Floor Loudspeaker Reviews

I performed the quasi-anechoic measurements of the Canton Reference 3.2 DC with DRA Labs' MLSSA system and a calibrated DPA 4006 microphone. I measured the speaker without its grille, as this was how Kal Rubinson had found it to sound best. My estimate of the Canton's voltage sensitivity was a high 91dB/2.83V/m; this loudspeaker will play loudly with just a few watts. However, its impedance plot (fig.1) indicates that the partnering amplifier needs to be comfortable driving a low impedance. The magnitude remains below 4 ohms for almost the entire bass and midrange regions, with a minimum value of 3 ohms at 360Hz and a current-demanding combination of 4.6 ohms and –45° electrical phase angle at 88Hz, a frequency where music can have high energy levels.

Fig.1 Canton Reference 3.2 DC, electrical impedance (solid) and phase (dashed) (2 ohms/vertical div.).

There is a glitch in both the impedance-magnitude and phase traces at 155Hz; an investigation of the enclosure panels' vibrational behavior found that while most of the surfaces were free from resonant modes, there was a fairly strong cabinet resonance on the sidewalls 12" from the bottom (fig.2). It is possible that because the radiating area of the affected region is small, this resonance will not color the Reference 3.2's sound. There was also a strong vibrational mode at 703Hz on the top panel (not shown); again, because the affected area is small and because, in this instance, the frequency is above the critical region, this behavior will not color the sound.

Fig.2 Canton Reference 3.2 DC, cumulative spectral-decay plot calculated from output of accelerometer fastened to center of side panel 12" from the bottom (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz)

The plot of the Canton's impedance magnitude (fig.1) doesn't resemble that of a typical port-loaded design, which has two peaks of approximately equal height in the bass. From the shape of the Canton's curve below 60Hz, however, it looks as if the downward-firing port is tuned to a frequency between 30 and 40Hz. The two woofers have identical acoustic outputs, and the sum of their nearfield responses (fig.3, blue trace) has the expected minimum-motion notch at 31Hz. (This is the frequency where the back-pressure from the port resonance prevents the woofer cones from moving.) Above that frequency, the woofers cover the region between 50 and 250Hz; below it, they roll out at a steep 24dB/octave rate rather than the usual 12dB/octave. The same is true for the port's output (fig.3, red trace)—in earlier Canton speakers we have reviewed, there was a large series capacitor in the woofer feed to increase the rate of rolloff below the port resonance; it looks as if the Reference 3.2 DC uses a second-order high-pass filter to drive the woofer, to prevent subsonic spuriae from restricting its dynamic range. The port rolls off smoothly above 40Hz, with just a slight return evident between 300 and 600Hz.

Fig.3 Canton Reference 3.2 DC, acoustic crossover on tweeter axis at 50", corrected for microphone response, with nearfield responses of port (red), woofer (blue), and midrange unit plotted below 1kHz, 350Hz, and 500Hz, respectively.

Higher in frequency in fig.3, the midrange unit rolls in smoothly above 250Hz, with it and the tweeter offering a superbly flat response up to 20kHz. This can also be seen in fig.4, which shows the Reference 3.2's farfield response averaged across a 30° horizontal window on the tweeter axis. Rather than the usual high-Q (very narrow) peak above the audioband, there is a very sharp notch at 22kHz, which appears from the impedance plot to be the frequency of the metal-dome tweeter's primary resonance. Most of the apparent boost in the upper bass will be due to the nearfield measurement technique used to generate the trace below 300Hz, but there is useful output down to the port tuning frequency. The speaker rolls off very sharply below 25Hz, however, due both to the fact that the outputs of the woofers and the port are in anti-phase below the port resonance, and to the presence of the additional high-pass filter.

Fig.4 Canton Reference 3.2 DC, anechoic response on tweeter axis at 50", averaged across 30° horizontal window and corrected for microphone response, with complex sum of midrange, woofer, and port nearfield responses plotted below 300Hz.

Up to 5kHz, the Canton Reference 3.2 DC offered textbook lateral dispersion (fig.5), with smoothly controlled and even off-axis behavior. The speaker became a little more directional in the top two octaves than is usual for a tweeter with a 1" diaphragm, with then an off-axis flare around 15kHz and an off-axis notch between 17 and 20kHz. The sharply defined suckout at 22kHz in the Canton's on-axis output appears to fill in somewhat to the speaker's sides, as it does at extreme off-axis angles in the vertical plane (fig.6). However, this graph shows that the Reference 3.2 is not fussy about the height of the listener's ears, the response not changing significantly over a fairly wide window.

Fig.5 Canton Reference 3.2 DC, 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.6 Canton Reference 3.2 DC, vertical response family at 50", normalized to response on tweeter axis, from back to front: differences in response 20–5° above axis, reference response, differences in response 5–10° below axis.

In the time domain, the Canton's step response (fig.7) suggests that its tweeter and midrange unit are connected in inverted acoustic polarity, its woofers in positive polarity. However, the tail of each unit's step smoothly blends with the start of the step of the next lower in frequency, which correlates with the superb frequency-domain integration of their outputs seen in fig.4. The Reference 3.2's cumulative spectral-decay plot (fig.8) is generally clean, if not quite in the league of the best speakers I have measured.

Fig.7 Canton Reference 3.2 DC, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).

Fig.8 Canton Reference 3.2 DC, cumulative spectral-decay plot on tweeter axis at 50" (0.15ms risetime).

Overall, the Canton Reference 3.2 DC offers excellent measured performance, as I have come to expect from the company's chief engineer, Frank Göbl.—John Atkinson