System Gain Structure 1227signal rich in harmonic distortion, Fig. 33-8. Gain struc-
ture, from a component perspective, is passing the
signal through at optimum amplitude—not too strong
and not too weak. As such, a system component can be
overdriven, underdriven, or optimally driven by a signal
source. It is important to note that the SNR of the
program source is often the determining factor for the
SNR of the entire system, since it can only be improved
with very specialized signal processing that is not found
in typical reinforcement systems. The old adage
“garbage in, garbage out” certainly applies. The SNR
will be degraded as it passes through other system
components, which is why care must be taken to prop-
erly calibrate each stage of the system.
33.4 System Gain Structure
Audio components have been evolving since early in
the last century. In the process the dynamic ranges of
system components have become similar, and in many
cases approach the theoretical limits dictated by nature.
While overall dynamic range may be similar, the clip-
ping levels and noise floors are not identical between
manufacturers or even within product lines. While we
won’t go into all of the reasons for this, it is unfortunate
that at least clipping levels aren’t standardized within
the sound reinforcement industry. As such, it is possible
for a system component to be operating optimally
within its own dynamic range, and yet be overdriving or
underdriving the next component. This reality forces us
to consider gain structure from system perspective.
Before discussing the gain structure of a sound
system, it is necessary to consider a method for deter-
mining the internal gain structure of a system compo-
nent. This can be done by introducing a stimulus to the
component and observing its output signal. It is
common practice for technicians to use a stable and
repeatable waveform for calibrating the signal-
processing chain. A sinusoidal waveform, commonly
called a sine wave, is such a waveform. The sine wave
is a single frequency tone that is easily generated, fed to
an input, and observed at the output of each component
in the chain. The previous graphs have shown sine
waves displayed on a magnitude versus frequency plot.
They resemble a vertical spike due to the narrow
frequency bandwidth. An alternate and equally valid
display is amplitude versus time. The oscilloscope
displays the amplitude of the waveform as a function of
time. More advanced models will even provide some
statistics, such as peak voltage, rms voltage, frequency,
crest factor, level, etc. Let us proceed. A 1000 Hz sine
wave is developed across the input terminals of one
channel of the mixer. An amplitude is chosen that does
not overdrive the input, and that is of sufficient level to
drive the mixer to its full undistorted output voltage
with all level controls in the signal path set at unity. For
microphone inputs, about 0.1 Vrms (–20 dB ref. 1 V) is
generally sufficient. As much as 1 Vrms might be
required for a line level input. The level controls of the
mixer are set as follows:- Master at unity or 0 dB.
- Channel at unity or 0 dB.
- Trim at unity or 0 dB.
Under these conditions, the voltage amplitude of the
output signal should be the same as the voltage ampli-
tude of the input signal—an amplification factor of one
or unity, and a gain of 0 dB.
The input voltage has been increased until the main
meter of the mixer reads zero. We will speak in more
detail about zero later, but for now we will assume that
it indicates a voltage in the optimum operating range of
the mixer’s overall dynamic range (typically 20 dB rel.
clipping). Since program audio waveforms are
constantly changing, this operating level allows some
room for peaks in the audio waveform to pass undis-
torted. It is instructive at this point to measure the
output voltage of the mixer at meter zero. Using the
oscilloscope, either the peak or rms value of the wave-
form can be measured. It is traditional to measure the
rms value, since it is readily measured with much less
sophisticated voltmeters than the oscilloscope and
correlates well with the loudness and heat production of
the signal.
For historical reasons, a common voltage measured
at a mixer output with the meter indicating zero is
1.23 Vrms, corresponding to an rms open circuit level
of +4 dB ref. 0.775 V (+4 dBu). This level might be
termed the operating level of the mixer. A volume indi-Figure 33-8. Overdriven signal characteristics.