Sidebar 3: Measurements
I used DRA Labs' MLSSA system and a calibrated DPA 4006 microphone to measure the Canton Reference 7K's frequency response in the farfield. My reference axis for the measurements was level with the center of the midrange unit, which is 36" from the floor, the height of my ears in my listening chair. I used an Earthworks QTC-40 mike for the nearfield and in-room responses and Dayton Audio's DATS V2 system to measure the impedance.
Canton specifies the Reference 7K's sensitivity as 88.5dB, presumably for 2.83V at 1m, which was confirmed by my measurement. The Reference 7K's impedance is specified as 4–8 ohms. The solid trace in fig.1 indicates that the impedance magnitude lies between 4 and 9.5 ohms over most of the audioband, with minimum values of 3.2 ohms at 120Hz and 3 ohms at 305Hz. The electrical phase angle (dashed trace) is occasionally high, which means that the EPDR (footnote 1) drops below 2 ohms between 16Hz and 29Hz, between 83Hz and 123Hz, and between 395Hz and 745Hz, with minimum values of 1.7 ohms at 36Hz, 1.4 ohms at 103Hz, and 1.8 ohms between 480Hz and 600Hz. The Reference 7K must be used with amplifiers that don't have problems driving 4 ohms.
The black trace below 300Hz in fig.4 is the complex sum of the nearfield woofer and port responses, with the latter's acoustic phase compensated for the fact that it is on the loudspeaker's base. The peak in the upper bass, which is due to the nearfield measurement technique, is a little greater in amplitude than I usually see. The low frequencies roll off rapidly below 30Hz. The Canton's farfield response, averaged across a 30° horizontal window centered on the midrange axis (fig.4, black trace above 300Hz), is impressively even between 400Hz and 8kHz, with very slight dips balanced by equally small excesses of energy. The tweeter's on-axis response starts to rise above 8kHz then drops off quickly above 20kHz. Figs.3 and 4 were taken without the vestigial grilles; repeating the measurement with the grilles reduces the level between 4kHz and 10kHz by up to 3dB.
Footnote 1: EPDR is the resistive load that gives rise to the same peak dissipation in an amplifier's output devices as the loudspeaker. See "Audio Power Amplifiers for Loudspeaker Loads," JAES, Vol.42 No.9, September 1994, and stereophile.com/reference/707heavy/index.html. Footnote 2: Using the FuzzMeasure 3.0 program, a Metric Halo MIO2882 FireWire-connected audio interface, and a 96kHz sample rate, I average 20 1/6-octave–smoothed spectra, individually taken for the left and right speakers, in a rectangular grid 36" wide by 18" high and centered on the positions of my ears.
Fig.1 Canton Reference 7K, electrical impedance (solid) and phase (dashed) (2 ohms/vertical div.).
The traces in fig.1 are free from the small discontinuities that would imply resonances of some kind. However, when I investigated the enclosure's vibrational behavior with a plastic-tape accelerometer, I did find a single resonant mode at 168Hz on all the surfaces (fig.2). Because this mode has a relatively low level and a high Q (Quality Factor), I doubt that it will result in audible coloration. (The lower the Q of a resonance the greater the possibility that it will be excited.)
Fig.2 Canton Reference 7K, cumulative spectral-decay plot calculated from output of accelerometer fastened to center of side panel level with the midrange unit (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz).
The saddle centered on 42.7Hz in the impedance magnitude trace suggests that this is the tuning frequency of the downward-firing port. The two woofers have identical responses; the minimum-motion notch in their summed output (fig.3, blue trace), which is when the back pressure from the reflex resonance holds the woofer cones still, lies a little lower in frequency, at 35Hz. The port's nearfield output (fig.3, red trace) peaks between 25Hz and 90Hz, with a clean upper-frequency rolloff. The rise in the impedance magnitude below 25Hz in fig.1 confirms that there is a series high-pass filter in the woofer feed. The effect of this filter can be seen in fig.3, which indicates that both the woofers and the port roll off below the port tuning frequency with a much steeper slope than the usual 12dB/octave.
Fig.3 Canton Reference 7K, acoustic crossover on midrange axis at 50", corrected for microphone response, with the nearfield response of the midrange unit (green), the summed nearfield responses of the woofers (blue), and the port (red), respectively plotted below 450Hz, 400Hz, and 500Hz.
