🔊

Crossover Frequency & Slope Calculator

Design 2-way and 3-way passive speaker crossovers. Pick the slope (6, 12, or 18 dB/oct), the alignment (Butterworth, Linkwitz-Riley, Bessel), and the driver impedance — the tool computes every L and C, picks the driver polarity that flattens the summed response, and plots the actual simulated network response (not just the ideal curves).

Input

Linkwitz-Riley and Bessel are available at 12 dB/oct only. 1st & 3rd order are Butterworth.
Typical 2-way crossovers: 1.5–3.5 kHz.

Result

Design notes & formulas
Schematic
Combined frequency response — actual simulated LC network

Choosing a Slope & Alignment

A passive crossover splits the audio signal into bands so each driver only sees the frequencies it's good at. Two design choices dominate: the slope (how fast each filter rolls off, in dB per octave) and the alignment (the mathematical filter response that defines exactly what happens at and around the crossover frequency).

Slopes

6 dB/oct (1st order) — the gentlest slope. One inductor for the woofer, one capacitor for the tweeter. Cheap, simple, and has zero phase shift at the crossover, but the drivers overlap broadly so both reproduce a wide region together. Demands well-behaved drivers without nasty out-of-band breakup.

12 dB/oct (2nd order) — the standard "good enough" slope. Sharper rolloff means less driver overlap, less out-of-band stress, and looser driver requirements. The trade-off is a 180° phase shift at the crossover, which is why 2nd-order crossovers usually wire one driver with reversed polarity to flatten the summed response.

18 dB/oct (3rd order) — sharper still. Very little overlap; ideal when you need to keep a tweeter well above its resonance, or protect a small driver from low frequencies. Wires up with normal polarity for both drivers (Butterworth 3rd-order sums flat in magnitude). Component count and cost go up.

24 dB/oct (4th order) is not offered here. Fourth-order passive crossovers are technically possible but rare in well-designed loudspeakers — the singly-terminated 4th-order ladder is extremely sensitive to component tolerances and driver-impedance variation. When 24 dB/oct is needed, the standard solution is an active Linkwitz-Riley crossover built from cascaded op-amp Sallen-Key sections (which is well covered by our other filter calculators).

Alignments at 12 dB/oct

Butterworth (Q = 0.707) — maximally flat amplitude. Each driver is −3 dB at the crossover; the sum has a +3 dB bump there. The classic choice when you want flat response in the passband.

Linkwitz-Riley (Q = 0.5) — the audiophile favourite. Each driver is −6 dB at the crossover; the sum is perfectly flat (0 dB) when one driver is wired with reversed polarity. The trade-off is the −6 dB acoustic crossover means more driver overlap.

Bessel (Q ≈ 0.577) — between the two. Most natural transient response (linear phase), favoured in studio monitors. Drivers are about −4.8 dB at the crossover and the sum is close to flat (within a dB) at fc.

Why polarity matters

Filters introduce phase shift. When two drivers overlap, they can add (in phase) or cancel (out of phase). The tool automatically picks the polarity combination — normal or inverted — that gives the flattest combined response at the crossover frequency. For most 12 dB/oct designs that means inverting one driver; for 6 dB/oct and 18 dB/oct Butterworth, both drivers stay normal.

Driver impedance — the lie that's mostly fine

These formulas assume the driver looks like a constant resistance R across the whole band. In reality drivers have a resonant rise around their free-air resonance, a mostly-resistive midband, and a voice-coil inductance rise at the top. The calculator's output is the ideal starting point — for a finished design you'd add a Zobel network on the woofer (to flatten its high-frequency impedance rise) and possibly an L-pad to attenuate a too-loud tweeter. Always measure the in-box response.

Frequently Asked Questions

What slope should I pick?
If the drivers behave well a full octave outside the crossover, 6 dB/oct is musical and cheap. For most real two-ways, 12 dB/oct Linkwitz-Riley is the default — flat sum, modest component count, forgiving of driver imperfections. 18 dB/oct when you need to keep a sensitive tweeter well above its resonance or protect a small driver from low-frequency excursion.
Why does the calculator say "invert the tweeter"?
Second-order filters introduce a 180° phase shift between their input and output at the crossover frequency. With the same polarity, the LP and HP outputs would cancel at fc and create a deep null. Reversing polarity on one driver (usually the tweeter) puts them back in phase at fc so the sum is flat. The calculator picks this automatically based on the alignment.
Why are component values different from another online calculator?
Two reasons. First, "alignment" matters — Butterworth, Linkwitz-Riley, and Bessel use different Q values and produce different L and C. Second, some calculators design for a doubly-terminated network (source R + load R), which gives values about 2× different than the singly-terminated (voltage-source amp, load R) network used in real speaker crossovers. This tool uses the singly-terminated formulas and verifies every coefficient by simulating the actual LC ladder.
Can I use these values without measuring the driver?
As a starting point, yes — the formulas assume a constant-resistance driver and will get you close. For a finished design, measure the drivers' impedance and SPL response in your enclosure. A Zobel network on the woofer (compensating its voice-coil inductance rise) and an L-pad on the tweeter (matching sensitivities) are the usual additions. The textbook formulas are a "Tuesday afternoon" design — the real design is finished by ear and by measurement.
What about a 4-way or sub-bass crossover?
For more than three ways, or for very low (sub-bass) crossovers, passive becomes impractical: the inductors get enormous and lossy, capacitor count balloons, and component tolerances dominate the result. Active crossovers (line-level filters before separate power amps per driver) are the standard solution. Our Band-Pass, Low-Pass, and High-Pass calculators cover the building blocks for active designs.
Why no 24 dB/oct?
A 4th-order singly-terminated passive ladder is genuinely finicky — component tolerances and driver impedance variation cause large deviations from the target Linkwitz-Riley response. In professional and DIY practice, 24 dB/oct crossovers are almost always built actively as cascaded 2nd-order Sallen-Key sections (where the op-amp's near-zero output impedance makes each stage isolated and predictable). To keep this tool honest, we ship only the orders we can verify by ladder simulation.
What is the "sum" curve in the plot?
The acoustic sum of all driver outputs at the listening position, assuming the drivers are coincident and ideally resistive. A flat sum curve means the speaker reproduces the input signal faithfully; a +3 dB bump (typical of Butterworth) or a notch (wrong polarity) shows up here. The tool picks the polarity combination that gives the flattest sum at the crossover frequencies.