1260 Chapter 34
system should be based on how well that loudspeaker
system will accurately reinforce a live performance.
For this reason, listening tests done with live sources
in acoustically well-designed rooms are ideal evalua-
tions. When it’s possible to do live evaluations, use a
strong-voice talker with a well-chosen microphone for
speech. Use a single, well-known instrument, such as an
acoustic guitar or acoustic piano for musical evalua-
tions. A singing voice, accompanied by a guitar or
piano is also a good choice. See Chapter 16 for a discus-
sion of microphones.
When a live test is not possible, use a CD or other
high-quality digital recording as a source. Certain
well-recorded vinyl LPs may also be suitable. Choose a
recording of a solo acoustic guitar, acoustic piano, or a
voice accompanied by a guitar or piano.
Choosing these simple musical sources makes it easier
to evaluate the fidelity of the loudspeaker system because
most people are familiar with the way they ought to
sound. If the loudspeaker system colors this in any way,
most people will recognize the coloration easily.
Using recordings of loud and dynamic rock music or
highly synthesized music of any kind may be a good
way to evaluate the ability of the loudspeaker system to
handle high-power live sources, but it is not a good way
to evaluate the fidelity of a system since distortions or
frequency response aberrations in the loudspeaker
system may be interpreted as intentional parts of the
original performance!
34.3.2 Loudspeaker Systems
There are several styles of loudspeaker systems used in
sound reinforcement systems. The most common are the
central cluster, split cluster, exploded cluster, and the
distributed system. Any of these loudspeaker systems
may be designed from component loudspeakers, pack-
aged loudspeakers, or line arrays. In addition, there are
variations and combinations of these types. This sec-
tion discusses some basic design criteria for loud-
speaker systems. Chapters 17 and 18 discuss additional
details of loudspeaker system design.
34.3.2.1 The Central Cluster
A central cluster is a group of packaged loudspeaker
systems or component horns and woofers placed in a
central location and aimed at a listening area. The tradi-
tional central cluster is placed above a stage (on the pro-
scenium) or above the primary microphone location,
Fig. 34-16. A modular line array, suspended in this loca-
tion, may be considered a central cluster.
A location above the audience makes the difference
between the distance from the cluster to the nearest
listener and from the cluster to the farthest listener more
nearly equal. This, in turn, makes the job of designing
the cluster for even audience coverage easier. In most
cases, however, the cluster should not be more than
about 30–45 ft above the heads of the listeners. This is
because listeners seated near the talker can often hear
both the unaided talker and the cluster. If the cluster is
more than about 30–45 ft above the heads of the
listeners, they will notice a hollow sound or even a
distinct echo due to the natural delay between the sound
from the talker and the sound from the cluster.
The human ear can accurately discriminate the loca-
tion of sounds from a left-right perspective, but not as
well from an up-down perspective. Thus, another
advantage of a central cluster, if it is placed near the
center of the room or approximately above the primary
microphone location (and assuming other factors are
favorable), is that the sound will appear to emanate
from the talker, and not from the cluster.
A final, significant advantage of a central cluster is
that, compared to an equally well-designed distributed
loudspeaker system, the central cluster is almost always
less costly.
Sometimes, aesthetic considerations prevent the
installation of a central cluster. For example, the loud-
speakers may block important architectural elements in
a religious facility or historical building.
In some rooms, the ceiling is too low compared to
the length of the room to allow a central cluster to work
well. This problem prevents adequate gain before feed-
back in the back of the room (PAG is too low). A good
rule of thumb is that D 2 should be no more than about
four times D 1 for a single central cluster to work prop-
erly. If the 45 ft height rule is followed, this seems to
limit central clusters to rooms with dimensions of 180 ft
or less. However, the 45 ft rule can be ignored if the
listeners cannot hear the talker without the aid of theFigure 34-16. Typical central cluster. Courtesy Community
Professional.