffler to simulate the actual environment. The STL was computed by measuring transfer functions between the two microphones placed at inlet as well as outlet of the muffler. The measured transmission loss data were compared to that obtained from the numerical method. The correlation between the numerical and experimental results was fairly good demonstrating that STL can be computed accurately and easily using the LMS-VL modeling method.