When traveling to different time zones, we need to correct our time-keeping tools in order to align our activities to the local standard time. Very simple.
Likewise in sound reinforcement systems, we often encounter multiple sources of misalignment, albeit in much smaller time increments. The importance of these misalignments is often subject to debate, but with the advent of a wide variety of very good digital processing products, correcting such anomalies is no longer out of reach for most of us.
Further, the presence of affordable, accurate time-based measurement systems (Rational Acoustics Smaart, Gold-Line TEF, etc.) means that it is by no means impractical to clearly determine the presence of multiple arrivals from a sound system.
Almost all sources of misalignment are based on the physical positioning of various transducers in relation to one another, to their surrounding physical boundaries, and to sound sources. The only place where there are no appreciable time-based alignment problems (other than phase response) is in a sound system’s electrical signal path.
The electrons that constitute electrical signals utilized in audio systems travel at close to the speed of light, and even when there are great variations in cable length or multiple stages of audio processing, the degree of leading or lagging across the audio spectrum is not perceptible to human hearing.
Digital signal processing (DSP) does present potential for misalignment, but it is rare that DSP is employed for selected single channels of audio.
Thus we focus on the transducers used for inputs and outputs when discussing misalignment.
Although there is a potential for microphones (or mic technique) to be just as much a culprit as loudspeakers, traditionally we have not had the means to treat separate inputs nearly as easily as we do the outputs feeding into the loudspeakers.
The proliferation of digital mixing consoles for sound reinforcement certainly helps change this, yet still the primary focus of misalignment too often remains on the “tail end” of our sound reinforcement systems.
Driver Misalignment
In loudspeakers with both high- and low-frequency transducers that are direct radiating (not very common in sound reinforcement), there will be offset in the arrival times of the high-frequency transducer versus the low-frequency transducer.
This is simply due to their required physical positioning. A high-frequency transducer consists of a smaller diaphragm with a shallower motor structure than its low-frequency counterpart.
When these drivers are positioned on a flat baffle (the most common baffle shape) the high-frequency signals arrive at the ears of listeners sooner than the low frequency signals, on the order of several hundred microseconds (See Illustration 1).
This misalignment is reversed when the high-frequency driver is horn loaded (the more common method used in sound reinforcement loudspeaker systems), requiring a waveguide of a certain depth to achieve the needed pattern control.
This is one of two primary reasons for using such horns; the other reason is an increase in acoustic efficiency.
With horns, high frequencies no longer lead low frequencies, but rather they lag behind, again on the order of several hundred microseconds (See Illustration 2).
In loudspeaker systems where high-, mid- and/or low-frequency components are in various configurations of direct radiating and/or horn loading, it is almost impossible to realize horn lengths that position the drivers in physical alignment to one another.
Note that the alignment of multiple drivers is not as simple as visualizing or physically measuring the location of each component’s diaphragm and then either positioning them to be in the same physical plane or applying electrical signal delay to the leading component. (See Illustration 3A and 3B)
Every driver has varying points of acoustic origin that shift with frequency. The point of acoustic origin, as applied to driver alignment, must be measured to be derive a mean value.
