Examination of the Trade-Offs Between Accuracy and Feasibility in Aspects of Ground-Borne Vibration Propagation Testing

Federal Transit Administration guidelines describe an empirical method to predict ground-borne vibration levels generated by rail vehicles. This method requires the measurement of soil propagation characteristics that is typically performed by dropping an instrumented hammer on the ground at various positions along the right-of-way to generate a line-source response proportional to the length of a train. The vibration pulses generated by the hammer are measured using ground-mounted accelerometers or geophones, and the soil response is calculated from a ratio of the vibration magnitude and the hammer force. While the method has proven effective, there are aspects of the procedure that can affect the results and subsequent predictions. Rajaram and Somerville have previously investigated measurement issues that can affect data quality and quantified the error caused by common problems. This paper examines three components of the propagation tests with respect to the tradeoffs between practical procedures and accuracy. First, different methods of coupling accelerometers to the ground are compared. Next, the length of the impact line required to generate an accurate estimate of the line source response for a given train length is discussed. Finally, we assess the Rajaram and Somerville opinion regarding the effect of hammer bounces on data quality.