Description:
The advent of microprocessors has made modern high-speed spectral analyzers possible and has given the test engineer new and powerful techniques for the analysis of structures. However, with new techniques have come new difficulties to overcome in obtaining valid data. This paper stresses the input/output nature of dynamic system measurements and builds upon this concept for the determination of valid data. First, consideration is given to various excitation techniques (swept sine, random noise, impulse) in addition to self-excitation. The distribution of energy in the frequency domain of the excitation signal offers insight as to what may be expected in the output measurement. Undesirable energy in a selected frequency range may need to be filtered out to better match the dynamic range of a signal with the input amplifier gain. Noise, which typically occurs on every signal, affects measurement accuracy. Averaging techniques are described which can sometimes help reduce the effects of noise and improve the measurements. In two-channel measurements, the use of the coherence function as a measure of the goodness of frequency response data is also introduced.