Verity Audio Sarastro loudspeaker Measurements

Sidebar 3: Measurements

The Verity Sarastro was almost 6dB more sensitive than average, at an estimated 92.5dB(B)/2.83V/m—within experimental error of the specified 93dB figure. Its impedance stays above 6 ohms for most of the audioband (fig.1), which would imply that the speaker is an easy lead for an amplifier to drive. However, the impedance magnitude drops to 3 ohms at 8kHz, and there is an awkward combination of 4.1 ohms and a –54 degrees electrical phase angle at 5.5kHz, which suggests a good 4 ohm–rated amplifier would work best. The general shape of the impedance curve indicates that driving the Sarastro with a tube amplifier's relatively high source impedance will shelve down the top couple of octaves, though there will also be a slight peak in the presence region that might well be perceived as extra detail.

Fig.1 Verity Sarastro, electrical impedance (solid) and phase (dashed). (2 ohms/vertical div.)

The small bump just above 500Hz in the impedance magnitude trace could be due to a cabinet resonance of some kind. When I examined the vibrational behavior of the Sarastro's panels with an accelerometer, I did find a relatively mild resonant mode at 537Hz in the head-unit cabinet that was present on all surfaces. However, as can be seen from fig.2, a cumulative spectral-decay plot calculated from the output of a plastic-tape accelerometer fastened to the center of the rear panel, two more severe resonant modes are also present, at 435Hz and 457Hz. That MF didn't notice any midrange congestion with the Sarastro is presumably due to the fact that the rear panel faces away from the listener, and that the radiating area affected by this mode is relatively small. Remember, too, that I place a speaker on upward-pointing cones to perform this measurement, which allows resonances to develop to their fullest extent my (see article on this subject). In actual use, the Sarastro head unit sits on mass-loaded Sorbothane pads, which will reduce the amplitude of cabinet vibrations.

Fig.2 Verity Sarastro, cumulative spectral-decay plot calculated from the output of an accelerometer fastened to the head unit's rear panel (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz).

The Sarastro's bass bin sounded quite lively to the knuckle-rap test, but using the accelerometer to look at its vibrational behavior (with the head unit sitting on it but disconnected) uncovered only a couple of panel resonances, at a relatively low level. The waterfall plot in fig.3, for example, was calculated from the accelerometer's output when it was fastened to the center of the front panel. Though ridges of resonant behavior can be seen at 185Hz, 220Hz, 309Hz, and 350Hz, these are down in level, presumably because the stimulus is being attenuated by the woofer's low-pass crossover filter. But they could be excited by vibrations transmitted from the head unit, which is presumably why Verity uses an elaborate, mass-loaded, compliant interface between the two cabinets.

Fig.3 Verity Sarastro, cumulative spectral-decay plot calculated from the output of an accelerometer fastened to the bass unit's front panel level with the woofer (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz).

As is often the case with speaker designs using more than one port, the Sarastro's low-frequency behavior is complex. Fig.4 shows the individual outputs of the midrange unit (black trace), the port in the head unit that loads the midrange unit (blue), the woofer (red), and the bass-bin port (green). Looking first at the woofer, it has the expected reflex notch in its response, this lying at 29Hz, the frequency of the "saddle" in the impedance magnitude trace (fig.1). The woofer rolls off above 110Hz with a relatively shallow slope and a peak evident at 400Hz. The port that loads the woofer (green) covers a broad bandpass centered on the tuning frequency of 29Hz, which suggests excellent low-frequency extension. However, the port output is significantly lower than is required to fully extend the woofer's output in the low bass; Verity here relies on a degree of room-boundary reinforcement to get a flat response. Note that the port has some severe resonant peaks apparent in its midrange output, even with the rolloff due to the crossover. Listening to pink noise through the bass bin alone, I found these peaks audible.

Fig.4 Verity Sarastro, nearfield responses of midrange unit (black), head-unit port (blue), woofer (red), and bass-unit port (green), scaled in the ratio of the radiating diameters.

Fig.4 also reveals rather peculiar behavior on the part of the midrange unit (black). It appears to be shelved-down a little in the midrange, before starting its high-pass crossover rolloff. However, the slope of the rolloff is very shallow, with a small peak present at 65Hz in its output, below which is a deep notch centered at 41Hz. This might be thought to correlate with the tuning frequency of the head unit's reflex port; however, fig.4 indicates that the maximum port output (blue) coincides with the peak at 65Hz in the woofer's response. Looking at the head unit's impedance (not shown) indicated that this port is actually tuned to 56Hz. Perhaps its tuning and that of the midrange unit are affected by the presence of a series capacitor in its feed, which is also revealed by the impedance plot. And, like the bass bin's port, the head unit's port has some peaks evident in the midrange.

