Crossovers

Print
Sunday, 16 January 2005 20:14

The audible frequency spectrum runs from 20Hz to 20kHz theoretically (in practice I can't hear much beyond 17kHz and that's pretty normal for most people and there's very little up that high anyway). Sound systems having to reproduce this wide range of frequencies are unable to reproduce all these frequencies from a single driver. A large diameter driver is very good at reproducing low frequencies but cannot reproduce high frequencies while a small driver can do the highs but not the lows.

In order to cover all the frequencies cleanly the drivers must be electronically "crossed over", so that the high frequencies are only fed to the HF driver and the low frequencies to the LF driver.

Basic speaker cabs perform this cross passively, feeding the whole range into the speaker and then separating the signals at speaker levels. This is inefficient and wastes amplifier power; it also results in speakers that are less "clean" sounding than active cabinets.

Active cabs use an electronic crossover to split the signal before it goes into the power amplifiers. This means you need more power amp channels, but the trade off is a much cleaner sounding system with much more available power for the same rating speakers as the amps are no longer amplifying signals the speakers can't reproduce. In effect the amp power is concentrated on a smaller part of the frequency spectrum.


Crossovers

Crossovers
Image copyright RANE coroporation, http://www.rane.com

This is a simple 2-way crossover from RANE. The first place a crossover is normally seen is when Subs are added to a passive system in order to increase the LF output. The crossover is placed in the system before the amplifiers and sends the low frequencies to the sub amp and the high frequencies to the amplifier for the tops.

The most critical control here is the frequency control, which sets the point at which the subs are "crossed" to the tops. The cross frequency is the point where the signal to both channels is -3dB from the nominal level (half the power to each channel and therefore summing two halves to make a whole).

Once the crossover point has been set, the next decision is to set the level of the Low and High output. Set the "Master" level to maximum, as the balance between the two is what is being adjusted. Next turn up both the output levels to about half way and consider if the sound is too "boomy" or too harsh. If the former increase the level of the "high", if the latter increase the sub output, continue to adjust until you are satisfied with the balance of your system.

This can of course also be adjusted with assistance from an RTA (see How to EQ) but in practice, in particular for recorded music I like to have the subs quite a bit louder, as much as 9dB louder, than the tops. This is not a reflection on any particular LF boost, but more of a compensation for the fact that to deliver equal LF power to the amount of HF power, you need to have more to start with as each LF wave contains more energy as it is longer. The final knob is the "delay" of the LF channel.

As the LF driver is normally physically in front of the top cab, it must be delayed by a few fractions of a millisecond so that all the sound arrives together. The amount of delay is calculated from the distance the LF driver is in front of the top driver and the speed of sound.


A suggested starting point for crossing subs in a smaller system is 120Hz, but experiment either side to see where it sounds best. Large professional systems often cross the subs in at around 80-90Hz while the passive crossover provided in Peavey and similar subs cross in the tops at around 150Hz.