While sound system operators have come to enjoy the considerable benefits of digital signal processing, it’s important to understand how these devices truly interact with loudspeakers.
There’s a lot of cause and effect that sometimes gets overlooked, resulting in less than optimum performance, and leaving the operator scratching his head as to what’s going wrong.
One particularly important aspect that is often overlooked is using a DSP to drive loudspeakers with a manufacturer’s prescribed settings (or other target drive function).
Specifically, it is absolutely vital that the transfer function (magnitude and phase response versus frequency) of each output of the DSP match the manufacturer’s prescribed (or other target) transfer function.
Simply matching the values of the parameters on the front panel is not sufficient. The reason is that DSP units from different manufacturers, or even different units from the same manufacturer, may implement filter algorithms that are not exactly the same.
Thus, front panel read-outs on two different DSP units of a 24-dB-per-octave Butterworth high-pass filter at 1.6 kHz may not have the exact same curve when measured. These filter (transfer function) differences may cause performance differences when driving loudspeakers. Therefore, it’s important to measure the transfer function of the DSP unit that will be used to verify that it is indeed set to yield the desired response.
Start by getting the target transfer function for driving the loudspeakers from the manufacturer. Alternatively, if you know the exact model of DSP the manufacturer used to derive their settings for the loudspeakers, the parameters can be programmed into that DSP and measured to get the target transfer function.
These parameters may then be compared to the measurements of the DSP that will be used.
Even if this is not done, it’s still advantageous to have a listing of the parameters, because it will help speed up the process of entering the initial values to set-up the DSP unit to be used.