Computer Aided Sound System Design 1343
35.1.3 Acoustic Sources and Loudspeaker Systems
35.1.3.1 Natural Sources
Sound reinforcement quite often has to deal with rein-
forcing natural sources like human voices or natural
music instruments. Therefore to know the amount of
reinforcing required, we have to know the quality of the
reference sources and the following aspects:
- Loudness.
- Frequency response.
- Directivity.
Fig. 35-6 shows the level and frequency range of
natural sources and instruments perceived by human
beings. In this range natural sources develop their sound
power and produce level components in the mentioned
frequency range. Everything outside this range is on one
side masked by noise (lower 30 dB in the midrange) or
dangerous for our health—pain threshold |120 dB.
Frequency components lower than 25 Hz are becoming
inaudible as well as frequency parts higher than
15–20 kHz—depending on age and health.
Natural sound sources like human voices or musical
instruments do not radiate sound in an omnidirectional
way. It comes close in case of a human voice, but the
higher the frequencies, the more the head becomes an
obstacle to the sound radiation backward. Fig. 35-7A
shows the directivity balloon curves of a female voice
in the vertical domain. Below 1000 Hz the pattern is
almost an omnidirectional radiation pattern, but for
higher frequencies the radiation dominates more and
more in front of the head. This is also the reason that in
such concert halls where the audience is behind the
orchestra, people sometimes complain about the
singer’s clarity.
The radiation behavior of musical instruments is
much more complex. Here a lot of investigations have
been done, especially by Meyer.^6 Fig. 35-7B shows the
3D presentation of the directivity balloon of a horn
instrument including the player. According to Meyer the
shadow effect of the player himself is also considered.
We observe with increasing frequencies a reduced radia-
tion into the front domain of the player.
To model all these different natural sources correctly,
their radiation behavior has to be known and this not
only as single instruments but also in groups. Here a
lack of corresponding data is still evident.
Figure 35-5. Same room model in four different views.
A. Wireframe model. B. Rendered view.
C. AutoCad presentation. D. SketchUp view.