Here are the 'Rosetta Stone' measurements that I have been providing each time that the measurement situation changes. It is all well and good to measure here and there, but the question arises 'What is the reliability and validity?' of the measurements across venues.
What is shown below are the responses from the 1S6T baffle in both the Listening Room and then again, months later, in the gymnasium. The gym data, for this comparison, has been processed to match the limitation imposed by the sound reflection constraints of the Listening Room measurements. That limitation is the nominal 7 milli-second 'reflectionless' data interval available in the Listening Room. Of course the gym provides for much longer reflectionless data samples, but we want to see the comparability of the baffle's responses across conditions, so the gym data is arbitarily constrained in this comparison (on this page only) to the length of the Listening Room data. The other gym data in this section has all been processed with the maximum 'reflectionless' data length avialable for full low frequency response.
Both responses have been 1/12 octave smoothed. The Listening Room data
includes the low frequency attentuation of the 150Hz 2nd order high pass
active filter in-circuit. The gym data has no such active filter in
circuit.
Beyond the low frequency attenuation by the active filter seen in the Listening Room plot the most obvious differences between the two plots is the increased dipole ripple also seen in the Listening Room plot below about 2500Hz. Currently these differences are being 'assigned' to the undefined factor of 'wall loading' in the Listening Room.
It is hoped that further quantification of this phenomena will be gained in future comparsions between baffles measured both at the gym and in the Listening Room. In all, I feel that this comparsion confirms the basic validity and reliabilty of the measuring techniques - in fact, I'm astonished that the responses under such differing conditions are so close.
A final comparison shows what was unknown until the gymnasium measurements - the ultimate low frequency response of the RD75/baffle combinations. The added yellow trace above is the plot of the 1.9KHz sampling rate data from the 1S6T baffle at 3m/1m in the gym. The low frequency sampling rate results in very high resolution data, but requires a very large space for use due to the problems of sound reflections in the data.
It is instructive to note the differences in the low frequencies of the plots where they overlap. It can be seen that the relative short duration (7mS) data of the full-range plots is negatively affected, with the information below 200Hz particularly inaccurate. It is wise to always be skeptical of low frequencies in plots derived through Fast Fourier Transforms, as they are particularly subject to inaccuracy.
The Listening Room Measurements - Table of Contents
The Gymnnasium Measurements - How Big is the Gym?
The RD75 Dipole Baffle Study - Table of Contents
Acoustic Line Source Research - Table of Contents.
