Hales Audio System Two Signature loudspeaker Measurements
Starting with the Listening Room Diagnostic Recording (LEDR) test on the Chesky Test CD (JD37), I heard the most accurate spatial positioning of any speakers I've had in my listening room. The "Up," "Over," and "Lateral" tests produced contiguous, stable images throughout their apparent travel beyond the loudspeaker boundaries.
Next, the Signatures were driven with a variable-frequency sinewave with my hand and ear on the cabinet walls to detect enclosure resonances. A loudspeaker will typically have several low-frequency modes that set the cabinet to dancing when stimulated. Not surprisingly, the Signatures exhibited virtually no cabinet resonances. The resonances I did detect were few and of low amplitude. The strongest mode was at 41Hz, with no other detectable modes until 111Hz. The rest of the band was resonant-free until 588Hz, where a slight mode was heard, along with one at 667Hz. Overall, the Signatures exhibited the least cabinet vibration of any speaker on which I have performed this test. Some loudspeakers, when driven with a sinewave at their enclosure's resonant frequency, will shake like a paint mixer.
The Signature's impedance magnitude and phase angle are shown in fig.1. Note the overall low impedance in the low-frequency region. A slight impedance peak is visible at the crossover frequency, typical of a second- or higher-order crossover. The minimum impedance was 3.75 ohms at 140Hz, in disagreement with the specified 2.9 ohms minimum impedance. Generally, this is not a difficult loudspeaker to drive, but an amplifier with the ability to deliver current into lowish impedances is indicated by the impedance plot, and suggested by the auditioning with the Krell KSA-200.
Fig.1 Hales System Two Signature, electrical impedance (solid) and phase (dashed) (2 ohms/vertical div.).
Moving on to MLSSA-derived measurements, the Signature's performance was exceptional, correlating very closely to the auditioning impressions. The impulse (fig.2) and step (fig.3) responses reveals that both drivers are connected in phase (something confirmed by looking at each driver individually, fig.4) and the MB tweeter's ultrasonic ringing can be seen overlaying the impulse "tail." Windowing the anechoic portion of the impulse response to remove room reflections and performing an FFT calculation gives the speaker's anechoic response, seen in fig.5. The on-axis woofer response and on-axis tweeter response are plotted separately. The apparent peaks and dips in the woofer response above 2kHz are probably measurement artifacts.
Fig.2 Hales System Two Signature, impulse response on tweeter axis at 45" (5ms time window, 30kHz bandwidth).
Fig.3 Hales System Two Signature, step response on tweeter axis at 45" (5ms time window, 30kHz bandwidth).
Fig.4 Hales System Two Signature, step responses of tweeter (red) and woofers (blue) on tweeter axis at 45" (5ms time window, 30kHz bandwidth).
Fig.5 Hales System Two Signature, acoustic crossover on tweeter axis at 45", corrected for microphone response, with nearfield woofer response plotted below 300Hz.
Looking at the Signature's full-range response, spatially averaged over a 30° lateral window, gives a more accurate view. The result is shown in fig.6, coupled with the nearfield woofer response taken with the measurement microphone nearly touching the dustcap. The –6dB point in the bass appears to be a high (for the size) 41Hz. In addition, a rising treble response within the ±15° lateral window is apparent, due to the domination on this measurement of the central response seen in fig.5.
Fig.6 Hales System Two Signature, anechoic response on tweeter axis at 45", averaged across 30° horizontal window and corrected for microphone response, with nearfield woofer response plotted below 300Hz.
The third-octave, spatially averaged in-room response (fig.7) corresponds very closely to my listening impressions. The Signatures have a remarkably flat response, especially through the critical midrange, with a marginally rising treble and slightly overdamped, lean-sounding bass. The superbly flat response through most of the spectrum correlates to my impressions of tonal neutrality. Also correlating with my impressions while experimenting with placement and toe-in is the set of response curves in fig.8. The top curve was taken directly on-axis, those below it 7.5°, 15°, and 30° off-axis respectively. Note how the top octaves change from having too much to too little energy as the lateral angle increases. The correct amount of HF energy is present with the listener 20° off axis, which is exactly the position I found by ear during the positioning experimentation. All the auditioning was done at this angle.
Fig.7 Hales System Two Signature, spatially averaged, 1/3-octave response in RH's room.
Fig.8 Hales System Two Signature, horizontal response family at 45", normalized to response on tweeter axis, from back to front: differences in response 30°–5° off-axis; reference response; differences in response 5°–30° off-axis.
In the vertical plane, a crossover-related notch appeared on the upper-woofer axis, which will be manifested as a lack of immediacy if the listener sits too high.
Fig.9 shows the cumulative spectral decay plot, also called a "waterfall" graph. Essentially, it is a series of FFT-derived frequency-response curves taken at discrete time slices. The measurement reveals resonances as it analyzes the decay of the impulse driving the loudspeaker. Not surprisingly, the Signatures exhibited a very clean decay through the entire band. There was a slight resonant mode apparent at 4.35kHz, but this is minor. The dark line near 15kHz is not part of the loudspeaker response, but the computer monitor's scanning frequency. Overall, the cumulative spectral decay was excellent.—Robert Harley
Fig.9 Hales System Two Signature, cumulative spectral-decay plot at 45" (0.15ms risetime).