Back in the ’60s, when I first began playing in bands, speaker crossover systems were almost unknown outside of higher-level professional applications. The first bands I played in used nothing more than a pair of Atlas Sound Reflex Horns connected directly to a little powered P.A. mixer. It was fortunate that the mixer had only enough channels to handle the vocals, because instruments pumped through the Atlas Horns would have sounded really bad.
We took a major step up when our guitarist bought a new Kustom PA 200 with columns that had cone-driver speakers. The Kustom looked very cool with its tuck-and-roll red-sparkle covering; still, something was missing from the sound. Our little P.A. system didn’t sound anywhere near as good as my home hi-fi system, but we just accepted the apparent fact that live P.A. systems sounded muffled while home stereos were capable of full-range sonic reproduction.
A few years later, I began experimenting with improving the sound of my P.A. system. I understood enough to know that large drivers are better at handling low frequencies than small drivers are and vice versa, so I built a full-range speaker cabinet with an 18-inch woofer, a 10-inch midrange driver, and a horn for the highest frequencies. Of course, without some way to direct the appropriate frequencies to each driver, it wouldn’t work. While attempting to solve the problem, I was introduced to the wonderful world of speaker crossovers.
These days there are many great-sounding speaker systems ranging in size from huge stacks to ultracompact units you can easily carry with one hand. Virtually all of these systems use multiple drivers for different frequency ranges, which means they require some sort of
So whether you have your own P.A. or just use the house systems at the clubs where you perform, a working knowledge of how crossovers function is helpful for understanding what’s going on soundwise at your gigs. The logical place to begin this discussion is the place from which sound emanates — the driver (in this article I’ll use the word driver to refer to the individual components, such as a woofer or horn, in a speaker system).
The very first audio drivers were full-range devices; that is, they reproduced the full bandwidth of whatever you pumped into them. However, it was quickly discovered that you could design a driver with a big diaphragm that would reproduce the bass notes well or with a small diaphragm that would reproduce the high notes well, but it was difficult to make a single driver that could do both well. To better understand the problem, let’s first take a close look at driver construction.
Generally speaking, drivers with larger diaphragms handle low frequencies more efficiently, and drivers with smaller diaphragms handle high frequencies more efficiently. Why is that the case? Basically, the lower the frequency, the longer its wavelength, and the more air that must be moved to transfer energy from the driver to the air.
Very high notes have wavelengths only a few inches long, while bass notes can have wavelengths of 30 feet or longer. If all a driver has to do is reproduce a single frequency, then you can tune it to efficiently reproduce just that frequency with no regard to any other. A good visual reference for this concept is a large pipe organ. The length of most pipes in such an organ is half the wavelength of the fundamental frequency they produce, so the longest pipe in a large organ is 28 feet long (half the wavelength of 20 Hz), while the shortest one is an inch or smaller.
Drivers designed to handle low frequencies have diaphragms (in this case, usually cones) that are larger in diameter, thicker, and heavier than those of high-frequency drivers. Those drivers, commonly called woofers for obvious reasons, have to move a lot of air — they work like a big piston pushing in and out with as much as an inch of travel, up to a few hundred times a second.
For a driver to reproduce, say, the metallic sound of a drumstick hitting a cymbal, it must move back and forth approximately 10,000 times per second. A big, heavy driver diaphragm is unable to do that, but a small, lightweight diaphragm can, which is why drivers designed to handle high frequencies are usually only a few inches across and weigh less than an ounce. A high-frequency diaphragm has to move only a tiny fraction of an inch to transform the electrical energy into lots of sound. High-frequency drivers — tweeters in hi-fi systems — are called horns in P.A. systems because they typically have a bell-like flare on the front.
If you run high frequencies into a large woofer, the diaphragm will attempt to move back and forth thousands of times per second, wasting lots of energy that is converted into heat rather than sound. Conversely, putting bass notes into a tweeter will force it to move in and out farther than it’s designed to, turning it into so much confetti in milliseconds. If you want to power both types of drivers with a single amplifier channel, you need some sort of routing system that will send the bass frequencies to the woofer and the high frequencies to the horn. Necessity, in this case, gave birth to crossovers.
Crossovers are divided into two basic categories: passive and active. Both types divide the entire frequency spectrum into two or more bands, and the boundaries between the bands are the cutoff frequencies or crossover points (see Fig. 1). Ideally, the split would be absolute, with all the high frequencies going to the horn and all the low frequencies going to the woofer, but in practice this is not easily accomplished, particularly with passive designs.
The actual rate at which unwanted frequencies fall off beyond the crossover point is referred to as the slope because of the way it appears on a graph. The slope varies according to the number and type of components used, but the most common crossover slopes are 6, 12, 18, and 24 dB per octave.
As the name implies, passive crossovers are circuits that have no active (powered) electronic components. They are usually built into the speaker cabinet and fed driver-level signals directly from the amplifier (see Fig. 2).
The steeper the slope, the better able the crossover is to keep unwanted frequencies away from a particular driver. However, steeper slopes require more components, a setup that wastes a lot of electrical energy, even at the frequencies the crossover is supposed to be passing. In addition, you need a separate crossover for every speaker cabinet in your system. To top it off, it’s not possible to change the crossover frequency without soldering in new components.
The disadvantages of passive crossovers soon became apparent to those using them in professional sound-reinforcement applications, and somebody came up with the idea of using active circuits instead. Rather than being placed between the amplifier and speaker, active crossovers go between the mixer and the amplifier (see Fig. 3). That arrangement requires separate amplifier channels for each driver; for example, to drive a two-way speaker system, you need two amps, a scheme called biamping. Carrying around extra power amps may seem like a hassle, but active crossovers and biamping have lots of advantages over passive crossovers.
