Ακουστική/ Acoustics 2002 Patras University
In the beginning
were some
historical moments :
Roundffusor1's announcement at the
University of Patras, September
30, 2002 ,here is a
photo report
The Patras paper in English, published in Proceedings of Hellenic Institute of Acoustics (HELINA), University of Patras, © 2003, pages 41-48, ISBN: 960-7620-27-5 . HELINA is a member of the European Acoustics Associations
Wide band sound diffuser with
self regulated low frequency absorption and methods of mounting
it
ΖΑΙΝΕΑ LIVIU NICOLAE
ESR [EDDY’S STUDIO RECORDING] http://www.zainea.com
E-mail
:
liviu@zainea.com
ABSTRACT
We present here a new kind of a sound diffuser,
enough simple in his construction but very active in doing his job. It is
composed from a main body and two lateral drivers, which are situated on the
inside part of the main body. This two plastic drivers serves to sustain very
steady the diffuser upon the surface to be mounted: walls or ceiling. There are
two wooden rails ‘’T’’ simply mounted on the wall surface. The main diffuser
body with his two drivers is mounted just like a drawer toward the two rails T.
The main body is composed from three basically 3D shapes. Each fourth diffusers,
displays together a new 3D shape at their nearest point. This diffuser may be
used in any kind of room where the critical listening is needed. His main
advantage is that it is working simultaneously in two ways: as a clean diffuser
for mid and high frequencies and as a self controlled absorber, below 250 Hz,
for low and very low frequencies. This is possible, because being mounted in a
new way, his compartmented volumes behind the diffuser surfaces works like a
complex Helmholtz resonator. Each air space behind each diffuser surface,
interact and resonate in a different way. In real time, there is a kind of
reconstruction of all music characteristics – parts diffused or absorbed or
both. Even for the trained ears, there is no time to percept when and how the
music is processed and the listener is just immersed in the sound without any
trace of coloration. The ‘’ historical’’ focus or stereo place of listening,
becomes obsolete. Diffuser dimensions: 58x58x12 cm, vacuum thermoplastic
polystyrene, weight: around 3 Kg (diffuser + 2 woods rails). The Roundffusor1 is
Greek Patent Nr. 1004186/22.05.2002.
Introduction
For many
years I’ve thought that, especially in acoustics, what was to be invented was
invented. The announced ‘’end of science’’ corroborated with an almost
monopolistic worldwide marketing situation, soften the overall ambition for
new. Once again, our country,
2. What is the Roundffusor1?
The shown
diffuser, is a device, said apparatus comprising two lateral rigid supports,
mounted to the inside of the said diffuser which lateral supports are received ,
like a drawer , by two wooden rails, section T. The two wooden rails ‘’T’’ are
simply mounted on the wall surface with screws or nails. The main diffuser body
with his two drivers is mounted just like a drawer toward the two rails T. The
main body is composed from three basically 3D shapes. Each fourth diffusers,
displays together a new 3D shapes at their nearest point. The angle between each
3D shapes, including the new common shape, is the same. From the total surface
of each diffuser there is more than 90 % diffusing surface. Except the wooden
rails, section T, the complex diffuser body and his lateral supports are
fabricated from vacuum thermoformed hard PVC. The aim of this invention is to
radically improve any kind of an existing diffuser, be theoretical or a
practical one. Is a big effort, redirecting the whole diffuser philosophy: from
absorptive to clean diffusing, simultaneously being only a low frequency self
adaptive - variable low frequency resonator.
2.1
Something
about
diffusers…
The diffusers, until now, may be described theoretically and theirs predicted characteristics named diffusion ability are shown as polar plots, for various incidence angles, the additional absorptive curves being supplementary. The diffusion and absorptive ability of predicted acoustical diffusers are common and expected and for the real, practical diffusers. All those polar plots , usually taken each 5 degree of microphone position , shows the acoustic pressure of the reflected waves, and a semi circle shape is considered the ‘’ideal’’ meaning that the named diffuser controls the interfering reflections in such a way that scatters the incident sound uniformly so that the acoustic glare in all directions is minimized. As much from the wide bandwidth source is scattered uniformly, and the polar plots is taking the semicircular shape, the diffuser have the diffusing ability D tends or equals 1. Fig. 1 shows, for 3 incident angles, such a ‘’ ideal situation ‘’ where the diffusion polar plot’s ark, is like a semi circle. The plots from Fig. 1 are taken from bibliography as an example for prediction for a diffuser of 2,4 x 2,4 meters. The measurement setup for the described situation is similar with our Roundffusor’s measurement and surface (1,8x1,8 meters) and also with the diffuser named as ‘’ from another company’’ which also have dimensions bigger than our Roundffusor. Note, that for the 1,74 x1,74 meters, there are 9 Roundffusors1 mounted in a rectangular formation, a 3x3 matrix.
