Crossovers: A Show and Tell

One of the best weekends of my life involved a train trip to visit a new, out-of-state girlfriend. I barely remember my time with her, but I vividly remember what I read on the way: Vance Dickason’s Loudspeaker Design Cookbook. Dickason’s now-classic tome taught me in great depth how speakers work, something I’d only dimly understood despite having built quite a few speakers myself.

For me, the most enlightening chapter of the Cookbook was the one that deals with crossovers, the part of a loudspeaker that people least understand. By breaking a crossover down into its component parts, Dickason made the subject — well, not simple, but as simple as it probably can be. When I got back to work Monday morning, I immediately yanked the crossover out of a speaker that was sitting in my office, and found that I suddenly understood what functions all those parts served.

If you want to comprehend the most complex crossovers, like the ones found in top-of-the-line models from high-end brands such as Thiel Audio, you’ll need to put in some quality time with the Loudspeaker Design Cookbook and maybe some additional references, too. But if you just want to learn the basics — i.e., how to figure out what the core functionality of a crossover is — keep on reading.

That’s an order

A speaker crossover consists primarily of electrical filters, simple circuits that attenuate sounds above or below a certain frequency. The crossover in a two-way (woofer/tweeter) speaker comprises a high-pass filter that prevents low frequencies from reaching the tweeter, and a low-pass filter that prevents high frequencies from reaching the woofer.

Crossovers are differentiated by the order of the filters. The order of a filter dictates its slope, the reduction in decibels per octave above or below the crossover frequency. A 1st-order filter reduces signals by -6 dB per octave. Thus, a 1st-order high-pass tweeter filter might cut signals by -3 dB at 2,500 Hz, -9 dB an octave below that at 1,250 Hz, -15 dB two octaves below at 625 Hz, and so on. Each additional filter order reduces signals by an extra -6 dB/octave, so a 2nd-order filter gives you -12 dB/octave, a 3rd-order filter gives you -18 dB/octave, etc.

The parts of a filter

The key components of a speaker crossover are capacitors and inductors (or chokes). A capacitor lets high frequencies pass but blocks low frequencies. An inductor lets low frequencies pass but blocks high frequencies.

The capacitors used in speaker crossovers look like little cans. The inductors look like coils of wire, which is exactly what they are.

To filter the low frequencies out of a tweeter, you simply put a capacitor in the circuit. A single capacitor, in series with the tweeter (i.e., connected between the tweeter and one of the speaker’s binding posts) will give you a 1st-order filter. (See figure 1.) Some crossovers in very inexpensive speakers have nothing more than that, just a capacitor and maybe a resistor in series with the tweeter.

If you need more filtering on the tweeter — perhaps to attenuate low frequencies more severely, which would reduce distortion and allow it to play louder — you might want a 2nd-order filter. No prob. Just put an inductor in parallel with the tweeter, i.e., connected across the tweeter’s terminals. (See figure 2.) How can the inductor filter out more lows, when we just said that it lets the lows through? Because the inductor is in parallel, so low frequencies pass through the inductor and effectively short themselves out before they ever reach the tweeter.

For a 3rd-order filter, you’d add one more capacitor in series with the tweeter, after the inductor. (See figure 3.) For a 4th-order filter, add one more inductor in parallel, after the second capacitor. (See figure 4.)

To filter the high frequencies out of the woofer, you do just the opposite. For a 1st-order filter, use an inductor in series with the woofer. (See figure 5.) For a 2nd-order filter, add a capacitor in parallel with the woofer. (See figure 6.) And so on.

“Reading” a crossover

Now that you understand the basics of speaker crossover filters, you should be able to figure out what the filters in a crossover are by tracing the circuits etched on the bottom of the circuit board. It helps to sketch out the circuit on a piece of paper so you can clearly see the function of the different parts.

In relatively simple, low-cost speakers such as the Wharfedale DX-1 shown here, figuring out the crossover may be as easy as counting the capacitors and inductors. Just remember one rule: The high-pass filter on the tweeter is almost always going to be of the same order or higher as the low-pass filter on the woofer. Thus, a speaker might have a 2nd-order low-pass filter on the woofer and a 3rd-order filter on the tweeter, but probably won’t have a 2nd-order low-pass filter on the woofer and a 1st-order high-pass filter on the tweeter. There are exceptions to this rule, but in 20-plus years of reviewing speakers, I can recall only a few.

Thus, if a crossover has three capacitors and two inductors (see figure 7), it probably has a 3rd-order filter on the tweeter and a 2nd-order filter on the woofer. Two capacitors and one inductor form a 3rd-order high-pass filter for the tweeter, while the remaining inductor and capacitor form the 2nd-order low-pass filter for the woofer.

You’ll encounter other components, too. The large, white, rectangular objects are resistors, which in simpler crossovers are typically used to reduce the level of the tweeter. (Tweeters are usually more sensitive than woofers, so their levels need to be brought down to match the woofers’ output.) For the most part, you can ignore these in your circuit analysis.

In more complex crossovers, you’ll find additional filters intended to smooth out response problems in drivers, flatten the impedance curve, align the relative phase of a driver with other drivers, etc. But by sketching out the circuit, you should be able to figure out which components form the core filters of the crossover and which ones are tacked on to tweak the speaker’s performance.

A great example is the crossover I linked to above, which is from a Thiel CS 3.7. Despite its complexity, that’s just a first-order, two-way crossover. So all it needs for its basic functionality is a capacitor and a resistor on the midrange/tweeter driver and an inductor on the woofer. All that other stuff is there just to fine-tune the speaker’s performance. You want to know why the late Jim Thiel was so respected by speaker cognoscenti? Just look at his crossovers.

Three, four, and more

So far, we’ve talked only about two-way speakers. What about three-ways? Four-ways? Well, it’s all pretty much the same, just more complicated.

A three-way speaker’s crossover incorporates four basic filters: a low-pass filter for the woofer, high-pass and low-pass filters for the midrange driver, and a high-pass filter for the tweeter. But not to worry: Now that you know the basics of speaker crossover filter design, with a little patience you should be able to figure out the core functionality of any crossover out there. Unless, perhaps, it was designed by Jim Thiel.

X