Sidebar 3: Measurements
I used DRA Labs' MLSSA system and a calibrated DPA 4006 microphone to measure the Zu Audio Soul Supreme's frequency response in the farfield; for the nearfield frequency responses, I used an Earthworks QTC-40, which has a ¼" capsule and thus doesn't present a significant obstacle to the sound.
My estimate of the Zu Soul Supreme's voltage sensitivity was 91.5dB/2.83V/m, which is lower than the specified 97dB/W/m but still much higher than average. This speaker will play loudly with low-powered amplifiers. In addition, its plot of impedance magnitude and electrical phase (fig.1, plotted with a wider vertical scale than usual) reveals the Soul Supreme to be a very easy load for the partnering amplifier. Though the phase angle is occasionally high, the minimum magnitude is a still-high 8.4 ohms at 260Hz and the average impedance is around 16 ohms. This partly explains the measured sensitivity being lower than specified; at 2.83V, the speaker will demand from the amplifier 0.5W rather than 1W.
.1 Zu Audio Soul Supreme, electrical impedance (solid) and phase (dashed) (5 ohms/vertical div.).
Several wrinkles in the impedance traces suggest the presence of resonances of various kinds. The largest discontinuity lies just below 200Hz, but when I tested the enclosure's panels with a plastic-tape accelerometer, I found nothing amiss in that region. However, there was a fairly strong resonance at 340Hz on all of the panels, and strongest on the back (fig.2), where it was joined by an even-higher-level mode at 590Hz. Herb Reichert didn't mention any midrange coloration that could be ascribed to these panel resonances, nor did I hear anything amiss when I listened to the Soul Supremes. It's possible, therefore, that the fact that these modes are of high Q (Quality factor), coupled with the speaker's high sensitivity, allows them to step out the music's way most of the time.
Fig.2 Zu Audio Soul Supreme, cumulative spectral-decay plot calculated from output of accelerometer fastened to center of rear panel (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz).
Though Zu takes pains to explain that the Soul Supreme's downward-firing, finger-shaped ports do not behave as they would in a typical reflex-loaded speaker, the impedance-magnitude trace does have the double-humped shape in the bass that indicates some sort of reflex design. The saddle in the trace centered on 45Hz implies that this is the ports' tuning frequency. However, the primary drive-unit's response, measured in the nearfield (fig.3, blue trace), has the expected minimum-motion notch a little higher in frequency, at 52Hz.
The output of the ports—again measured in the nearfield, but with the speaker raised a little higher off the floor than the recommended ¼" so that I could poke the tip of the Earthworks microphone under the cabinet—peaks between 40 and 70Hz (fig.3, red trace). However, a significant peak in its upper-frequency response at 182Hz coincides with the largest wrinkle in the impedance traces. Because this resonance is out of phase with the woofer's output, the complex sum of the woofer and port responses (fig.3, black trace below 300Hz) has a notch centered at 182Hz. However, as the ports face the floor, it's probable that their upper-frequency output at the listening position will not be strong enough for this notch to fully develop. Again, when I listened to the Zus, I was unaware of any problem in the transition region between the upper bass and lower midrange. This graph reveals no sign of the usual nearfield "bump" in the upper bass, which suggests that the Soul Supreme's low-frequency alignment is tuned to favor control and definition rather than boom.
Fig.3 Zu Audio Soul Supreme, anechoic response on tweeter axis at 50", averaged across 30° horizontal window and corrected for microphone response, with nearfield responses of woofer (blue), port (red), and their complex sum (black), respectively plotted below 300Hz, 700Hz, and 300Hz.
Higher in frequency in fig.3, the Zu's response on the tweeter axis, which is 36" above the floor, is superbly flat between 300Hz and 1.8kHz; ie, throughout the entire midrange. Though the response looks very uneven above 2kHz, it reminded me of something Martin Colloms said to me in the late 1970s, when I used to take part in the blind listening tests he then organized for Hi-Fi Choice: that when narrowband peaks and dips in a speaker's treble are equal in amplitude and width, they tend to balance each other. The perceived balance is thus more neutral than the measured response would suggest. I note that though HR mentioned that he listened for "strangeness, sibilance, or discontinuity in the upper harmonics of the ranges of male and female voices," he heard nothing amiss, though he did comment on a rise between 2 and 5kHz.
The perceived balance will also be affected by the fact that the peaks and dips in the Zu's on-axis response tend to be compensated for in the off-axis behavior. This can be seen in fig.4, which plots the difference between the off- and on-axis responses at 5° intervals, from 90° on one side of the speaker to 90° on the other. For example, the cursor is positioned at 1.7kHz, the center frequency of one of the on-axis peaks, and it shows that that peak is balanced by a notch to the speaker's sides. It looks as if the primary driver's "whizzer" cone does improve the Soul Supreme's dispersion in the treble, as Zu claims. Though the tweeter's output above 10kHz looks very ragged in fig.3, fig.4 shows that its dispersion smoothly decreases to the speaker's sides. The speaker's vertical dispersion (fig.5) is difficult to interpret, but I believe it reveals that the optimal balance will be on or just below the tweeter axis.
Fig.4 Zu Audio Soul Supreme, lateral response family at 50", normalized to response on tweeter axis, from back to front: differences in response 90–5° off axis, reference response, differences in response 5–90° off axis.
Fig.5 Zu Audio Soul Supreme, vertical response family at 50", normalized to response on tweeter axis, from back to front: differences in response 15–5° above axis, reference response, differences in response 5–10° below axis.
In the time domain, the Soul Supreme's step response (fig.6) has a superb, time-coincident, right-triangle shape, and it is in the correct polarity. I checked this with an old-fashioned test: briefly connecting a D-cell battery across the speaker terminals and watching which way the cone moved. It did indeed move toward me when hit with the 1.5V DC voltage.
Fig.6 Zu Audio Soul Supreme, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).
Finally, the Zu Soul Supreme's cumulative spectral-decay or waterfall plot (fig.7) looks awful, with a significant resonant mode at 2.8kHz—that accounts for Herb's presence-region rise—and hashy-looking delayed energy in the top audio octave. Yet, other than noticing a somewhat clanky quality with recordings of acoustic piano, I didn't find the Zu to sound as bad as this graph implies. In fact, I enjoyed the afternoon I spent listening to the speakers in Herb's system. As with Zu Audio's Essence speaker, reviewed by Art Dudley in October 2009, Sean Casey appears to have carefully balanced the Soul Supreme's performance to sound more neutral than its measured behavior would suggest, allowing the listener to appreciate its high sensitivity and evenly balanced midrange.—John Atkinson
Fig.7 Zu Audio Soul Supreme, cumulative spectral-decay plot on tweeter axis at 50" (0.15ms risetime).