If the
diffuser shows one or more irregularities in his polar plots measurements, at
that particular angle, the diffusing ability is reduced and D tends to zero.
Must be noted that for those frequencies and angles of incidence, the acoustical
diffuser is working like as an absorptive surface, the notch being more or less
obtrusive toward the supposed source’s linear shape. When the diffusion ability
becomes high, a kind of source reinforcement may be perceived. This
reinforcement may be associated with the loudness. If it is sound coloration or
not, this is a difficult task to respond. Around practical commercial diffusers,
there is a confuse approach , because they
appears as having either high diffusing ability and negligible absorption
, or the characteristics are shown but the explanations done are misleading even
for the specialist. For many years, was kept as advantageous, the absorption
ability or more correct the disability of the named diffusers to not have a
small percent of absorption. There is a sensible difference as what is perceived
and understood as good or bad around the term absorption and his role in music –
live or recorded. This difference is wider than expected and differentiated
between US and European listeners. Even the highest ‘’ golden ears’’ gifted
person are in difficulty of being totally objectives in their appreciations,
simply because the educational background is not similar. From the all those
notch, the music, as well recorded as may be, will be distorted. There will be
certain area in x,y,z directions where the sound distribution / diffusion will
be irregular, far from the required symmetry.
We are
limited here, but we must note, that the interactive phenomena between purely
physical and perceptual aspects of what kind and where the absorption is
necessary and if the simple concept of diffusion and nothing else is better
practically but less well understood. To conclude, whatever the place for
diffusers to be installed, their dimensions, especially the depth, the smaller –
the better.
2. Study and effectiveness
In order to be sure that this diffuser doesn’t have any resemblance with anything else. Being restricted to less than 60x60 cm and less than 15 cm depth (in order to be competitive) was a very difficult task. Practically in study was all existed bibliography. The first advantage of this diffuser is that it works clean and effectively in the 250-6300 Hz, the large area where the voice and musical instruments spectrum are prominent and the human ear is more sensitive. As a result of very small distance between each Roundffusor [R1] – less than 1 mm, and the special mounting procedures, the whole surface of a rectangular team of diffusers [R1] is like a compartmented plate, simply ( but very steady) supported at the two edges, at some distance from the mounting surface. The plate works like a Helmholtz resonator. There is a known experience equation [T1] relating basically the diffuser’s dimensions, distance from the wall and distance between all diffusers. The Helmholtz type of resonance is double. First, the diaphragmatic absorption as if supposed to result from the grouped diffusers, where this total surface is related with the (supposed) bigger wave length for the low frequencies. ‘’ Supposed’’ because until now, especially in larger halls, the panels reflected bigger wave length than that corresponding to their dimensions. Those big panels redirect or reflect the low frequencies. If the panels are mounted at some distance from the mounting surface, they work as an air cavity resonator. Here we have the same surprise, and we encountered absorption down to 5 Hz. For 174x174 cm, the 9 [R1’s] area, this low frequency is much lower than the <critical> 300x300 cm ceiling music hall panels. It is known that those very low frequencies are sub harmonics usually generated in small rooms by the low of modal resonance corroborated with loudspeakers inherent distortion in the very low band. Fact is that a rectangular group of 9 [R1] deal with such a very low frequencies.
The parallel
architectural acoustics for music hall with the small and medium size rooms is
not brought here to confuse, but to show that are still many not clearly
explained problems in the two distinct fields of acoustics. Are they so distinct
or not, this is another story. The second aspect of the Helmholtz type of
resonance encountered in our grouped [R1] is that being mounted compartmented,
each line or row of minimum 3 [R1] is a ‘’ sub class’’ or part of the overall
diffuser’s surface. All the above analysis is maintained, only that, the
resonance frequency bandwidth is higher than the resonance of the whole grouped
[R1]. So, the grouped [R1] works simultaneously as an air resonator for
different bandwidths. I couldn’t prove ( at least at this stage of our research
) if the total surface of the grouped [R1’s] have some ‘’ classic’’
diaphragmatic absorption , because , being in such a way designed geometrically
,compartmented and steady supported, each [R1] doesn’t resonate as will be
expected from a plastic box, meaning a very affordable diffuser. The
measurements done with a miniature accelerometer, indicated similar ‘’ box’’
plastic resonance with the heavy wood furniture located in the room. For
monaural source / s , all [R1] acts alike, but as soon as the source become
stereo, each [R1] above his surface and below, at least at the finger’s edges,
feels continuously different. If applied, for 25 mm distance from the wall of 4
grouped [R1], the equation [T1] shows that the smaller resonance frequency will
be 58 Hz. This means and imply that the [R1] as hard plastic box don’t have any
significant flexures, even at high SPL’s.
