There are two general types of crossover networks in use today: active and passive. Used within their limitations, both are equivalent, given equivalent specs but as you will see, the passive network has very definite limitations.
First, what is the difference between an active and passive crossover? Very simply stated, the passive crossover requires no external power source besides the audio signal itself. It does not depend on any sort of external power. An active crossover requires an external power source for operation (i.e. you must feed it batteries, or plug it in).
A second difference arises in terms of design. Passive crossovers require one reactive component per 6dB of slope. A reactive component is either an inductor (coil) or a capacitor. Active crossovers use resistors and capacitors to achieve the same end result. Furthermore, when the crossover design begins to get complicated, say beyond 18dB/octave, active crossovers take the lead because they’re much easier to design. You can literally get any slope you want...from 6dB/octave to 72dB/octave...and beyond.
Passive crossovers that are typically used in conjunction with loudspeakers operate best when presented with a resistive load, which a loudspeaker isn’t. Most better designs attempt to compensate for the non-resistive nature of the loudspeaker load, but you can only do so much. In addition, the low frequency speaker has an inductor in series with it, which degrades the damping factor (ability of the amplifier to tell the speaker what to do) of the amplifier and causes small, but measurable amounts of signal loss.
A passive crossover isn’t always used directly with a loudspeaker. Indeed, many of the earliest electronic crossovers were totally passive...textbook implementations of the appropriate network. The difference here is the signal levels involved. A passive crossover connected between the amplifier and loudspeaker system is termed a high-level crossover. A passive crossover (or active crossover for that matter) used for biamplification is termed a low-level crossover. An active crossover uses active filter technology to create the response characteristic of a similar passive filter network. Active crossovers are sometimes called electronic crossovers.
Active crossovers have the following advantages:
- The crossover is isolated or buffered from the loudspeaker load by the power amplifier.
Since the speakers are connected directly to the amplifiers, the damping factor is limited
only by the speaker impedance and the wire resistance. The amplifier has better control
of the speaker's cone motion. - The insertion loss of the crossover network is no longer a factor.
- The power rating of the crossover is no longer a factor.
- It is much easier to tune to precisely the crossover frequency that you want.
- Steeper slopes are possible without the insertion loss of a high-slope passive crossover network.
- Intermodulation distortion caused by amplifier clipping is reduced.
How to choose the right crossover frequency
A consequence (or privilege) of using an active crossover is that you can easily tune the crossover
to almost any crossover frequency. In some cases, you may have control over slope and other
aspects of the filter characteristic. Changing a module to change the crossover frequency is trivial.
Picking the right crossover frequency is anything but trivial.
The following factors influence your choice of crossover frequency:
- Driver (loudspeaker) limitations
- Polar pattern of the driver (dispersion)
- Performance goals
Driver limitations
One of the purposes of a crossover is to allow the driver to operate in the optimum region of its
frequency response. For tweeter drivers, the low frequency rolloff provides
protection to the driver from excessive diaphragm excursion as well as preventing the driver from
operating where it can produce little or no output. In a similar vein, the high frequency rolloff of the bass and tweeter crossover outputs prevents
these drivers from operating in a region of their response curve where they may have irregular
frequency response, or severe dispersion problems.
The consequences of operating a low frequency speaker at too high a crossover point are never
fatal. The usual problem is simply that of a hole in the tweeter because the low frequency
speaker may not have much output in that region.
On the other hand, picking a tweeter crossover point that is too low may be fatal to the tweeter
driver. There are two basic failure modes here: thermal and over-excursion. Thermal failure
occurs because the low crossover point directs additional energy into the driver, which may not be
equipped to handle it (the voice coil burns up). Over-excursion failure occurs because the lower
crossover point requires additional cone movement from the driver. If you exceed the cone excursion
limit of the driver, it will most likely fail.
The Driver’s Polar Pattern
The polar pattern of a driver is simply its dispersion characteristic, which is anything but consistent with frequency. Why does it matter? It matters because the crossover allows the two loudspeakers operating throughout the crossover region (that is, the region where the two curves cross) to contribute simultaneously to the system’s total acoustical output. Dispersion effects the ability of the system to project a stable stereo image over the sound stage. Irregular dispersion cause imaging problems such as the inability to localize certain instruments, inconsistent localization throughout the listening are or frequency response problems in the crossover region. The optimum crossover point is that where the dispersion characteristics of the two drivers involved match. If you are not an equipment manufacturer, this may be difficult to do. Remember, the limitations of the drivers come first.
Performance Goals
The performance goals of the completed system may not effect the actual crossover frequency so
much as they may effect the number of crossover points. This is not to say that more is better. A
well designed two-way system can easily hold its own against three-, four-, and five-way systems.
Generally, getting more performance out of a system means obtaining drivers whose performance
characteristics are optimized over a wider range. It can also mean buying more drivers, and
operating them over a smaller portion of their operating characteristic. This is the idea behind
most multi-way (three-way and up) systems. The multi-way system is not a panacea for poor
design...the design can be more difficult because of the multiple crossover points (phase interaction,
lobing errors, time offset correction, etc.).
For some systems, performance goals do influence the choice of crossover frequency. In horn
loaded systems, especially those where the high frequency portion of the system is horn loaded
(such as a traditional “theatre” system) you generally want the crossover point to be as low as
possible to take advantage of the horn loading on the driver. The limiting factors are the low frequency limitations of the high frequency driver (exceed them and it will die on you), how loud you want the thing to go (the lower crossover point restricts the maximum output
because the diaphragm must move further. . . it can only move so far), finally the low frequency limit of the horn (it can no longer load the driver below a certain frequency).
Bi-amplification (or tri-amplification) is a logical step for any high performance audio system. Well
designed systems can play louder, for longer periods of time, without listening fatigue.
If you are
searching for an answer to the rigors of digitally recorded source material,
bi-amplification may be
the answer, providing that your speakers are up to the task!