We hear the term crossover all the time in pro audio, and crossovers play an extremely important role in our P.A. systems. A crossover (sometimes called a “crossover network” or “frequency dividing network”) is an audio circuit that divides the full frequency range into high- and low-frequency bands, and sometimes also mid-frequency bands.
I like to think of a crossover as a traffic director, routing low frequencies to woofers, mid frequencies to midrange drivers, and high frequencies to tweeters. It ensures that each driver in a loudspeaker receives only the frequencies it was intended to reproduce.
Crossovers are particularly important in live sound applications because we’re dealing with high levels of amplification that could potentially damage drivers. You’re probably aware of the fact that it’s not a great idea to route an audio signal of, say, 50 Hz to a small transducer (i.e., a tweeter), but if you think about headphones or IEMs, that’s often exactly what happens. In such applications, the transducers aren’t damaged because the amount of power required to drive them is miniscule compared to what we need to make a P.A. system fill a stadium. We need to filter low frequencies from the feed to HF drivers, and that’s one of the things that a crossover does. Conversely, it’d probably be a good idea to filter high frequencies before they reach a woofer because (1) most woofers aren’t very good at high frequency reproduction anyway (and those that are probably can’t produce the SPLs we require), and (2) when multiple drivers produce the same frequency range, we’re inviting phase issues. In general we can reduce distortion by removing the frequencies that a driver is not optimized to reproduce.
Passive Crossovers
There are two broad categories of crossover: passive and active. A passive crossover is usually located inside the cabinet of a full-range speaker and you may never have to worry about it. One or more filters are created using resistors, capacitors, and inductors (coils), the values of which are determined through careful analysis of frequency response, impedance and sensitivity of each driver. A passive crossover can be as simple as a single capacitor wired in series with a tweeter (that’s probably what’s in your car), but make no mistake — crossover design is a high art/science and is crucial to the sound of a speaker. Passive crossovers don’t require any electricity other than the power delivered to them by an amplifier, and they live in the signal path after the power amplifiers (See Fig. 1).
The drag about a passive crossover is inefficiency. Let’s suppose you have a three-way loudspeaker rated to handle 100 watts. You go out and buy a power amp rated to deliver 100 watts. (If you’re smart, you buy an amp that can deliver more than 100 watts, but that’s a subject for another time…). Inside the speaker cabinet there is a woofer, a midrange and a HF driver. These individual transducers will never see 100 watts, because the full-range audio coming from that external power amplifier is going to be separated into three bands, and probably attenuated on its way to at least one of the drivers (more on that in a second). The crossover components need to be able to handle the full 100 watts (and then some) from the power amp, and unfortunately, they are going to waste a lot of that power as heat while they are directing
traffic.
Insertion loss is the loss of signal that happens when a crossover is placed between an amp and a speaker. A crossover with an insertion loss of only 1 dB, when connected to the output of a 100-watt amplifier, reduces the effective maximum power that the driver sees to about 79 watts.
A crossover also matches the sensitivity of individual drivers. For example let’s say a woofer has a sensitivity of 90 dB at 1 meter/1 watt, and a HF driver has a sensitivity of 94 dB at 1 meter/1 watt. Right off the bat, the tweeter is too loud for the
woofer, making the speaker sound bright. The high frequencies destined for the tweeter must be attenuated in order to make them match the output of the woofer. The extra power diverted from the tweeter is usually dissipated by the crossover as heat. It’s kind of a waste.
There are other disadvantages to passive crossovers, such as lack of adjustability and low-frequency distortion from inductors in the low-pass band. It’s also possible that low-frequency distortion generated by the amplifier can produce high-level harmonics that can damage a high-frequency driver.
Active Crossovers
Active crossovers live in the signal path before the amplification and typically operate at line level (See Fig. 2). The placement in the signal flow means that the crossover doesn’t need to handle high levels of
power — increasing reliability and opening up a broad world of components to the designer that might not be available with high power ratings.
As active crossovers divide the frequency range before the power amp, separate amplifier channels are required for each driver or set of drivers, opening up the idea of biamplification (separate amps for LF and HF drivers) and triamplification (separate amps for LF, midrange, and HF transducers). No one says that you have to use the same amps for the high- and low-frequency amplification — you can now use different amps for high-, mid- and low frequencies, and put the power where you need it. (e.g. more power in the bottom). In fact, that’s what many manufacturers of powered speakers already do. Instead of using the components in a passive crossover to compensate for different driver sensitivities, the crossover outputs or power amp input level controls can be adjusted, making the entire system more efficient. Removing passive components between an amplifier and speaker also improves the amplifier’s damping factor (its ability to control the motion of the driver).
Other advantages of active crossovers include increased headroom and reduced intermodulation distortion.
Headroom: When both high- and low-frequency material are present in a signal (say.. flute and bass guitar) the high-energy bass frequencies use up most of the power, leaving little power for the high frequencies. The result is distortion of high frequency material. With an active crossover, the HF’s are routed to their own power amp, eliminating the problem.
Intermodulation distortion (IMD): Intermodulation distortion occurs when at least two frequencies interact to form new, non-harmonically related frequencies. IMD is usually more audible to most listeners than harmonic distortion. It can be caused by the Doppler effect. A large diaphragm moves back and forth at a low frequency. The high frequency ‘rides’ on the moving diaphragm, causing its frequency to shift up as the diaphragm moves closer to the listener, and shift down as the diaphragm moves away — the same way the pitch of a truck (or train) horn rises as it gets closer, and falls as it gets farther from you. Bi- and tri-amplified systems reduce this type of distortion.
In the past, setting up a bi- or tri-amped P.A. system was a chore because you had to do a lot of the calculating manually. These days, manufacturers do a lot of the work for you. Active speakers already have the correct amp for each driver, and the crossover is optimized for those drivers. Your only concern might be determining the crossover between a subwoofer and a full-range system, which is often made even easier by using “drive processors” that have presets for boxes from specific manufacturers. Next month, we’ll look at crossover parameters and what they mean.
Steve “Woody” La Cerra is the tour manager and front of house engineer for Blue Öyster Cult.
