Coaxial Audio and the Point Source Paradox: Why Perfect...

=== The Sound That Arrives Twice ===

You hear it before you understand it. A piano chord strikes, and something feels subtly wrong. The low notes and the high notes do not land at the same instant. The sound is not late. It is split. Two versions of the same event reach your eardrum separated by a fraction of a millisecond, and your brain registers this as a vague sense that the music is not quite whole.

This is not imagination. It is physics. And it happens inside every multi-driver speaker ever built.

When a loudspeaker uses separate drivers for different frequency ranges, those drivers occupy different physical positions. The woofer sits here; the tweeter sits there. Even when the crossover network splits the signal perfectly in the electrical domain, the acoustic output from those two drivers travels different distances to reach your ear. The result is a time smear, a phase offset, a tiny but measurable fracture in the coherence of the sound wave. In recording studios, where engineers spend thousands of dollars chasing phase accuracy, this problem is well documented. In the earbuds you wear on the train, it is barely discussed.

=== What a Point Source Actually Is ===

In acoustics, a point source is a theoretical radiator that emits sound from ...

In acoustics, a point source is a theoretical radiator that emits sound from a single infinitesimal location in space. All frequencies originate from exactly the same point. No path difference between drivers exists because there is only one driver. No crossover timing error can occur because there is no crossover.

Nature provides close approximations. When a branch snaps in a forest, the sound radiates from one physical location. When two objects collide, the acoustic energy originates from a single point of contact. These are not perfect point sources, but the radiating area is so small relative to the wavelengths involved that the approximation holds with negligible error.

The appeal of a point source for audio reproduction is straightforward: if al...

The appeal of a point source for audio reproduction is straightforward: if all frequencies leave from the same point, they arrive at the listener's ear as a single coherent wavefront. Phase relationships between frequencies are preserved. The spatial image is stable. The sound behaves the way it did at the moment of recording.

The problem is that a true point source cannot reproduce the full audible spectrum. A driver small enough to behave as a point source at high frequencies cannot move enough air to produce low frequencies. A driver large enough to generate bass radiates high frequencies directionally, like a flashlight beam rather than a bare bulb. The point source ideal and the full-range requirement are fundamentally at odds.

=== The Multi-Driver Compromise and Its Cost ===

Speaker designers resolved this contradiction by dividing the frequency range among multiple drivers, each optimized for a specific band. A woofer handles the lows. A tweeter handles the highs. A crossover network routes the appropriate signal to each driver.

This solution works well enough that it dominates the industry But it introdu...

This solution works well enough that it dominates the industry. But it introduces a physical separation between drivers that creates a path-length difference to the listener. The equation is simple: the path difference is the driver spacing multiplied by the sine of the listening angle. When that path difference equals half the wavelength of a given frequency, complete phase cancellation occurs at that frequency.

Consider a typical bookshelf speaker with 150mm between the woofer and tweeter centers. At 2kHz, where the crossover region often sits, the wavelength is approximately 171mm. A listening angle of roughly 35 degrees produces a path difference of about 86mm, which is close to half the wavelength. At that angle, a deep null appears in the frequency response. Move your head slightly, and the null shifts to a different frequency. This is comb filtering: a series of alternating peaks and dips caused by constructive and destructive interference between two sources.

In controlled measurements, non-corrected multi-driver designs show comb filt...

In controlled measurements, non-corrected multi-driver designs show comb filtering nulls of 2 to 4dB at specific frequencies. The effect is most pronounced in the 2 to 4kHz range, precisely where human hearing is most sensitive to phase anomalies. Listeners describe the resulting sound character with words like "phasy" or "bleedy," terms that attempt to capture the sensation of a sound arriving in slightly misaligned layers.

The crossover region deserves particular attention because it is where both drivers reproduce the same frequency simultaneously. Even a well-designed crossover cannot eliminate the physical offset. The tweeter's output arrives slightly before the woofer's output, or vice versa, depending on geometry. Step response measurements reveal this timing error clearly: the tweeter transient leads the woofer by 50 to 100 microseconds in typical designs.

