Floor Loudspeaker Reviews

Because of the extended spell of sub-zero temperatures in New York in late January, when this issue was being prepared, I was unable to measure the Linn Akurate 242 outdoors on a high stand. This, my usual practice, effectively eliminates the interfering effects of room boundaries on the measured response. I therefore had to measure the speaker indoors, which reduced the anechoic time window of my MLSSA measurements from 6ms to around 2.5ms. I might well have missed response problems in the midrange, therefore. With that qualification, here are my measured findings.

Assessed on the supertweeter axis, which is a low 33" from the floor with the speaker sitting on its metal base, the Akurate 242's voltage sensitivity was very slightly below average, although within experimental error of the specified figure, at an estimated 86.5dB(B)/2.83V/m. Its impedance plot (fig.1) also suggests that the speaker will be a somewhat demanding load, with a magnitude that remains between 3 and 6 ohms over almost the entire audioband, and a tricky combination of 5 ohms and–47 degrees phase angle at 90Hz. The generally higher impedance in the bass will boost this region with amplifiers having a higher-than-normal output impedance; eg, tube designs. I suspect that this might be partly why AD found the Lamm amplifiers to sound too dark with the Akurates. The saddle at approximately 35Hz between the twin impedance peaks in the bass suggests that the Akurate's twin ports are tuned to this frequency, which in turn implies good low-frequency extension.

Fig.1 Linn Akurate 242, electrical impedance (solid) and phase (dashed). (2 ohms/vertical div.)

The Linn's impedance curves are free in the midrange from the wrinkles and glitches that would indicate the presence of cabinet resonances. Investigating the vibrational behavior of the Akurate's enclosure walls with an accelerometer and a stethoscope did indeed indicate that it is superbly braced and damped. The only significant resonant mode I found, at 492Hz (fig.2), is both high enough in frequency and low enough in level to have no subjective effect. (The speaker was sitting on its metal bases and spikes for these measurements.)

Fig.2 Linn Akurate 242, cumulative spectral-decay plot calculated from the output of an accelerometer fastened to the cabinet's side panel level with the upper woofer (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz).

The Akurate's twin woofers and ports mean that its low-frequency behavior is complex. The lower woofer and its port both roll off a little earlier than the upper woofer and its port, which both extend higher into the midrange. But to produce a legible graph, the red trace in fig.3 shows the sum of the two woofer outputs, the green trace the sum of the two port outputs. The blue trace in this graph is the lower-midrange dome's nearfield response, while the black trace is the sum of all five nearfield LF outputs (each scaled in the ratio of the square root of its radiating diameter), taking into account acoustic phase and the different distance of each source from a nominal farfield microphone position.

Fig.3 Linn Akurate 242, midrange output (blue), complex sum of the woofer outputs (red), and complex sum of the port outputs (green), measured in the nearfield and scaled in the ratio of the square roots of the radiating areas. Black trace is the complex sum of the nearfield responses, taking into account acoustic phase and distance from the nominal farfield point.

The woofers (red trace) show only a vestigial reflex notch, this lying at the 35Hz frequency suggested by the impedance graph, with the ports covering a broad bandpass of 25–80Hz. The midrange hash evident in the ports' output is well down in level, and the possibility of coloration will be further reduced by the fact that the ports face away from the listener. The woofers cross over to the lower-midrange unit just below 500Hz.

Fig.4 repeats the complex sum of the nearfield LF unit responses below 300Hz; above that frequency, this graph shows the Akurate's farfield response, averaged across a 30 degrees horizontal window on the supertweeter axis. The speaker's bass is down 6dB at the 35Hz tuning frequency of the ports—good bass extension—but the absence of the usual nearfield "bump" in the upper-bass response suggests that the Akurate's bass alignment is actually somewhat overdamped. This is probably why AD found the speaker's bass to be unfazed by the various changes he made in placement.

Fig.4 Linn Akurate 242, 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.

Higher in frequency, what would be an impressively even response on this axis, extending up to the 30kHz limit of this graph, is disturbed by a lack of mid-treble energy offset by a peak at 4kHz. The "furriness" of the trace above this frequency is subjectively innocuous, but results from what appear to be some early reflections of the Akurate HF output, perhaps from the cavity behind the upper-frequency drive-unit array.

As alarming as the mid-treble response anomaly might look, it needs to be considered in combination with the Akurate's dispersion. Fig.5 shows the Linn's horizontal radiation pattern. It's hard to see what's happening in the mid-treble, but careful comparison of fig.5 with fig.4 reveals that the on-axis peak at 4kHz is compensated by a lack of energy off-axis, while the on-axis suckout between 5 and 7kHz is also compensated by an excess of off-axis energy. Only the on-axis dip between 2 and 3kHz continues to the speaker's sides.

Fig.5 Linn Akurate 242, lateral response family at 50", normalized to response on tweeter axis, from back to front: differences in response 90–5 degrees off-axis, reference response, differences in response 5–90 degrees off-axis.

In the vertical plane (fig.6), the Akurate's balance doesn't change significantly over quite a wide window, with the exception of the 4kHz peak on the supertweeter axis flattening out at 10 degrees above this axis and higher. As noted earlier, the supertweeter axis is a quite low 33" from the floor; taking the Akurate's vertical radiation pattern into account, along with the manner in which its horizontal dispersion will compensate for the mid-treble deficiencies in its on-axis behavior, I'm not surprised that AD felt that the speaker offered a well-balanced sound. However, I do feel that the balance between the Akurate's on- and off-axis behaviors was the reason he thought that voices and instruments took on a hollow sound when he spaced the speakers too far apart or toed them in.

Fig.6 Linn Akurate 242, vertical response family at 50", normalized to response on tweeter axis, from back to front: differences in response 20–5 degrees above axis, reference response, difference in response 5 degrees below axis.

The Akurate 242's step response is shown in fig.7. With the possible exception of the supertweeter, all the drive-units appear to be connected in positive acoustic polarity; and while the speaker is not time-coincident, the step of each driver smoothly hands over to that of the next lowest in frequency. In order to produce a cumulative spectral-decay plot from the speaker's impulse response, the latter had to be aggressively windowed to eliminate the reflection of the sound from the floor between the speaker and the microphone. The region where this results in invalid data is dotted in the resultant graph (fig.8). The 4kHz peak seen in the on-axis response again makes an appearance, associated with a small amount of delayed energy. The graph is less clean in the high treble than I would have liked to have seen, but the slight amount of low-level hash evident is perhaps related to the early reflections I noted before.

Fig.7 Linn Akurate 242, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).

Fig.8 Linn Akurate 242, cumulative spectral-decay plot at 50" (0.15ms risetime).

Overall, the Linn Akurate 242 demonstrates some impressive audio engineering, though its on-axis behavior needs to be balanced by placement and in setup.—John Atkinson