Implementation and evaluation of a fast method for THD analysis of sound transducers under demanding operating conditions

The total harmonic distortion (THD) is an important parameter to specify an instrument performance. In general, this parameter is specified in a single frequency and as the maximum level allowed in the operational range of the device under test (DUT). Müller (2007) formulated the sweep level method for fast THD evaluation, in which THD vs. level is obtained using a single tone modulated by a sweep level envelope. Thus, is possible to evaluate the THD in the function of carrier frequency and also of the input level of the system. Moreover, the relationship between the input and output levels can be used to analyze the non-linear behavior of the DUT. The method was validated using both numerical and experimental comparison with the classical stepped sine method. In the numerical model, a hyperbolic tangent was applied to the excitation signal to simulate non-linear behavior. The most important conclusion obtained refers to the level parametrization that was made in an increment of the signal amplitude per wave cycle, instead of increment per time unit. Using an increment of 0,05 dB per wave cycle, the maximum error between sweep level and stepped sine was less than 0,01%. Using a loudspeaker and microphone, the THD of the system was tested in the frequency range between 125 Hz and 4 kHz. The results for the maximum THD vs level presented strong frequency dependence. Thus, the linear operational range of electroacoustical transducers is frequency dependent and it is important to specify for high fidelity applications, mainly if loud signals must be reproduced by the system. As the next step, it is intended to purpose a standardization of THD measurement and results presentation. So, is possible to offers a more detailed parameters to the customers.