1354 Chapter 35
It stands to reason that point sources or line loud-
speakers of a certain dimension cannot exhibit this radi-
ation behavior automatically. Here the manufacturer has
to supply along with the arrays, not only the electronic
driver unit for the same, but also the parameter setup
algorithm. By means of attached software this algorithm
will then be controlled according to the desired
application.
35.1.4 Wall Materials
To simulate the radiation behavior of sources in rooms
or open spaces we need to construct corresponding
models. All boundary walls of these models need to
have the corresponding acoustic properties:
- Absorption.
- Scattering.
- Diffraction.
These properties have been discussed in Section
7.3.4, therefore they will not be repeated here. Instead,
some specialties important to know when doing
computer modeling should be added here.
35.1.4.1 Absorber Data
Absorbing behavior has been known for hundreds of
years, but data has been available only for 80 years. We
distinguish between data measured in a reverberation
chamber (Standard ISO 354)^23 or data that is angle
dependent. The latter absorption coefficients are very
seldom measured and only available for special applica-
tions. For computer simulation, the absorption coeffi-
cient measured in the diffuse field will be used. This
coefficient is measured by the corresponding
manufacturers and published in specification brochures.
It is measured in octave or -octave bands and starts
normally at 63 Hz and goes up to 12 and even 16 kHz.
In most simulation programs the low end is skipped
because the actual simulation routines do not cover fre-
quency ranges below 100 Hz. The highest-frequency
band is quite often only 8 kHz.
All this data is meanwhile published in table form
and some simulation programs have more than 2000
materials from different manufacturers on board.
35.1.4.2 Scattering DataScattering data is not found in textbooks except for
some special scattering materials or samples. Here
should be mentioned the products of RPG Diffuser Sys-
tems Inc., which produce special modules with sound-
scattering surfaces.
On the other hand it is known that the absolute value
of the scattering coefficient s is less important. The fact
is that there is almost no material not scattering (s=0)
or only scattering (s= 1). The practical values for the
scattering coefficients are between 0 and 1. So there are
some rules of thumb to define the actual scattering coef-
ficient in simulation software programs. Some programs
give some guidance to estimate the coefficients, other
programs like EASE 4.2 use special BEM routines
(compare to Fig. 7-46) to derive the coefficient in a way
as it should be measured according the proposals of
Mommertz,^24 see also Standard ISO 17497-1.^25
A scattering coefficient will never generally be avail-
able in tables (except the mentioned special module
values), because the way the interior architect uses the
materials in a hall affects the scattering behavior.
Therefore the scattering behavior of wall parts in a
computer model must be determined model specific.35.1.4.3 Diffraction, Low-Frequency AbsorptionAs we will see in Section 35.3.2, the computer simula-
tion programs use different ray-tracing algorithms to
calculate the impulse responses in model rooms. But
these routines of using particle radiation are only valid
above a certain frequency determined by(35-24)where,
K is a constant (2000 in metric units, 11,885 in U.S.
units,
RT 60 is the apparent reverberation time in seconds,
V is the volume of the room in cubic meters or cubic
feet.For lower frequencies and especially in small rooms the
particle assumption cannot be applied. Here the wave
acoustics routines are applied. An analytical solution is
impossible, so numeric routines have been developed.
Mainly the finite element method (FEM) and the bound-
ary element method (BEM) are used. First for applying
the FEM, the computer model must be subdivided in
small volumes (meshes), where the dimensions of the(^1) » 3
f 1 K
RT 60
V
------------=