=== The Coaxial Solution: Physics Against Physics ===

Coaxial driver design attempts to collapse the multi-driver arrangement back ...

Coaxial driver design attempts to collapse the multi-driver arrangement back toward a single acoustic origin. The principle dates to 1947, when Tannoy filed the first patent for a Dual Concentric driver. In this configuration, the high-frequency horn is loaded through the center of the low-frequency cone. The two drivers share the same axis. The acoustic centers are aligned vertically, minimizing the horizontal path difference that causes comb filtering.

KEF refined the concept in 1988 with the Uni-Q driver, which places the tweeter in the physical center of the woofer. The woofer cone acts as the tweeter's waveguide, and because the two radiating surfaces are concentric, the off-axis frequency response is more uniform than in a conventional two-way design.

But here is the paradox: even in a coaxial arrangement, the voice coil center...

But here is the paradox: even in a coaxial arrangement, the voice coil centers of the two drivers cannot occupy the same point in space. The woofer's voice coil sits behind its cone. The tweeter's voice coil sits behind its diaphragm. In a typical coaxial monitor, the physical offset between these centers ranges from 5 to 8mm. That separation is small against the 150mm spacing of a bookshelf speaker, but it is not zero. And zero is what a true point source demands.

The math is unforgiving. At 1kHz, the wavelength is approximately 343mm. A 1mm path difference corresponds to roughly 0.3 degrees of phase shift. An 8mm offset between voice coil centers produces about 2.4 degrees of phase error at 1kHz. At 8kHz, the same offset produces roughly 19 degrees of phase error. The error grows with frequency because the wavelengths shrink.

This is the point source paradox: the closer you get to a true point source, ...

This is the point source paradox: the closer you get to a true point source, the more the physical constraints of driver construction push you away from it. A coaxial driver is a better approximation than a side-by-side arrangement, but it remains an approximation. The ideal is mathematically impossible because two physical structures cannot share the same center of mass.

=== Digital Correction: When Physics Needs Help ===

Some manufacturers accept that physics alone cannot close the gap and turn to...

Some manufacturers accept that physics alone cannot close the gap and turn to digital signal processing. Genelec's 8361A SAM studio monitor pairs a coaxial driver with an AutoCal system that measures each driver's group delay individually and applies correction filters. The system uses 0.5 milliseconds of look-ahead latency to pre-align the drivers' outputs in the time domain. The frequency response specification of 45Hz to 20kHz within plus or minus 1dB is achieved through this combination of physical alignment and DSP correction.

The approach is effective but carries tradeoffs. The look-ahead latency introduces a small but real delay between input and output. In a studio monitoring context, this is acceptable. In a live performance or gaming application, 0.5ms of additional latency may be perceptible. And DSP correction can only address what it can measure; nonlinear distortions and mechanical resonances that vary with temperature and aging are harder to compensate.

The Genelec benchmark illustrates an important principle: coaxial geometry re...

The Genelec benchmark illustrates an important principle: coaxial geometry reduces the magnitude of the problem, and DSP can further reduce it, but neither eliminates the fundamental fact that two physically distinct radiators are involved. The goal is not perfection. The goal is to push the residual errors below the threshold of audibility.

=== Miniaturization: The TWS Constraint Problem ===

When coaxial driver principles are applied to true wireless stereo earbuds, the engineering constraints become dramatically more severe. A typical TWS earbud cavity provides approximately 1 to 2 cubic centimeters of internal volume for the driver, versus 3 to 5 cubic centimeters in a wired in-ear monitor. The driver diameter is constrained to roughly 6 to 10mm for a coaxial arrangement.

Within that volume, the coaxial driver must compete for space with the batter...