The black trace in fig.5 shows how these nearfield measurements sum in the farfield (taking acoustic phase and the relative distance of each sound source from the nominal measurement point), spliced at 400Hz to the measured farfield response averaged across a 30 degrees horizontal window on the tweeter axis. The shelved-down lower midrange is evident, as is a significant midbass peak. This peak to some extent will be due to the nearfield measurement technique, but the combination of peak and dip is responsible, I'm sure, for Mikey's finding the Sarastro's bass to be both "prodigious" but also "somewhat rich and ripe" and "underdamped." That he found the bass tuning more acceptable at low levels is due to this balance mimicking the so-called "loudness" contour.

Fig.5 Verity Sarastro, anechoic response on tweeter axis at 50", averaged across 30 degrees horizontal window and corrected for microphone response, with the complex sum of the nearfield responses, taking into account acoustic phase and distance from the nominal farfield point (black), along with the complex sums of the nearfield head-unit responses (red) and nearfield bass-unit responses (blue).

The bass peak is apparently due to the coincidence of the woofer's bass tuning with that of the midrange unit. This is illustrated by the colored traces in fig.5: the blue trace is the complex sum of the woofer and its port, the red trace the sum of the midrange and its port. Sarastro owners who cannot tame the speaker's midbass in their rooms might well experiment with blocking the ports of the head units with foam to give a sealed-box alignment.

Higher in frequency, the Sarastro's response can be described as either an upper-midrange peak combined with an elevated top octave and a half, or a flat response marred by a lack of presence-region energy. Which of these the listener will perceive will very much depend on the music played. I note that MF described the speaker as having a "subjectively polite though extended top end"; I suspect that part of the reason for this is that his ear was locking on to the relative lack of presence-region energy.

But a speaker's perceived balance is also affected by its dispersion, and the Sarastro suffers from a diminished presence region in the horizontal plane as well (fig.6), which adds to the on-axis lack. This behavior will give a speaker a polite character, and a listener the feeling that it never quite plays loudly enough to be fully satisfying—both of which MF commented on. Other than that region, the Sarastro's horizontal radiation pattern is commendably wide and even.

Fig.6 Verity Sarastro, lateral response family at 50", normalized to response on tweeter axis, from back to front: differences in response 90 degrees–5 degrees off-axis, reference response, differences in response 5 degrees–90 degrees off-axis.

I was puzzled by MF's finding the top octave to sound depressed; nothing so far in the speaker's measured behavior correlates with this. However, the vertical dispersion (fig.7) reveals that that short ribbon tweeter is quite directional in the high treble; this, coupled with the fact that the ribbon axis is a high 44" from the floor, and thus fires above a seated listener's head, will result in less top-octave energy in the speaker's balance than is suggested by the on-axis response.

Fig.7 Verity Sarastro, vertical response family at 50", normalized to response on tweeter axis, from back to front: differences in response 45 degrees–5 degrees above axis, reference response, differences in response 5 degrees–45 degrees below axis.

In the time domain, the Sarastro's step response (fig.8) suggests that all three drive-units are connected in the same positive acoustic polarity but that the speaker is not time-coherent. Some high-frequency ringing is evident in the impulse response (not shown) and results in a ridge of delayed energy at 6.7kHz in the speaker's farfield waterfall plot (fig.9). There is also something going on at 1kHz, but this may well be an interference effect rather than a resonance per se. Other than these two modes, the waterfall plot is pretty clean, which may well correlate with Mikey's enthusiastic reception of the speaker's presentation of recorded detail.

Fig.8 Verity Sarastro head unit, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).

Fig.9 Verity Sarastro head unit, cumulative spectral-decay plot at 50" (0.15ms risetime).

The Verity Sarastro's idiosyncratic measured performance suggests, to me at least, that the designer carefully arrived at the speaker's balance by listening rather than go for a clean measured performance and then fine-tune the speaker by ear. As MF found, the Sarastro can produce a fine, if polite, in-room balance, but the speaker's bass will always sound a little excessive except in very large rooms (though the speaker will then not get the necessary boundary reinforcement to extend its bass). Even more than usually, a home audition will be essential with the Sarastro before a final purchase decision is made.—John Atkinson

Verity Audio
US distributor: Audiophile Systems
8709 Castle Park Drive
Indianapolis, IN 46256
(888) 272-2658