The
equation Τ1
: fr=√ e/Dd and e= r/(r+w ) ,
where
r = distance between diffusers
w = diffuser
width (for Roundffusor, width = length)
d = depth of
diffusers
D = distance
from the mount surface
e = percent of open surface between
diffusers
fr = fundamental resonance frequency
All this indicate that the grouped [R1] works as a complex Helmholtz air cavity resonator offering absorption , enough linear , as shown on Fig. 5a, down to some 5 Hz. Such a performance, for only 12 cm depth, is for the first time achieved practically.
The other
Roundffusor1 [R1] function is his clean diffusing ability. As known, the polar
plots are measured for 0, 45 or 60 degrees. Here we have to compare 3 kinds of
polar plots : the ‘’ ideal’’ ones from Fig. 1, the Roundffusor1’s from Fig. 2 ,
3 and 4 and, the polar plots of a commercial diffuser produced by a known US
company. It is also known that except for incident angle at zero degree , where
many commercial diffuser present enough near semi circle polar plots , but still
unregulated, as we proceed toward 60 degree off axis polar plots, the results
are highly irregular for all commercial diffusers. More, the 3D lobes are
distinctly irregular. The sole exception until now, the diffuser ‘’ from another
company’’ where the overall lobe is somewhat better and may be ‘’ felt’’ at 3-4
meters. What is felt is the diffused field. In our situation, for the
Roundffusor1, if we compare the 3 groups of polar plots, the result are more
than obvious. May be expressed in words as such: the [R1] is a
wide band acoustical diffuser having the ability of low and very low frequency
absorption. The Roundffusor1 will be extremely useful in the control
rooms, HI-FI, HI-END or music hall and of a big help in music school- for voice
or instruments. At the actual technical / cultural level , as much as the
electronic device will replace the real organs, the ‘’golden ears’’ of gifted
people, able to hear 1dB or
1Hz J.N.D.’s and people who loves
and remember the timbre of real organs, will become less and less. Thus, the
difficult obtained hearing abilities will be slowly
lost.
2.1.1 Comparisons, perceptual
tests
To make
the story shortly, doing the listening tests in a room filled with only 9
Roundffusors, we obtained similar aural impressions either for near field, far
field, and all incidence angles. Until now, the ‘’ historical announcements’’
were outside
If we compare
the Fletcher-Munson curves with the waterfall plot of [R1] and we make their
algebraic additions, an almost straight line will result. In a way this means
that for our ears, the grouped [R1] is doing an inverse equalization. Because,
in any room where the [R1] were installed the aural impression was similar, this
means that the inverse equalization observed phenomenon is a self adaptive one;
I dare to say, like a digital equalizer. As in all ‘’ historical moments’’ we
are accepted to say that the all flat curves and polar plots measured from the
Roundffusor1 are very near to the said ‘’ near perfect diffused field’’
situation.
2.1.2 ‘’ The perfectly diffuse
sound field ‘’
Here is an extract from bibliography, showing
which are the conditions for a ‘’ perfect diffused’’ field*. Please note that we
not agree with the term perfect, so, near perfect seems more
approachable.
‘’ Even
though unattainable, it is instructive to consider the characteristics of a
diffuse sound field. Randall and Ward’ have given us a list of these: • The
frequency and spatial irregularities obtained from steady-state measurements
must be negligible.• Beats in the decay characteristic must be negligible.•
Decays must be perfectly exponential, i.e., they must be straight lines on a
logarithmic scale.• Reverberation time will be the same at all positions in the
room.• The character of the decay will be essentially the same for different
frequencies.• The character of the decay will be independent of the directional
characteristics of the measuring microphone.
