An important addendum to the free air material on the bare driver is:
RD75 driver S/N...54 in free air at 3m - High-Resolution
This is a high resolution (0.5Hz) 3m plot with extension to 50Hz and a
highly compressed amplitude scale to show the final fallrate of the
driver below it's low frequency resonance point. There is no
smoothing in the plot. Note that the cursors are positioned over the
octave 325-650Hz, and that the amplitude fall in that octave is
5.7dB. This is very close to the theoretically expected -6dB/octave
fallrate, and given that the driver's Q at resonance is high the
descrepancy is minor.
Yet in the octave 50-100Hz is measured an amplitude reduction of 30.9dB! This far exceeds the theoretical expectation of -18dB/octave. One might suppose that the tensioned membrane diagphram is being excursion limited in this range? This extreme fallrate has direct implications for crossover/woofer integration when the driver is used to it's maximum low frequency limit (125Hz).
Here is an overlay comparison of that high resolution (0.5Hz) data with the 3m data of the driver on the floor:
Comparison: Bare RD75, suspended and on floor. No smoothing.
In free air the response over the octave 325-650Hz was seen to fall -5.7dB, on the floor the fall is -6.08dB. This is very close to the value expected from dipole theory. In the octave 50-100Hz the free air response fell -30.9dB, on the floor the response fell only -21.25dB. A not particularly useful increase given the required minimum crossover frequency of 125/150Hz, but the value is much closer to the expected -18dB/octave fallrate expected from theory.
In the final analysis, floor placement compared with free air use of standalone dipole baffles using the RD75 driver produces relatively trivial effects on the frequency response.
18"W Two-sided Baffle, RD75 BG S/N..54 in free air.
This two-sided baffle dates back to last June/July. It is a 1.5"
thick baffle using a composite constuction for low resonance. For
this test the edges were left bare, and are hence square and should
therefore produce maximum diffraction effects.
18"W Two-Sided baffle in free air at 1m, 2m, and 3m - unsmoothed responses.
18"W Two-Sided baffle in free air at 1m, 2m, and 3m - 1/12 octave smoothed responses.
Analysis:
John's Bump appears in all it's glory in the three plots. Take serious note of something very unusual occurring in the frequency response below 1000Hz. Further commentary is left for later as the 'arguement' builds during presentation of the remainder of the data. Comparison plots across conditions will be presented later.
The cavity resonance remnant at ~6KHz takes on it's fully rounded form in the 2m and 3m plots. It is understood that this is a correctable problem, best achieved through electronic equalization. The passive notch filter is far from a perfect solution. Cavity resonance is the natural amplification of the signal that occurs from it's generation within an enclosure. In this case the inside of the clamshell steel enclosure that forms the RD75. It's frequency range is determined by the internal dimensions of the case.
The frequency response of the two-sided baffle, like the bare driver, changes with distance. It will be seen that this is the rule. The Proximity Effect is still evident, as is the first inkling of a 'Moving Dip' at ~1000Hz.
Next time a one-sided 'Blondia Baffle' in free air.
John Whittaker
Dipole Baffle Study Report#17.
Dipole Baffle Study Report#19.
The RD75 Dipole Baffle Study - Table
of Contents
Acoustic Line Source Research -
Table of Contents.