The amount of power an active crossover is required to handle is minuscule compared with the hundreds of watts pumped into a passive crossover; as a result, it can have an adjustable design. Instead of working at a fixed frequency of, for instance, 500 Hz, it can include a control that sweeps from 100 to 5,000 Hz, allowing it to be tuned for the frequency preferences of a given driver.
In addition, active crossovers don’t waste power the way their passive counterparts do; attenuation slopes of 18 or 24 dB per octave are therefore common. A greater slope removes the offending frequencies more completely from each driver, resulting in a cleaner signal with less damage. Another benefit to active crossovers is that each driver gets exactly the frequencies it can produce most efficiently, so it can be driven harder without sounding bad or having to dissipate a lot of heat — the nemesis of all drivers.
In addition to offering greater efficiency and controllability, active crossover systems are generally less expensive than passive systems, because a single active crossover can easily drive dozens of amplifiers without any problems. For larger sound-reinforcement systems, it’s simply a matter of adding as many amplifiers and speakers as required. Also, more sophisticated crossovers have limiters, delays, and other processors built into them, providing additional control over the sound (see the sidebar “Delay Tactics”).
Most active crossovers offer a wide range of settings to accommodate the needs of different speaker systems. How do you know which frequencies a particular driver can accept without damage? Begin by examining the literature that came with the speakers; nearly all manufacturers provide recommended crossover frequencies and slopes for their products. If you don’t have documentation, try contacting the manufacturer directly.
An additional consideration is the way in which crossovers connect to speaker systems. Some speakers have individual inputs for each driver (usually on ¼-inch or XLR connectors or on five-way binding posts that accept dual and single banana plugs, spade lugs, or bare wires). However, Neutrik NL4 Speakon connectors are becoming increasingly common. NL4s are able to switch between passive and active routing; in the active setting, the low frequencies go to the Low terminals and the high frequencies go to the High terminals.
Three, Four, or More
A bi-amped system splits the audio spectrum somewhere in the middle, usually around 1,600 Hz, for two-way speakers. However, you can add as many crossover points as necessary for really large speaker systems. A system that uses three-way active crossovers is triamped; a system that has four-way active crossovers is quad-amped. Within the heady world of professional sound systems, tri-amping is extremely common because it can split the speakers into three spectral ranges: 20 to 120 Hz, 120 to 1,600 Hz, and 1,600 to 20,000 Hz. Of course, these are all approximate crossover points; each speaker manufacturer has specific frequency values for each particular speaker type and enclosure.
One of the larger systems I regularly use is a three-way active system with a four-way passive split. The woofers, mid-range drivers, and horns are each powered by their own amplifier set, but the high range is passively split at approximately 8 kHz or so between a dome tweeter and a horn driver. If necessary, I can add in some big 18-inch subwoofer cabinets, using the lowpass 100 Hz filters common on many professional amplifiers (from companies such as QSC, Crown, and Mackie). These filters transform an amplifier into a subamp that outputs only the frequencies below 100 Hz. In this case, the system becomes a quad-amped, five-way affair with a passive split.
However, more is not necessarily better, because each crossover point tends to introduce phase errors and makes the system more prone to feedback. That’s the reason that sound engineers are invariably interested in the crossover points of an unfamiliar sound system. Any big feedback frequencies will probably appear somewhere around the crossover points.
Over and Out
I hope this simple introduction to the complex and wonderful world of crossover technology has improved your basic understanding of sound-reinforcement concepts. Of course, variations on the technology show up in everything from the simplest two-way home-stereo speakers to multicomponent studio-monitoring systems costing tens of thousands of dollars.
You don’t have to understand how a particular technology works to use it, but in this case, a little knowledge can go a long way toward maximizing your tweeters and keeping your woofers in line. Rock on!
Crossover units sometimes include certain features that enable them to become an entire speaker-management system. The more sophisticated products don’t simply regulate the crossover frequencies and slope; they also provide limiters, parametric equalizers, and even digital delays, or they allow for the insertion of those devices into the signal path. The need for limiters and EQ should be obvious; however, why it might be desirable to delay one or more drivers may not be as readily apparent.
Basically, in order for a large array of speakers to work together cohesively, some of the drivers must be delayed to allow the sound from a far driver to catch up with the sound of the drivers closest to the audience. Sound travels at approximately 1,100 feet per second, or 1 foot per millisecond, so if sets of speakers are located hundreds or thousands of feet apart, the time delay can be substantial. There is even a slight delay when horn drivers are placed only a foot or so behind the woofer diaphragms, as is often the case. Delaying the woofer by the appropriate amount of time — 1 millisecond for a distance of 1 foot — aligns the entire speaker cabinet so that the sound emerges from it in a single pulse rather than stretching out.
Excellent examples of this variety of high-end crossover include the BSS Omnidrive (an industry standard) and the dbx DriveRack. Each of those units include all the goodies you could want in addition to storing all the crossover information for dozens of speaker types by name. All you have to do is select the speaker type and press the Recall button, and within a heartbeat you have all of that speaker’s crossover points, delays, slopes, equalization, and limiters as specified in the manufacturer’s recommendations. While these high-end units have all the bells and whistles, they can be very expensive costing a few thousand dollars or more. But Galaxy Audio is now offering their DSPOT processor series, with very affordable crossovers, compressors, and equalizers. And the Galaxy DSPOT processors are very cost effective.
Copyright Mike Sokol
Sound Advice Links
A short primer on crossovers by Siegfried Linkwitz on the Linkwitz Lab site. www.linkwitzlab.com/crossovers.htm
Another short crossover primer by Wayne Larmon on the scrounge.org site.
Copyright 2008/2016 Mike Sokol – All Rights Reserved