A. Accelerometer measurements

This  is a normal music listening level

Here is a higher signal level then above level 

Here below, are shown the results from the highest signal level applied to de loudspeaker, relative to the two above measurements

* with white - the recorder signal from the accelerometer glued upon the loudspeaker's wood cabin

** other colors- the recorded signals from the accelerometer, glued upon the two different R1's surfaces - exactly in the middle of each R1, at the cross point.

1. The source- a loudspeaker IMF Professional Reference Standard MK IV [a transmission line loudspeaker, down to 17 Hz], positioned at 3,8 meters and 45 degrees lateral to the 12 grouped R1's axis [a matrix of 3 R1 horizontally by 4 vertically ] driven by an amplifier [one channel, 130W, class A] directly from a TEF20 units.

2. The sensors:

a. a PVR1 - OMNI ELCTRET mike , covered with protective air plastic foil in order not to touch the R1's body.

b. a miniature Bruel accelerometer model 4500, serial nr.1804165, driven by the TEF20 towards a DeltraTron power supply WB 1372.

A calibration was not necessary, what was asked were:

a. the relative levels between the loudspeaker's wood cabin - at the lateral wall at the woofer height [the accelerometer glued using double face adhesive tape ]and two places upon the outside R1's surface, and...,

b. the relative levels between a multitude of positions, recorded by the above mike inside the R1's column long of 232 cm, the loudspeaker being at 3,8 meters away from the R1's grouped surface .

 Conclusions:

 A.  Accelerometer's measurements

1. In all situations, the displacements recorded by the accelerometer on the loudspeaker's wood cabinet were bigger than that sensed on two different [ but grouped] R1. Please note that each IMF speakers weights about 47 Kg and each R1 only 2 Kg.

2. Somehow, the spectral irregularities recorded on the wood cabin's surface are filtered out upon the R1's surface.

This means that in now way, and at any source's levels, the R1's plastic body resonate even as much as the heavy loudspeaker's cabin, the small displacements waves , following the speaker's ones.

B. Microphone measurements

A laterally column of 3 vertically columns of R1

The central column of a 3 vertically R1's columns [ in the matrix of 3 horizontally x 4  vertically]

B.  Microphone's measurements

1. The low frequency point were the grouped R1 are changing towards the mid and high frequency is 181 Hz.

2. It is clear that the R1 must be kept absolutely empty and not be filled with any kind of porous material, the air back absorption being between 10 and 25 dB lower that the normal low and very low frequencies adaptive tuning.

3. The back air absorption is almost linear and this explain the same almost linear frequency responses as in the older site posted room's polar plots measurements [ obtained with a mike rotated at 37 positions of 5 degree each, at some distance from the R1's surface].

The recorded phase is remarkable stable at all points of measurements.

The superposition principle is somehow accomplished practically, even if extremely hard to be mathematically modeled.

4. As described earlier in the site, each R1's column, behave differently, reacting dynamically with the rooms modes and the always changing stereo signal, much more to the lateral edges and less towards the R1's grouped surface center.

Please note, that in the microphone's measurement case, the frequency resolution is higher at 22.3 Hz than the 52.2 Hz in the accelerometer's case.

For clarity, all measurements were smoothed to 33% or 1/3 octave.