Infinity Modulus loudspeaker & Modulus subwoofer Measurements
I use a mixture of nearfield, in-room, and quasi-anechoic FFT measurement techniques (using primarily DRA Labs' MLSSA system with a B&K 4006 microphone, but also an Audio Control Industrial SA-3050A 1/3-octave spectrum analyzer with its calibrated microphone) to investigate objective factors that might explain the sound heard. The speakers' nearfield low-frequency responses and impedance phase and amplitude were measured using the magazine's Audio Precision System One.
There are no surprises in the Infinity's measured impedance (fig.1). The sealed-box tuning and the lightweight balance are shown by the peak in the bass, reaching a maximum of 26.8 ohms at 84Hz, while the peak centered on 2kHz is due to the crossover. Although its impedance drops to between 4 and 5 ohms in the low bass and lower midrange, the Modulus shouldn't be too hard a load for a good amplifier to drive, particularly as, in common with nearly every speaker I've ever measured, the maximum phase angle does not occur when the magnitude is at a minimum. Above 8kHz, the tweeter presents pretty much a 6 ohm resistive load with its control set to 12 o'clock. The Modulus's estimated sensitivity was around 85dB/W/m at 1kHz.
Fig.1 Infinity Modulus, electrical impedance (solid) and phase (dashed) with treble control at 12:00 (2 ohms/vertical div.)
The Modulus's impulse response 48" away (footnote 1) on the HF axis is shown in fig.2 and the step response in fig.3. (The grille was left off for these and all the other measurements, except where noted.) Again, there are no surprises, though I would have expected slightly more overhang from the high-order crossover.
Fig.2 Infinity Modulus, impulse response on tweeter axis at 48". (5ms time window, 30kHz bandwidth.)
Fig.3 Infinity Modulus, step response on tweeter axis at 48". (5ms time window, 30kHz bandwidth.)
Fig.4 shows how the speaker's frequency response changes as that impulse decays: pretty clean, the "waterfall" is only marred by two slight areas of hash, around 2kHz and between 5kHz and 10kHz, which might correlate with the slightly bright treble character noted during the auditioning. The "crinkly" appearance of the latter region seems to be typical of planar drive-units, though what it represents I'm not sure. The step responses of the individual units show that they're connected with the same polarity, and the effectiveness of the stepped baffle is shown by the fact that the output from both units is coincident in time on the tweeter axis. The woofer seems very clean toward the top of its passband, as can be seen from its waterfall plot (fig.5), though again a mild mode is visible at 2kHz or so, this possibly tying in with the treble congestion heard at high levels.
Fig.4 Infinity Modulus, cumulative spectral-decay plot at 48" (0.15ms risetime).
Fig.5 Infinity Modulus woofer only, cumulative spectral-decay plot at 48" (0.15ms risetime).
Fig.6 shows the anechoic response of the two drive-units on the tweeter axis at a distance of 48" with the tweeter control set to 12 o'clock (as it was for all these measurements). Note the steep acoustic rollout of the tweeter below 3kHz, as well as its rising response in the top octave and a half. Looking at how the drivers' outputs integrate, fig.7 shows the anechoic response of the Modulus from front to back, 7.5° below the woofer, on the woofer axis, on the tweeter axis, and 7.5° above the tweeter, all, again, at a distance of 48". The responses in front of the loudspeaker are equally smooth, but those above and below the Modulus feature a significant notch at crossover. The vertical listening window will therefore be reasonably critical, and it is possible that the room reverberant field may also lack energy in the presence region.
Fig.6 Infinity Modulus, acoustic crossover on tweeter axis at 48", with nearfield woofer response shown below 300Hz.
Fig.7 Infinity Modulus, vertical response family at 48", normalized to response on tweeter axis, from back to front: difference in response 7.5° above axis, reference response, differences in response 7.5–15° below axis.
I looked at the effect of the grille by repeating the HF-axis response in fig.7 with it on. Although the resultant response is not shown, the grille appeared to reinforce the depth of the on-axis crossover suckout, something that I had suspected from my auditioning. The tweeter-level control increases or reduces the drive-unit's output by +2 or –2dB above 5kHz compared with the level with it set to 12 o'clock.
To examine the Modulus's anechoic response in a manner that minimizes interference-type features that will be too position-dependent to have any serious subjective effect, I average five responses across a ±15° lateral window. This is shown on the righthand side of fig.8. Notably smooth overall, it is broken only by a minimal notch at crossover and the rising trend in the high treble seen in fig.6. The lefthand side of fig.8 shows the nearfield response of both the Modulus woofer, plotted below 180Hz, and that of the servo-controlled Modulus Subwoofer. The two plots in the low bass show the difference in nearfield bass extension when the control unit is set to "22Hz" or to "35Hz." Note the restricted low-frequency extension of the Modulus when used on its own, –6dB at 74Hz referred to the level at 200Hz, though the sealed-box alignment does roll out at a relatively gentle 12dB/octave.
Fig.8 Infinity Modulus, anechoic response on tweeter axis at 50", averaged across 30° horizontal window and corrected for microphone response, with the nearfield response of the woofer below 180Hz, and the nearfield response of the Modulus subwoofer with the low-pass filter set to 22Hz and 35Hz and crossver set to 210Hz.
In-room and spatially averaged to remove the effect of low-frequency room resonances (fig.9), these quasi-anechoic responses translate to a broadly similar balance: restricted bass extension, a superbly smooth midrange, a definite lack of energy around crossover (correlating with the lack of immediacy to the speaker's sound), and a little too much energy in the top two octaves with the tweeter control at 12 o'clock.
Fig.9 Infinity Modulus, spatially averaged, 1/3-octave response in JA's listening room, tone control set to 12:00, no subwoofer.
Finally, I examined the characteristics of the electronic control unit by measuring the subwoofer's nearfield response with the crossover frequency set to between 210Hz and 40Hz with the extension set to 22Hz (fig.10). This measured subwoofer performance, together with the continuous phase and level adjustment, will allow the maximum flexibility in integrating the subwoofer's output with that of both the Infinity satellites and loudspeakers from other manufacturers.—John Atkinson
Fig.10 Infinity Modulus subwoofer, nearfield response with low-pass filter set to 210Hz, 175Hz, 120Hz, 90Hz, 70Hz, 60Hz, and 50Hz.
Footnote 1: I've been asked why I use this "nonstandard" distance for MLSSA measurements, rather than 1m or 2m. The answer is that there are two mutually exclusive conditions at work with these quasi-anechoic measurements. Measuring at a 2m distance gives the optimum correlation with how the outputs from each drive-unit of a multiway design actually integrate at the listening position. However, the frequency resolution of FFT and TDS-type measurement systems is best when the microphone is very close to the loudspeaker because then you can get the largest time window between the initial arrival of the impulse and the first reflection from an adjacent boundary. (The anechoic frequency resolution is then proportional to that time window.) In a typical room with a 9–10' ceiling, a 48" distance gives a good compromise between the two needs when the loudspeaker is placed with its drive-units roughly midway between ceiling and floor, giving an anechoic time window of 4ms or so (which is why my cumulative spectral decay or waterfall plots also feature a time axis of some 4ms). Ideally, however, I would like a room with at least a 20' ceiling. Maybe I can persuade LA to excavate a cubical room, 20' to a side, in Stereophile's parking lot.—John Atkinson