These six
factors are observation oriented. A professional physicist specializing in
acoustics might stress fundamental and basic factors in his definition of a
diffuse sound field such as energy density, energy flow, superposition of an
infinite number of plane progressive waves, and so on. The six characteristics
suggested by Randall and Ward point us to practical ways of obtaining solid
evidence for judging the diffuseness of the sound field of a given room.’’ From: '' The Handbook of Acoustics-3rd
Editions- F. Alton Everest, TAB Books, page 223’’.
3. Listening tests results for
the Roundffusor1
The
good diffuser has more advantages than induced problems. We will present some of
them, as a result from personal listening test and as a kind of statistic based
on other’s perception. The degree of openness and spaciousness contained within
the recording, such as characteristics of transparency and depth, can also be
more easily assessed. This and the followings may be just literature, but when
listening test are at such an exceptional level as the measurements, the things
must receive the proper attention.
Regarding the precedence
effect, it is slightly reduced. Axial or left/right head movements don’t arrives
to dramatically induced changes. There is an incredible stability of sound
image. At the first sound heard from the loudspeakers, the listener,
instinctively [meaning very quickly, something between 0.1 to 1 second] focus
the ears/ eyes on the location of each musician. During the music, all those
spatial references, are easily memorized / visualized by the listener. More, if
the musician changes their places, or are moving here and there, their movements
are easily sensed. Because the Roundffusor1 is not working on '' phase grating
effect '', which involves and produce variable on surface absorption, the
subjective phase perception from our diffuser is near zero phase, regardless the
center frequencies or music bandwidth. All this phenomena are valid at more
than 6-7 meters from the music source and the '' critical focused point of
listening '' or ‘’ stereo seat’’ is obsolete. A kind of ''aura'' envelope
reconstruction from the room's recording session is obtained. Think that
all this arrives from a plastic area [of at least 9 Roundffusors] mounted on the
back wall or ceiling, there is nothing to be changed on your electronics. One
first conclusion: less absorption on high and mid frequencies: more music.
Second conclusion: the automatic bass absorption offered by the Roundffusor1,
removes the dynamic masking effects produced by the low and very low frequencies
toward the right of the audio band.
Sound
impressions in rooms were installed: unveil more and more details in the mix,
realistic sound stage and instruments location, outstanding voice &
instrument quality & musicality, less fatigue at high levels of listening.
Fig.
1 Roundffusor1 – patent pending, 4 Roundffusor1 mounted together
Fig.2 The
''ideal'' predicted polar diagrams (from bibliography) for situations
similar with the Roundffusor's measurement setup. The heavy lines represents 99%
diffusion factor, and the incidence angles are: a) 0, b) 30 and c) 60 degree.
Fig. 3 Off
axis polar diagrams for Roundffusor1. All polar plots were obtained from a 45
deg. incidence angle.
Fig. 4 Off
axis tile polar diagrams for Roundffusor1
Fig. 5a-
Waterfall plots for Roundffusor1, same source and setup, sweeping bandwidth
12-19368 Hz
Fig. 6 Off
axis polar diagrams from another company’s best
diffuser
Fig.
7 A big
control room in
Thanks
Such an
invention like this, appear each 40 or 50 years. Without God’s help or an exceptional
sudden inspiration, it is no possible to appear an evident result like this. In
it’s simplicity but highly complex functioning, even the patent’s text doesn’t
explains all, and further research is needed, at least from other researchers if
still in disbelieve.
Bibliography
1.
D‘ANTONIO , P., The Directional Scattering Coefficient: Experimental
Determination, J. Audio
2. D’ANTONIO,
P., KONNERT, J.H. and KOVITZ, P. The Disc Project. Experimental Measurement of
the Directional Scattering Properties of Architectural Acoustic Surfaces,
IpAAd2, 14 1144 (June 1994).
3. COX, T.J.
and LAM, Y.W. Evaluation of Methods by Predicting the Scattering from Simple
Rigid Panels. Applied Acoustics.40 123-1 40 (1993).
4. COX, T.J.
and LAM, Y.W., Prediction and Evaluation of the Scattering from Quadratic Residue Diffusers , J.
Acoustic. Soc. Am. 95(l). 297-305. (1994).
5. ISO
66:1997, Acoustics — Preferred frequencies.
6. IEC
61260 (1 995-08) Electroacoustics — Octave—band and fractional—octave band
filters.
7. DRAFT
AES-4id-xxxx , AES information document for room acoustics and sound
reinforcement systems-Characterization and measurement of surface scattering
uniformity ,
AES 2000.