Within that volume, the coaxial driver must compete for space with the battery, the Bluetooth chipset, the noise cancelling microphones, and the printed circuit board. The woofer and tweeter voice coils are separated by mere millimeters, and the tolerances for alignment are measured in tenths of millimeters. A 0.1mm misalignment that would be negligible in a studio monitor represents a significant fraction of the total driver spacing in a TWS unit.

Small drivers also face a bass extension problem. Less membrane area means less air displacement per cycle, which limits low-frequency output. Two engineering responses are common: high-flux neodymium magnets compensate for the reduced membrane area by increasing the force on the diaphragm, and Helmholtz resonance tuning of the earbud cavity extends the low-frequency response by creating a resonant air mass in the sound path.

A further complication specific to coaxial TWS designs is intermodulation bet...

A further complication specific to coaxial TWS designs is intermodulation between the two drivers. The low-frequency driver's diaphragm undergoes large-amplitude excursions to produce bass. Those excursions alter the acoustic environment around the high-frequency driver, which sits within or adjacent to the low-frequency diaphragm. The result is a risk of intermodulation distortion, where the low-frequency motion modulates the high-frequency output. Metal nozzle construction, with its higher hardness and density against plastic, can reduce unwanted resonance in the sound path, but it does not eliminate the coupling problem between the two drivers.

=== A Coaxial TWS in Practice ===

The AVIOT TE-W1 applies a dual dynamic driver coaxial configuration it calls ...

The AVIOT TE-W1 applies a dual dynamic driver coaxial configuration it calls Sound Coaxial 3D, derived from the technology developed for its flagship TE-Z1PNK model. Two dynamic drivers are placed coaxially to minimize the phase difference between them, covering the frequency range from low to high in a balanced arrangement. The metal nozzle and LDAC codec support round out the acoustic engineering, though the core proposition is the coaxial alignment itself.

In this context, the TE-W1 represents a specific engineering answer to the point source paradox: accept that perfect alignment is impossible, but reduce the offset enough that the residual phase error falls below the threshold where most listeners perceive it. The 8mm dynamic driver and balanced armature in a coaxial arrangement cannot achieve zero path difference, but the offset is small enough that comb filtering effects are measurably reduced against a non-coaxial dual-driver TWS.

=== The Boundary of Approximation ===

Coaxial design is not a guarantee of superior sound. The geometry reduces one category of error, phase misalignment between drivers, but it does not address others. The quality of the drivers themselves, the precision of the crossover implementation, the acoustic impedance of the earbud cavity, and the interaction between the nozzle and the ear canal all influence the final result. A poorly executed coaxial design can sound worse than a well-executed single dynamic driver.

The materials matter A metal nozzle reduces resonance com-ed to plastic, bu...

The materials matter. A metal nozzle reduces resonance, but it also changes the acoustic impedance of the sound path, which affects how the driver's output couples to the ear canal. The magnet strength determines how precisely the diaphragm tracks the electrical signal. The crossover frequency and slope determine how cleanly the two drivers hand off to each other in the overlap region.

And there is a philosophical dimension. The point source ideal is an artifact of Euclidean geometry applied to wave propagation. It assumes a listener at a fixed point in space, receiving a wavefront from a fixed source. Real listening conditions violate these assumptions constantly. The listener moves. The earbud shifts in the ear canal. The acoustic environment changes. A coaxial driver that measures well in an anechoic chamber may or may not maintain its advantage in a noisy subway car.

The pursuit of point source coherence in a device the size of a jellybean is ...

The pursuit of point source coherence in a device the size of a jellybean is an exercise in managing the distance between the ideal and the achievable. Every millimeter of offset that is eliminated, every degree of phase error that is reduced, moves the output closer to the coherence of a single acoustic origin. The paradox remains: the last millimeter is always the hardest to close, and it is the one that matters most at high frequencies.

The next time a piano chord sounds whole through a pair of small earbuds, consider that what you are hearing is not perfection. It is a carefully engineered approximation of something that mathematics says cannot exist, built into a space that physics says is too small to work. That the approximation works at all is the quiet achievement of coaxial audio engineering.