An Analytical Study of the Direct Method of Measurement of Source Impedance for Time-Variant Sources

摘要

Most fluid machines, for example internal combustion engines and fans, can be considered as periodic noise sources when operated under constant conditions, which allows for a frequency domain representation of the source and the associated acoustic field in the duct. In such a representation, the source is characterised by frequency-dependent values of both strength and impedance. The source impedance is a measure of the reflectivity of the source and knowledge of its form can be gained from experiment. One such experimental technique is the direct method, where an injected signal much larger than that from the original source is employed, and the reflection of this signal by the source is measured, to give the source impedance. The technique is known to work well on time-invariant sources, and for a time-variant source such as an IC engine does seem to always give a positive source resistance. This is in contrast to the indirect method, another experimental technique, which frequently gives irrational negative source resistance values when used on an IC-engine source. This paper addresses the question of whether or not the source impedance values given by the direct method for a time-variant source are of value, or indeed have any meaning.