Sunship Audio

How Impedance Networks Shape Crossover Attenuation in Real Rooms

Behind every high-efficiency horn loudspeaker sits a quiet network of resistors, inductors, and capacitors doing the work most listeners never see. The impedance network inside a passive crossover determines how much energy each driver receives, how smoothly that energy is rolled off, and how well the system behaves with a real amplifier. When that network is treated as an afterthought, even the finest compression drivers sound uneven. When it is engineered as carefully as the horns themselves, the result is a coherent, fatigue-free presentation that holds together across volume levels and acoustic spaces.

In Australia, where listening rooms range from converted inner-Sydney terraces with exposed brick to sprawling Queenslander verandas with high ceilings, the demands on a crossover are particularly varied. Add the long supply chains that force audiophiles in Perth to order components well in advance, and the case for getting attenuation right the first time becomes clear. Sunship Audio treats the impedance network as the connective tissue between driver, horn, and room, and the listening room in Berlin confirms every choice before a single pair ships.

Passive crossover basics for horn-loaded systems

A passive crossover is, at its heart, a frequency-selective filter network that splits an amplified signal into bands suitable for each driver. In a horn-loaded loudspeaker, the high-sensitivity compression driver and the large-format woofer rarely share much common frequency territory, so the crossover must deliver each band with minimal overlap and clean roll-off. The attenuation network — the part that actually reduces the level feeding a particular driver — relies on reactive components to achieve these slopes.

Resistive elements are added to the impedance network whenever a driver is too sensitive for its assigned band. A horn-loaded tweeter, for instance, may need two or three decibels of padding to match the output of a midrange horn that handles 800 Hz to 6 kHz. Pure L-pad designs, made from a fixed resistor and a series resistor with a bypass capacitor, appear in commercial speakers but rarely satisfy the precision expected of a hand-built studio monitor or a high-end domestic system.

How attenuation networks shape driver output

Attenuation differs from filtering, though both share components. A second-order high-pass on a tweeter provides both roll-off and some attenuation. A dedicated attenuation stage is added when the natural sensitivity of a driver is higher than the system reference. This is where the impedance network earns its keep, shaping the final balance between a TAD-Pioneer compression driver and the wooden bi-radial horn in front of it.

The challenge is that real drivers do not present a flat impedance. A compression driver can swing from 6 ohms at its resonance to over 30 ohms in its operating band. If the attenuation network is designed around a single nominal value, the level reaching the driver will vary with frequency. Skilled designers therefore model the driver impedance and place the network's components in compensation configurations, producing a flat response through the crossover region rather than the bumpy curve a naive resistor would deliver.

The behaviour of reactive components

Inductors block high frequencies and pass lows; capacitors do the opposite. Together they form the slopes of the crossover, but their behaviour changes with frequency in ways that affect attenuation profoundly. A first-order inductor behaves predictably above a few hundred hertz; below that, its core can saturate, the resistance of its windings adds unwanted loss, and the very slope you intended begins to drift.

This is where material choice becomes audible. As explained in why we choose copper over aluminum in our inductors, the company specifies heavy-gauge copper wire wound on laminated iron or air cores rather than the lighter aluminum alternatives that have crept into the industry. Copper's lower resistance preserves damping, while careful core selection keeps inductance stable across the signal swing. The result is an attenuation network that does exactly what it was modelled to do, instead of slowly surrendering to heat and saturation during a long listening session.

Matching driver sensitivity with resistor networks

Resistive pads are the simplest way to drop a driver's level. An L-pad places a resistor in series with the driver and another across the input, dissipating excess energy as heat rather than passing it on to the voice coil. The arrangement is cheap, predictable, and appears in most commercial loudspeakers. In a hand-built horn system, however, it serves as a starting point rather than a final answer.

Resistors behave the same at every frequency, while the drivers they connect to do not. A compression driver's rising impedance above resonance amplifies the effect of any series resistance, and a horn's acoustic loading shifts the impedance curve further. Skilled designers therefore treat resistor networks as one tool among many, often adding small inductors or capacitors to compensate for the driver impedance and produce a smoother curve through the crossover region.

Tuning attenuation for bi-radial horns

Bi-radial horns — the curved wooden flares that define the look of a Sunship Audio speaker — present their own demands. Their acoustic impedance varies with frequency, and the compression driver mounted at the throat sees a load that changes as the horn transitions from its lower cutoff to its upper operating range. The attenuation network has to anticipate these variations so that the perceived level at the listening seat remains constant.

A first-pass design rarely survives this checklist intact. Designers typically rebuild the impedance network two or three times before the horn integration sounds seamless, and the time spent in the Berlin demonstration room catches the small discontinuities that a measurement alone can miss.

Time alignment and phase coherence

Attenuation networks also influence phase. A purely resistive pad shifts signal in a predictable way, but a reactive network introduces frequency-dependent delay that can pull a driver out of step with its neighbours. In a multi-way horn system, the midrange horn and the tweeter horn sit at different physical depths inside the cabinet, so the crossover must do some of the work that mechanical alignment cannot.

Time-aligned passive crossovers achieve this by selecting component values that produce the correct acoustic centre for each driver at the crossover frequency. When the attenuation stage is designed in sympathy with the slope and the physical offset, the wavefronts from both horns arrive at the listener together. When it is not, the stereo image softens and the soundstage loses its grip on low-level detail — the kind of micro-dynamics that make a solo violin sound alive in a live recital recording.

Listening results in real Australian rooms

The proof of any attenuation network is what happens outside a Berlin demonstration room. Australian listening spaces are acoustically diverse: a Federation-era home in Adelaide with plaster walls and high ceilings behaves nothing like a modern apartment elsewhere with glass on three sides and polished concrete floors. The attenuation chosen for a system needs to be robust enough to sound musical across these differences.

Customers often describe how a Sunship system held its tonal balance when they took delivery in Brisbane and reinstalled it in a converted warehouse studio. The impedance network was tuned for the drivers and horns, not for a particular room, so the speakers adapt rather than fight the acoustics. That portability of character explains why the brand has loyal owners in Sydney and Perth despite the absence of an Australian distributor — every pair leaves Berlin pre-tuned, and only minor resistor swaps are needed if a customer wants to fine-tune the balance for an unusually live or dead space. The electrical safety regime under AS/NZS 60065 also requires careful attention to component ratings and internal wiring, and the company specifies insulation and grounding consistent with Australian standards so imported units can be commissioned by a licensed electrician without modification.