We had in total 4 drivers available for testing. Testing was done in a similar way, MLS with 8 averages, 4096 point FFT with microphone calibration applied. The first graph is the frequency response of the four units without notch filter, clearly showing the cavity resonance. The second graph shows the effect of the supplied notch filter. The peak is greatly smoothed out.
John's drivers were slightly broken in, mine were still wrapped. It would be interesting to repeat this test in a couple of weeks when John's drivers are fully broken in. According to B&G this would take 60 to 100 hours.
Above graph is the super position of the frequency data on 4 B&G RD75 ribbons. There was no notch filter installed, hence the resonance peak at 5 to 5.5 kHz. To make up for the ragged low end of the graph we made a new measurement at a lower sample rate, yielding the curves below. In both cases was the microphone to driver distance 1 meter. Height of the microphone was also 1 meter. The drivers were actually standing on the "floor", much the same way as one would use them in a practical application.
The test results with notch filters are posted below. Be advised that the horizontal scale is not the same as in the top graph on this page. Vertical scale has been kept the same. The notch filter supplied with the B&G RD75 effectively suppresses the resonance at 5.5kHz as shown below.