Fig.4 Canton Reference 7K, anechoic response on midrange axis at 50", averaged across 30° horizontal window and corrected for microphone response, with the sum of the nearfield responses plotted below 300Hz.
Fig.5 shows the Reference 7K's horizontal dispersion, normalized to the response on the midrange axis, which thus appears as a straight line. The contour lines in this graph are smooth and evenly spaced, this correlating with the stable stereo imaging I noted in my auditioning. While there are ridges of off-axis energy between 13kHz and 22kHz, with a gulley between them, these will be inconsequential. In the vertical plane (fig.6), a suckout in the crossover region develops 15° above and below the midrange axis, but the response doesn't change by much over a ±5° window centered on the midrange axis.
Fig.5 Canton Reference 7K, lateral response family at 50", normalized to response on midrange 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 7K, vertical response family at 50", normalized to response on midrange axis, from back to front: differences in response 15–5° above axis, reference response, differences in response 5–15° below axis.
Fig.7 shows the Canton Reference 7Ks' spatially averaged response in my room. (The spatial averaging, footnote 2, tends to average out the peaks and dips below 400Hz that are due to the room's resonant modes.) The Reference 7Ks get some help from the lowest frequency mode in my room, and the bass is well-extended. There is a little too much upper-bass energy in-room—that's the richness I heard with the lower register of the marimba—but the balance at the listening position is impressively even from 500Hz to the top of the audioband. A loudspeaker that offers a flat on-axis response and well-controlled lateral dispersion gives a gently sloped-down treble in the spatially averaged room response due to the increased absorption of the room's furnishings and the narrowing of the tweeter's radiation pattern at high frequencies. By that yardstick, the Cantons have a little too much output in the top two audio octaves.
Fig.7 Canton Reference 7K, spatially averaged, 1/6-octave response in JA's listening room.
Fig.8 compares the Reference 7K's spatially averaged response (red trace) with that of the Marten Oscar Duo that Michael Fremer reviewed in the November 2020 issue (blue trace). The Oscar Duo costs the same as the Reference 7K—$6995/pair—though the matching stands add $995/ pair to its price. The Canton speakers' in-room response is smoother in the treble with a little more top-octave output than the Martens'. The Oscar Duos excite the lowest frequency room mode just above 30Hz to a slightly greater extent, and the Reference 7Ks roll off faster below that frequency. The Marten is a two-way design; the three-way Canton, with its separate midrange unit and twin woofers, will play louder without strain.
Fig.8 Canton Reference 7K, spatially averaged, 1/6-octave response in JA's listening room (red) and of the Marten Oscar Duo (blue).
Turning to the time domain, the Reference 7K's step response on the midrange axis (fig.9) reveals that the tweeter and midrange unit are connected in positive acoustic polarity, the woofers in inverted polarity. (I confirmed this by looking at the step responses of the individual drive-units.) The decay of the tweeter's step smoothly blends with the positive-going start of the midrange unit's step, and the decay of the midrange unit's step smoothly blends with the negative-going start of the woofers' step. This implies optimal crossover design. The Reference 7K's cumulative spectral-decay plot (fig.10) is superbly clean.
Fig.9 Canton Reference 7K, step response on midrange axis at 50" (5ms time window, 30kHz bandwidth).
Fig.10 Canton Reference 7K, cumulative spectral-decay plot on tweeter axis at 50" (0.15ms risetime).
The Canton Reference 7K's excellent measured performance belies its relatively affordable price.—John Atkinson
Footnote 1: EPDR is the resistive load that gives rise to the same peak dissipation in an amplifier's output devices as the loudspeaker. See "Audio Power Amplifiers for Loudspeaker Loads," JAES, Vol.42 No.9, September 1994, and stereophile.com/reference/707heavy/index.html. Footnote 2: Using the FuzzMeasure 3.0 program, a Metric Halo MIO2882 FireWire-connected audio interface, and a 96kHz sample rate, I average 20 1/6-octave–smoothed spectra, individually taken for the left and right speakers, in a rectangular grid 36" wide by 18" high and centered on the positions of my ears.
