Measuring sensitivity in a horn-loaded loudspeaker system
Sensitivity is one of the most useful specifications for a horn-loaded loudspeaker, yet it is often quoted without enough measurement detail. A figure such as 105dB SPL may refer to a single driver, a complete speaker, a narrow frequency band, or an idealised laboratory condition. Those are very different results.
A reliable measurement connects acoustic output with a known electrical input and a defined distance. Once the method is controlled, you can compare a compression driver, a woofer, a passive three-way system, or a complete custom loudspeaker without being misled by room gain or amplifier volume settings.
What sensitivity actually tells you
Loudspeaker sensitivity describes the sound pressure level produced by a specified input, commonly 1 watt at 1 metre. The result is expressed in decibels SPL. Efficiency is related, but it is not identical: efficiency describes the proportion of electrical power converted into acoustic power, while sensitivity is the practical sound level achieved from that power.
Horn systems often produce high sensitivity because the horn improves acoustic loading and directivity. A compression driver may be extremely efficient through the midrange and treble, while a low-frequency horn or woofer section may have a different output profile. The final system sensitivity is therefore governed by the least efficient section, crossover attenuation, impedance, and the chosen frequency range.
A quoted number should always include its conditions. “108dB” without a frequency range, weighting, distance, input voltage, or averaging method has limited value. A fair comparison might specify 1W/1m, an 8-ohm nominal load, and an average across a defined band such as 500Hz to 10kHz.
Prepare the system and room
Start with the loudspeaker fully assembled in its intended configuration. Include the actual horn, compression driver, woofer, passive crossover, grille if normally fitted, and any padding resistor or protection network. Measuring a bare compression driver will reveal its capability, but it will not predict the sensitivity of the finished loudspeaker.
Use a calibrated measurement microphone on a stand, aimed at the acoustic axis. Place it one metre from the horn mouth or the manufacturer’s stated reference point. In a domestic room, keep the speaker away from walls and corners when possible. Early reflections from a timber floor, ceiling, or nearby furniture can easily add peaks and dips that have nothing to do with the driver itself.
For Australian listeners, a detached-room test in Melbourne or Adelaide can be quieter than a busy inner-city Sydney location, but traffic, air-conditioning and birds still matter. A concrete slab can create strong floor reflections, while a typical Queensland room with hard surfaces may exaggerate upper-midrange energy. Short measurement windows or outdoor ground-plane testing help separate the direct sound from these local room effects.
The horn construction details are relevant here because mouth size, flare profile and directivity affect how much acoustic energy reaches the microphone. A measurement made directly on the axis will not represent the same output as one taken 30 degrees off-axis.
Choose the right test signal
A swept sine is a practical choice for a frequency-response measurement because it shows the complete sensitivity curve and makes resonances easy to identify. Pink noise can be useful for checking average level, but it requires a stable analyser and careful bandwidth control. A single tone is suitable for a quick spot check, provided the frequency is safely within the driver’s operating range.
Set the amplifier to a known RMS voltage rather than relying on its volume knob. For a nominal 8-ohm load, 2.83V RMS is approximately 1 watt. The relationship is:
[ P = \frac{V^2}{R} ]
This means 2.83V produces 1W into 8 ohms, but about 2W into 4 ohms and 0.5W into 16 ohms. That difference explains why voltage sensitivity and power sensitivity can appear to disagree.
| Measurement condition | What it shows | Main limitation |
|---|---|---|
| 1W/1m | Acoustic output for a defined power input | Requires known impedance |
| 2.83V/1m | Convenient amplifier-voltage comparison | Equals 1W only at 8 ohms |
| Half-space ground plane | Strong low-frequency repeatability | Does not represent every room |
| Anechoic gated result | Direct sound and controlled reflections | Limited low-frequency resolution |
| In-room average | Practical listening level | Strongly affected by placement |
Keep the test level high enough to rise above the room’s background noise, but low enough to avoid compression, limiter action, horn distress or woofer excursion. With a high-sensitivity compression driver, a few volts can produce a surprisingly high SPL. Hearing protection is sensible, especially in a small listening room.
Calculate level from voltage and impedance
Measure the amplifier voltage at the loudspeaker terminals with a true-RMS multimeter or an audio analyser. Do not assume that the amplifier’s display is accurate. If the crossover presents a varying impedance, use the impedance magnitude at the test frequency or report the result as voltage sensitivity rather than claiming a precise watt-based figure.
For example, a speaker producing 105dB SPL at 2.83V and measuring 8 ohms at that frequency is approximately 105dB at 1W/1m. If its impedance falls to 4 ohms, the same 2.83V is 2W, so the power-based sensitivity is roughly 3dB lower. The distinction becomes important when comparing a high-impedance compression driver with a lower-impedance woofer.
Measure SPL using linear or Z-weighting where available, rather than A-weighting. A-weighting reduces bass and is designed around human hearing at moderate levels, so it can make a full-range loudspeaker appear more sensitive than it really is. Report the microphone distance, test voltage, bandwidth, weighting and averaging method alongside the result.
Interpret multiway and crossover results
A three-way horn system does not have one perfectly flat sensitivity number. The high-frequency horn may exceed 110dB through part of its range, while the bass section may sit closer to 95dB. The crossover determines how these sections combine, and padding on the compression driver may intentionally reduce its level to match the woofer.
Time alignment also affects the measured sum around each crossover point. If acoustic centres are displaced or polarity is incorrect, the response may show a deep cancellation even though each individual driver measures well. The principles behind a three-way crossover design are therefore inseparable from sensitivity measurement.
Take separate traces for each way at a safe level, then measure the complete system. The individual traces help identify the source of a problem; the full-system trace shows the performance that matters for listening. Record both on-axis and at useful off-axis angles, since a horn’s directivity narrows as frequency rises.
Turn measurements into useful design decisions
A useful sensitivity report might state: 102dB SPL average from 500Hz to 10kHz, measured at 1m on axis, 2.83V RMS, with a 4-ohm minimum impedance and a gated or ground-plane method. That description is far more informative than a single peak value. Include the frequency-response graph when possible, because averaging can conceal a sharp resonance or crossover dip.
Use the result to estimate amplifier requirements, headroom and listening distance. In free-field conditions, SPL falls by roughly 6dB each time distance doubles. Room reinforcement changes that relationship, especially in the bass, but the estimate remains useful. A 102dB-sensitive system needs vastly less amplifier power for a given peak level than a conventional 87dB loudspeaker.
For a home in Perth, Canberra or regional New South Wales, the final decision may also depend on room size, neighbours and available mains power. Australia’s 230V, 50Hz supply is more than adequate for normal high-end amplifiers, yet amplifier quality, noise floor and protection behaviour matter more than a large wattage figure with a horn system. Measure carefully, publish the conditions, and judge the complete loudspeaker rather than its loudest isolated driver.