Totem Acoustic Model 1 loudspeaker Measurements

Sidebar 2: Measurements

The Totem was about 2dB more sensitive than the BBC LS3/5A, placing it around 84.5dB/W/m, which, while being rather lower than specified, is what I would expect from such a small cabinet. It also needs to be driven by a good amplifier, its impedance magnitude dropping to 3.6 ohms in the upper bass (fig.1). The saddle in this plot also reveals the port to be tuned to 42Hz, the lowest note of the 4-string double bass, while the reduced magnitude of the lower bass peak suggests a slightly under-damped LF alignment. While this, if carried to an extreme, can result in an objectionable "one-note" bass quality, if carefully managed in a minimonitor it can give the subjective effect of a musically appropriate amount of low frequencies.

Fig.1 Totem Model 1, electrical impedance (solid) and phase (dashed). (2 ohms/vertical div.)

Fig.2 is a composite plot, showing the quasi-anechoic response of the Totem under the following conditions: with its grille off, averaged across a 30 degrees horizontal window on the tweeter axis at 45", above 200Hz, and the individual nearfield responses of the woofer (red) and port (blue), plotted below 200Hz and 1kHz, respectively, as well as their complex sum below 200Hz (black), taking acoustic phase into account. The woofer starts to roll out below 100Hz, with the port being responsible for reproducing everything below 75Hz or so. LG found the Totems to be able to deliver good, clean bass down to 50Hz in-room. The port also has a couple of pipe resonances noticeable at 500Hz and 950Hz, though the first of these is well down in level. The main response is somewhat upper-midrange-dominant, with a slightly suppressed top octave, as suggested by LG's auditioning. The tweeter's main resonance is very high in level, reaching some 20dB above the midband reference level; but at 26kHz, this will be inaudible.

Fig.2 Totem Model 1, anechoic response on tweeter axis at 45", averaged across 30 degrees horizontal window and corrected for microphone response, with the nearfield responses of the woofer (red) and port (blue) and their complex sum (black) plotted below 200Hz, 1kHz, and 200Hz, respectively.

In the vertical plane, the Totem is relatively uncritical regarding the optimal listening axis, as can be seen from fig.3, which shows just the changes from the response on the tweeter axis as the listener moves up and down. Only when listened to above the top of the baffle will a suckout at the crossover frequency start to appear. Horizontally (fig.4), the top two octaves progressively roll off with increasing off-axis angle, while a peak appears at the bottom of the tweeter's passband, due to the woofer being more directional at the crossover frequency. The off-axis behavior does compensate somewhat for the on-axis peakiness.

Fig.3 Totem Model 1, vertical response family at 45", normalized to response on tweeter axis, from back to front: differences in response 15 degrees-7.5 degrees above axis, reference response, differences in response 7.5 degrees-15 degrees below axis.

Fig.4 Totem Model 1, lateral response family at 45", 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.

In the time domain, the Totem's impulse response (fig.5) is dominated by the tweeter resonance, with the ultrasonic ringing predominant. The DRA Labs MLSSA software allows a speaker's step response, the output of the speaker when presented with a DC voltage step, to be calculated. The ideal shape should resemble a right triangle, a perpendicular step away from the time axis followed by a sloped line back to it, due to the speaker basically behaving as a high-pass filter. That for the Totem is shown in fig.6; though the tweeter and woofer are connected with the same polarity, the tweeter's output (the sharp up-and-down spike at 3.4ms) can be seen to precede the woofer's output (the lazier hump, overlaid with both the tweeter's ringing and some lower-frequency component) by about 0.5ms. This reflects the fact that, in a speaker with a flat front baffle, the physically less deep tweeter is mounted more forward than it needs to be for correct time alignment.

Fig.5 Totem Model 1, impulse response on tweeter axis at 45" (5ms time window, 30kHz bandwidth).

Fig.6 Totem Model 1, step response on tweeter axis at 45" (5ms time window, 30kHz bandwidth).

Finally, by taking progressive "slices" of the impulse response and calculating the frequency response of the slice with the Fast Fourier Transform, the MLSSA system can reveal the presence of resonances as "ridges" in the resultant cumulative spectral-decay, or "waterfall," plot. The die-away of the Totem 1's sound is basically clean, as can be noted from fig.7, with the exception of what is presumably a woofer-cone breakup mode at 2.3kHz (the cursor position). This resonance is the reason for the low-treble-prominent response in fig.3 and for the low-frequency ringing on the tail of the woofer's step response (fig.6). It might be expected to add a little hardness at high levels, though I note that LG noted no such character to the speaker's sound. Logistical problems meant that I was unable to perform any cabinet-resonance measurements. Nevertheless, the Totem's enclosure seems solid and well-damped.—John Atkinson

Fig.7 Totem Model 1, cumulative spectral-decay plot at 45" (0.15ms risetime).