Search Results

Customer feedback indicated that there was a difference in noise levels and frequency content between nominally identical diesel engines manufactured in two different plants. A test program was created to determine whether this perceived difference was real. This program opened the opportunity to investigate several levels of variability in SAE J-1074 noise measurements. From the data, estimates of confidence intervals were developed to predict true mean noise levels in the presence of back-to-back, install-test-remove, and sample-to-sample test variability. The data from this project can be used to estimate the number of tests required in order to achieve a given level of accuracy in estimating the true mean noise level of an engine or of an engine population. The results can also be used to design a test program that will provide a given level of confidence in determining whether there is a statistically significant difference between two engine populations.