Canton Karat Reference 2 DC loudspeaker Measurements

Sidebar 3: Measurements

The big Canton was significantly more sensitive than average, at an estimated 92dB(B)/2.83V/m. However, as shown by its plot of impedance magnitude and electrical phase against frequency (fig.1), it actually draws more like 2W from the amplifier to achieve this rating. With a true 1W drive, its sensitivity will be exactly to specification at 89dB/W/m.

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

The impedance stays within tight 3 and 6 ohm limits from 70Hz to 40kHz, meaning that there will be only small modification of its response when used with a tube amplifier having a higher-than-average source impedance. However, the phase angle does become large in the midbass, with a combination of 5 ohms magnitude and -45 degrees phase, implying that the partnering amplifier should have good current delivery. The rise in impedance below 20Hz suggests the presence of large capacitors in series with the woofer which limit excursion when it is unloaded at subsonic frequencies.

Two discontinuities are apparent in the fig.1 magnitude trace. The one above the audioband, at 22kHz, is due to the ubiquitous metal-dome tweeter resonance and will be benign. However, the one between 100 and 200Hz suggests the presence of one or more vibrational resonances in the 2 DC's enclosure. Fig.2, a cumulative spectral-decay plot calculated from the output of a simple PVDF accelerometer fastened to the back panel level with the woofer and ABR, indicates two quite-well-defined resonances—the higher-frequency mode could be detected on all surfaces. All things being equal, I would have thought these low in enough in level not to mess with the speaker's sound quality. But I could hear them with the Test CD 3 warble tones, and I suspect they added to the extra warmth heard in Joni Mitchell's voice and in the double bass.

Fig.2 Canton Karat Reference 2 DC, cumulative spectral-decay plot calculated from the output of an accelerometer fastened to the cabinet's back panel above terminals. (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz.)

The saddle at 35Hz in the fig.1 impedance magnitude trace suggests that this is the tuning frequency of the big ABR. However, the minimum-motion point in the woofer's nearfield output (fig.3, blue trace) lies a little lower than this, at 30Hz, while the ABR's output (green trace) peaks a little higher in frequency, at 40Hz. Note that both the woofer and ABR roll out below resonance with twice the usual reflex slope—24dB/octave rather than 12dB/octave—due to Canton's Displacement Control technology. This results in a very steep rolloff for the combined output of the units (black trace). The crossover between the woofer (blue) and midrange units (red) appears to lie at 150Hz, as specified, but with asymmetrical filter slopes. The midranges roll off at 24dB/octave below 100Hz, the woofer at 12dB/octave above the same frequency.

Fig.3 Canton Karat Reference 2 DC, nearfield midrange (red trace), woofer (blue), and ABR (green) responses plotted below 1kHz, 1kHz, and 800Hz, respectively, with their complex sum (black) plotted below 1kHz.

Fig.4 splices the black low-frequency trace from fig.3 to the Karat Reference 2 DC's farfield response, averaged across a 30 degrees horizontal window on the tweeter axis. There is a notable absence of peaks and dips in the response. While the upper bass is humped up by 5dB or so, this will be due at least in part to the fact that this is a nearfield measurement, which assumes a 2pi (hemispherical) environment. The tweeter seems to be balanced a couple of dB hot, but as the top octaves only occasionally sounded lifted up, I suspect that the listener identifies the relative lack of energy though the upper midrange as "politeness" rather than the treble as being excessive.

Fig.4 Canton Karat Reference 2 DC, 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 Canton's midrange units get a little directional at the top of their passband (fig.5), but the corresponding flare at the bottom of the tweeter's passband doesn't extend as high in frequency as is usually seen, the driver also starting to get directional above 5kHz. All things being equal, this might tend to work against the on-axis energy excess in rooms with typically absorptive furnishing. The radiation pattern in the vertical plane (fig.6) reveals that the listener's ears need to be at a level between the centers of the two midrange units—33"-41" if the low treble is not to sound recessed. This vertical-plane behavior contributes to the in-room balance (fig.7), meaning that it pretty much corresponds to the on-axis response, with the low treble somewhat lower in level than the upper octaves.

Fig.5 Canton Karat Reference 2 DC, 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.

Fig.6 Canton Karat Reference 2 DC, vertical response family at 50", normalized to response on tweeter axis, from back to front: differences in response 20 degrees-5 degrees above axis, reference response, differences in response 5 degrees-10 degrees below axis.

Fig.7 Canton Karat Reference 2 DC, spatially averaged, 1/3-octave, freefield response in JA's listening room.

In the time domain, the step response (fig.8) indicates that the tweeter and midrange units are connected in inverted acoustic polarity, while the woofer is connected in positive polarity. Because of the speaker's mass, I couldn't lift it off the ground for the measurements, which means that an early reflection from the floor can be seen at the far right of this graph. I windowed this reflection out before performing any FFT-derived responses, and also to generate the farfield cumulative spectral-decay plot (fig.9). This is overall quite smooth and clean, though a very slight amount of delayed energy can be seen at 5.5kHz, the frequency of a small step in the on-axis response.

Fig.8 Canton Karat Reference 2 DC, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).

Fig.9 Canton Karat Reference 2 DC, cumulative spectral-decay plot at 50" (0.15ms risetime).

These measurements indicate a well-engineered design that, while not being completely neutral, will offer excellent sound in all but small, lively rooms.—John Atkinson

1723 Adams Street NE
Minneapolis, MN 55413
(612) 706-9250