During the outdoor tests with changing baffle width, John and I noticed a dip in the response growing deeper with wider baffles. This was new and hinted that the cavity formed behind the ribbons by the lumber used to hold the ribbons and also the slide-in baffles was at least interfering with the measurements.
I wanted to find out if this was indeed the case, so I repeated some of the measurements indoors. I used the same stand for holding the RD 75, used no baffle and made a measurement using the MLS technique. Since LAUD has also a SINE instrument, I repeated the measurement using a gated SINE. Correlation was extremely good but I had all those ugly dips and peaks again. (supplied notch filter was used for suppression of the 5.5 kHz resonance.)
The lumber was removed and the ribbons were suspended from the first floor balcony, keeping the same position. The same measurements were repeated and look, the dips were gone and at least there was a trace of the dipole low frequency cancellation as discussed. Time to trash the lumber and all my previous indoor measurements I collected previously.;-(
The graphs below show the results of both sets of measurements, with and without lumber, using the MLS and the gated SINE instrument. (The traces with "hanging" in the description are without lumber.)
The peaks and dips caused by the lumber are very noticeable, the curves from the "hanging" position are significantly flatter. There is no smoothing applied to above curves as I received quite a few requests to see the "raw"data. The smoothed (1/3) octave version of these graphs are shown below. It is all to easy to "miss" the narrow peaks and dips and to get a "different" impression of the performance.
