Mission Pilastro loudspeaker Measurements

Sidebar 3: Measurements

The big Mission was very much more sensitive than the average loudspeaker, at an estimated 93.7dB(B)/2.83V/m. However, its plot of impedance magnitude and electrical phase (fig.1) indicate that it is really a 4 ohm design over much of the audioband. The Pilastro's minimum impedance of 2.9 ohms at 70Hz and a phase angle that varies between ±40 degrees mean that a good high-current amplifier is advised. The impedance traces are free from the wrinkles and discontinuities that would otherwise reveal the presence of panel resonances, and examining the cabinet's vibrational behavior with an accelerometer revealed nothing worth showing in a graph. The saddle at 30Hz in the magnitude trace indicates the tuning frequency of the six passive radiators, which in turn implies good bass extension.

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

The individual responses of the drive-units are shown in fig.2. The output of the passive radiators is the trace on the left that peaks at 32Hz, the frequency of the notch in the woofers' response. However, the ABRs also peak a little between 65Hz and 80Hz, which is where the woofers also have their maximum output. This might be real, but it might also be due to leakage from the woofers into the measured ABR response. Both woofers and ABRs show a slight peak around 130Hz, but this is above the units' passbands. The woofers appear to roll out with a second-order 12dB/octave acoustic low-pass slope, but the twin midrange units roll in with what, initially at least, is a third-order, 18dB/octave slope.

Fig.2 Mission Pilastro, acoustic crossover on tweeter axis at 50", corrected for microphone response, with the nearfield midrange, woofer, and ABR responses plotted below 400Hz, 900Hz, and 600Hz, respectively.

The upper crossover lies at around 2.2kHz, as specified, with the midrange units rolling off sharply above this frequency. However, the tweeter's output features a strong notch just above the crossover, and the tweeter appears to be balanced a couple of dB too high in level. Usually, it is the tweeter that limits a speaker's ultimate sensitivity; in this case, the tweeter is actually held back by the lower-frequency drivers. On its axis, the HF unit's output is maintained to above 20kHz and shows signs of returning to its full level at 30kHz, the current upper limit of my response measurements.

Fig.3 shows both the complex sum of the lower-frequency nearfield measurements and the Mission's farfield response above 300Hz, averaged across a 30 degrees horizontal window centered on the tweeter axis. The peak between 65Hz and 75Hz will partly be due to the nearfield measurement technique, and it might also be a measurement artifact due to the leakage from the woofers into the measured ABR response mentioned earlier. Without the peak, the Pilastro's low frequencies usefully extend down to 30Hz, though there is then a fairly sharp rolloff.

Fig.3 Mission Pilastro, 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 plotted below 300Hz.

The midrange in fig.3 is impressively flat, as is the treble, though the 2dB-high plateau seen in fig.2, covering the tweeter's passband, also makes an appearance in this graph. In addition, the notch at 2.7kHz in the tweeter's output leads to a slight lack of measured integration between the upper-frequency drive-units. However, the notch is both narrow and shallow enough that it should be inaudible. The top-octave response is maintained to 20kHz, though shelved down a couple of dB. However, the output above 20kHz in this spatially averaged response rolls off significantly. This is due to the tweeter's ultrasonic output being very directional, owing to the radiating diameter being larger than the wavelength in this frequency region. Fig.4 reveals that the grille slightly emphasizes the low treble, pulls down the octave between 5kHz and 10kHz, and introduces peaks in the top octave and above. (All the listening and measurements were performed without the grilles in place.)

Fig.4 Mission Pilastro, effect on tweeter-axis response of adding grille (2dB/vertical div.).

Because of the Pilastro's bulk and mass, I couldn't place it on my computer-controlled speaker turntable for the measurements. I had to examine the speaker's lateral off-axis behavior by rotating it by hand, using a protractor for guidance. I plotted the response at 5 degrees intervals out to only ±15 degrees to generate the average shown in fig.3. Beyond that angle, because of the error inherent in the manual rotation method, I measured the off-axis response at 15 degrees steps.

The Pilastro's lateral dispersion, normalized to the tweeter-axis response, is shown in fig.5. The indicated peak at 2.7kHz in the off-axis traces indicates that the on-axis notch in the tweeter's output does fill in to the speaker's sides. Despite the large radiating diameter of the midrange cones, there is no sign of beaming at the top of their passband. The even spacing of this graph's contour lines through the midrange to the high treble correlates with the stable, precise stereo imaging I heard. But, as noted earlier, the ring-radiator tweeter does become very directional above 20kHz.

Fig.5 Mission Pilastro, 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.

This ultrasonic rolloff is also evident in the vertical plane (fig.6), but it can also be seen that the Pilastro is remarkably uncritical with respect to listening height. Pretty much the same response will be heard by listeners with ear heights ranging from 36" to 48". For standing listeners, however, a severe notch develops at the upper crossover frequency.

Fig.6 Mission Pilastro, vertical response family at 50", normalized to response on tweeter axis, from back to front: differences in response 15 degrees-5 degrees above axis, reference response, differences in response 5 degrees-10 degrees below axis.

The spatially averaged response taken at the listening position in my dedicated listening room (fig.7) reveals a remarkably even delivery of energy into the room from the lowest bass through the upper midrange. Below 400Hz or so, the small peaks and dips will be due to residual room modes, which have not been completely eliminated by the speaker positions and the spatial averaging. However, as indicated in the quasi-anechoic response measurements, the region covered by the tweeter is indeed a couple of dB too hot in-room, which correlates with the lively, forward high-frequency balance I noted in my auditioning.

Fig.7 Mission Pilastro, spatially averaged, 1/3-octave, freefield response in JA's listening room.

Again because of the Pilastro's bulk, I could not lift it off the ground for the acoustic measurements. As a result, a strong reflection from the ground can be seen at the 7.5ms mark in the speaker's step response (fig.8). But this graph still shows that the outputs of the tweeter and midrange units are all in positive acoustic polarity, the former leading the latter by 400µs or so and its overshoot neatly splicing into the midranges' slower rise away from the time axis. The individual step responses of the woofers and midrange units (not shown) reveal that the woofers are connected in inverted polarity. This combines with the acoustic phase shift associated with the crossover filter to give good integration between the lower-frequency drivers.

Fig.8 Mission Pilastro, on-axis step response at 50" (5ms time window, 30kHz bandwidth).

I windowed out the ground reflection seen in fig.8 to calculate the Pilastro's cumulative spectral-decay plot on the tweeter axis (fig.9), which results in the dotted area, indicating invalid data, in the graph. Nevertheless, enough valid time data are available to show that the Mission tweeter has a very clean decay. While the black ridge around 16kHz is due to the computer monitor, the ridge of delayed energy at 8kHz is real enough, and is probably associated with the peak at the same frequency in the midrange units' output. However, this is low enough in level that it shouldn't have audible consequences. Lower in frequency, the on-axis notch at 2.7kHz is associated with some delayed energy, which suggests that it is some sort of diffraction or interference effect.

Fig.9 Mission Pilastro, cumulative spectral-decay plot at 50" (0.15ms risetime).

Too often for my taste, very expensive loudspeakers have puzzlingly poor measured performance. That this is not the case with Mission's Pilastro is reassuring, given the enthusiasm I felt about its sound quality.—John Atkinson

US distributor: Denon Canada Inc.
505 Apple Creek Blvd., Unit 5
Markham, Ontario L3R 5B1, Canada
(905) 475-4085