In the time domain, the Grand Piano's step response (fig.8) puzzled me: it implies that the tweeter and woofer are connected with opposite polarities, which is not what you'd expect from a design using a first-order crossover. This lack of time coherence is confirmed by a glance at the drive-units' individual step responses (fig.9). The tweeter response is the sharp negative-going trace just before the 4ms mark: it plunges to the negative baseline, then returns and overshoots the time axis, followed by a small amount of ringing that dies away by the 5ms mark. The woofer's response is the slower positive-going rise just after the 4ms mark, which then recrosses the time axis at 6ms. But, as revealed by the on-axis frequency response and the vertical dispersion plot, whatever the speaker's crossover topology, it is engineered to give a smooth integration between the drive-units on the typical listening axis—and that, rather than nominal time alignment, is what matters most.
Fig.8 Sonus Faber Concerto Grand Piano, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).
Fig.9 Sonus Faber Concerto Grand Piano, individual step responses of woofer and tweeter on tweeter axis at 50" (5ms time window, 30kHz bandwidth).
Finally, the Grand Piano's cumulative spectral-decay plot (fig.10) is clean from resonant modes, suggesting a finely detailed presentation.—John Atkinson
Fig.10 Sonus Faber Concerto Grand Piano, cumulative spectral-decay plot at 50" (0.15ms risetime).