Columns Retired Columns & Blogs |
Dunlavy Audio Labs Signature SC-VI loudspeaker Measurements part 2
The bulk of the loudspeaker, coupled with the lack of any turntable facilities in the DAL anechoic chamber, made it impossible to examine the SC-VI's off-axis behavior. However, during my own auditioning, I did the usual stand-up/sit-down test as well as listening critically at various off-axis angles. The speaker's dispersion does appear to be well-behaved.
In the time domain, the Dunlavy's impulse response can be seen in fig.3. It has an excellent, time-coherent shape, though with some slight HF ringing evident. A reflection can also be seen about 2.5ms after the main pulse, which surprised me considering that this measurement was taken in an anechoic chamber. However, I believe it is due to the speaker's sound reflecting off the manual forklift that John Dunlavy uses to lift this massive speaker off the ground in the chamber.
Fig.3 Dunlavy SC-VI, impulse response on tweeter axis at 10' (5ms time window, 35kHz bandwidth).
The step response on the SC-VI's tweeter axis (fig.4), calculated from the impulse response, is about the best I have ever seen. The initial rise is almost vertical, confirming that the individual sounds from the seven drive-units reach the microphone at the same time, with then a very smooth relaxation of the curve back to the time axis, broken only by some slight ripples and the reflection seen in fig.3. This is a very time-coherent speaker. In fact, although I haven't shown it here, the SC-VI on its tweeter axis is one of the very few loudspeakers to produced well-shaped acoustic squarewaves.
Fig.4 Dunlavy SC-VI, step response on tweeter axis at 10' (5ms time window, 35kHz bandwidth).
Finally, figs.5 and 6 show the SC-VI's cumulative spectral-decay plot. Fig.5, calculated with a 0.01ms risetime and a time-correct Thomson anti-aliasing filter, reveals the quickness of the speaker's decay (although the filter introduces a degree of high-treble droop). Fig.6, plotted using Stereophile's usual 0.15ms risetime and a flat-passband Butterworth anti-aliasing filter, sacrifices time resolution to more easily reveal energy storage problems. (Because of the reflection from the supporting hardware in the impulse response, I have windowed out any data more than 2.5ms after the exciting impulse.) This is excellent performance by any standard, but a slight ridge of decaying energy can be seen at 4kHz. However, SS didn't remark on any subjective problems in the mid-treble, so I must assume that this is benign.—John Atkinson
Fig.5 Dunlavy SC-VI, cumulative spectral-decay plot at 10' (Thomson antialiasing filter, 0.01ms risetime).
Fig.6 Dunlavy SC-VI, cumulative spectral-decay plot at 10' (Butterworth antialiasing filter, 0.15ms risetime).
- Log in or register to post comments