Every speaker with more than one driver needs a crossover. Designing a good crossover is part of the black magic of loudspeaker design. With the Quarks you also have to work with severe space constraints. Yet Jeff Bagby pulls it off with only a handful of components…

A Little Bit of Theory at First…

Most decent loudspeakers consist of more than one driver. Each driver is responsible for a different frequency range. This means that the incoming audio signal, which is a mixture of all the frequencies that the music you are listening to is comprised of, has to be split up, such that each driver only “sees” the frequencies it is supposed to reproduce.

To make things more challenging, this is not an all-or-nothing thing. The crossover point (where the woofer cuts off and the tweeter cuts in) is actually not a single point, but a region in the audio spectrum, where the total output of the loudspeaker is comprised by the sum of the signals of both the woofer and the tweeter.

For example, if your crossover is at 1.5kHz, at the crossover point, 50% of your signal is coming from the tweeter and 50% from the woofer. Go up a bit, to say, 1.6kHz and this might shift to 60% tweeter and 40% woofer. Same of course, but in the reverse, when going down in frequency from the crossover point.

Because of this, you also have to take the relative phase that the two drivers are working at with respect to each other into consideration. If they are out of phase, the signals might cancel each other out. This might or might not be desirable, depending on how you designed the speaker, but something you need to be aware of.

Crossover Components

Conventional crossovers are usually comprised of the following parts

• Resistors — A resistor will attenuate an electrical signal passing though it by dissipating parts of its energy as heat. This works regardless of the frequency of the signal
• Capacitors – A capacitor can store a small amount of electrical charge. Higher frequencies can pass through a capacitor better than lower frequencies (direct current cannot pass through a capacitor at all). Also, a capacitor introduces a phase-shift to the signal passing through it.
• Inductors – Similar to a capacitor, an inductor can store tiny amounts of the energy of an electrical signal. Different from a capacitor, it does that by creating a magnetic field in which the energy is stored, rather than by storing electrical charges. This effect gets stronger for higher frequencies, so that higher frequencies cannot pass through the inductor as easily as lower frequencies. Also, an inductor introduces a phase shift but it is opposed to the phase shift introduced by a capacitor.
• Drivers (speakers) themselves: They are basically a combination of an inductor and a resistor.

So there you have it…

The Crossover in the Quarks

This thread on the Parts-Express forum contains amongst other things a schematic of the crossover design for the Quarks. — Looking at the crossover that Jeff Bagby has designed for the Quarks I can get a basic glimpse of how these principles are put to work.

• There is an inductor wired in series with the woofer: Since inductive resistance increases with frequency, this means that higher frequencies get attenuated. Exactly what we want.
• The tweeter is a little more complicated: There is another inductor wired in parallel with the tweeter. Since inductive resistance decreases with decreasing frequency, the inductor is basically shorting out the tweeter for lower frequencies, which is exactly what we want. However, there is also a capacitor and a resistor in series with the tweeter/inductor pair. A resistor dissipates signal energy regardless of frequency, so this generally attenuates the tweeter. I presume this is done because the tweeter is more efficient than the woofer and you need to even out the effect. — The capacitor in series further attenuates lower frequencies (since capacitive resistance increases with decreasing frequency). — I presume this combination of inductor and capacitor is designed such that the phase shifts introduced by both add up to 180%, which is why the polarity on the tweeter is reversed.