Sidebar 3: Measurements
I used DRA Labs' MLSSA system and a calibrated DPA 4006 microphone to measure the Alta Hestia Titanium's frequency response in the farfield, and an Earthworks QTC-40 for the nearfield responses. (I measured S/N HST02706.) The Alta's voltage sensitivity is specified as a higher-than-average 90dB/2.83V/m; my estimate was much lower, at 82.5dB(B)/2.83V/m. In part, this will be due to the fact that my figure is based only on the speaker's frontal radiation, whereas the Hestia Titanium, being a dipole design in the midrange, will also be putting out almost as much energy behind it as in front of it. But this is still a low sensitivity, and will be exacerbated by the speaker's impedance magnitude, which drops to just 1.75 ohms at 264Hz (fig.1). There are also demanding combinations of magnitude and electrical phase at 50Hz (4.3 ohms and –46°) and 450Hz (2.9 ohms and +49°). Though its impedance remains above 6 ohms in the treble, the Hestia Titanium will need to be driven by a powerful amplifier if it is not to be starved of current.
But it was when I examined the nearfield response of the front-panel port that loads the woofer that I found disturbing behavior that correlated with the discontinuities in the impedance plots. The port's output (fig.3, blue trace) covers a wide passband, from 15 to 85Hz, but then has a sharply defined peak visible at 100Hz, with peaks lower in level at 200 and 400Hz. The two lower-frequency resonant modes are sufficiently severe to disturb the woofer's nearfield output (red trace). When Jim Austin writes, "While bass instruments had great impact through the Hestias, sometimes they didn't sound precisely as a conscientious sound engineer might want them to sound," I believe he is referring to the effect of this behavior.
Higher in frequency in fig.3, the sum of the midrange units' outputs—all three seem to have an identical frequency response—is shown as the green trace. This splices the sum of the nearfield outputs to the farfield response at 450Hz. The peak between 100 and 500Hz appears to be real and not an artifact of the nearfield measurement technique, as it was also evident in wide-range farfield measurements. The upper-midrange output is flat, but then there is a 5dB step down in output before the midrange units hand off to the ribbon tweeter. I suspect that this lack of presence-region energy affected my estimate of the Hestia Titanium's sensitivity.
Fig.4 shows how these individual responses sum in a 30° horizontal window centered on the tweeter axis. As recording engineer Phil Schaap said when he heard the Altas in Jim Austin's room, "These speakers have too much bass." I admit that the rise in low-frequency output in this graph will be due in part to the nearfield measurement technique, which assumes a half-space environment for the drive-units; ie, a baffle that extends to infinity in both planes. Nevertheless, the Hestia Titanium does have too much bass.
The Hestia Titanium's step response on the tweeter axis (fig.7) reveals that the tweeter and midrange units are connected in inverted acoustic polarity, the woofer in positive polarity. (I checked the woofer's polarity by momentarily connecting a 2V dry cell, positive to positive, across the Hestia's terminals; the cone moved away from the enclosure, as predicted by fig.7.) The cumulative spectral-decay or waterfall plot on the tweeter axis is shown in fig.8. The decay is impressively clean in the region covered by the ribbon tweeter, but there is a ridge of delayed energy at 1.2kHz that, all things equal, I would expect to add a small degree of nasality to the speaker's sonic signature.
Fig.1 Alta Hestia Titanium, electrical impedance (solid) and phase (dashed) (5 ohms/vertical div.).
There is a major discontinuity at 100Hz in the impedance traces, as well as minor ones at 200 and 400Hz. This behavior suggests that the enclosure has resonance problems, and indeed, when I investigated the behavior of the woofer bin with a plastic-tape accelerometer, I found a high-level mode at 203Hz on the rear panel (fig.2), and a lower-level mode at 98Hz on both sidewalls.
Fig.2 Alta Hestia Titanium, cumulative spectral-decay plot calculated from output of accelerometer fastened to rear of bass bin (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz).
Fig.3 Alta Hestia Titanium, acoustic crossover on tweeter axis at 50", corrected for microphone response, with nearfield responses of midrange units (green), woofer (red), and port (blue), respectively plotted in the ratios of the square roots of their radiating areas below 350Hz, 420Hz, and 1600Hz.
Fig.4 Alta Hestia Titanium, anechoic response on tweeter axis at 50", averaged across 30° horizontal window and corrected for microphone response, with complex sum of nearfield midrange, woofer, and port responses plotted below 300Hz.
It also looks as if it has too much lower-midrange output, at least on axis, but the fact that the speaker behaves as a dipole above the woofer's passband needs to be taken into account. The Hestia Titanium's lateral dispersion, normalized to the response on the tweeter axis, is shown in fig.5. You can see from this graph that the Alta's response rolls off rapidly to the sides between 300 and 700Hz, which in the listening room will tend to ameliorate the excess of energy in the same region shown in fig.4 and result in a more neutral midrange output. Higher in frequency, the contour lines in fig.5 are even and uniformly spaced, something that tends to correlate with stable, accurate stereo imaging. The tweeter does become relatively directional above 10kHz, which might make the speaker sound too sweet in large or well-damped rooms.
Fig.5 Alta Hestia Titanium, 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.
The vertical dispersion is shown in fig.6, again normalized to the response on the tweeter axis, which is a high 43" above the floor. The speaker's balance doesn't appear to be as fussy about listening axis as JCA's listening suggested, but a suckout does develop at 1.6kHz more than 10° below the tweeter axis; I think that is what he was compensating for when he tilted the speakers forward for his auditioning.
Fig.6 Alta Hestia Titanium, 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 5–15° below axis.
Fig.7 Alta Hestia Titanium, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).
Fig.8 Alta Hestia Titanium, cumulative spectral-decay plot on tweeter axis at 50" (0.15ms risetime).
There is a lot to admire in the measured performance of Alta Audio's Hestia Titanium, and it looks as if designer Michael Levy has worked hard to balance the omnidirectional radiation pattern in the bass with the dipolar behavior in the midrange and the tweeter's forward-firing pattern. The excessive low frequencies can be adjusted to some extent with careful placement, but I remain bothered by the resonances I found in the woofer's and port's outputs, and that delayed energy at the top of the midrange units' passband.—John Atkinson
