Coincident Speaker Technology Troubador loudspeaker Measurements

Sidebar 2: Measurements

With a calculated sensitivity of 91dB/W/m, the Troubador will play loud with only a few amplifier watts. Its impedance (fig.1) is also relatively benign. Fig.9 reveals the port tuning to lie at 55Hz, quite low for the enclosure size. Note the wrinkle in the impedance magnitude trace at 350Hz, a sign of some kind of cabinet resonance. And indeed, the cabinet vibrated strongly at that frequency. As the designer points out, the higher in frequency you can push a cabinet resonance, the smaller the effect it will have on music. And WP didn't remark on any midrange congestion.

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

In the frequency domain, the Troubador's low-frequency response (shown to the left of fig.2) is almost textbook: the port output is centered on 55Hz, as implied by the impedance plot; the woofer rolls off to reach its minimum-motion point at the same frequency; while the complex sum of the woofer and port outputs (allowing for the physical distance between them) is maximally flat to 70Hz, rolling off with a 24dB/octave slope below that frequency. The –6dB point is an impressively low 39Hz.

Fig.2 Coincident Troubador, anechoic response on tweeter axis at 50", averaged across 30° horizontal window and corrected for microphone response, with nearfield woofer and port responses and their complex sum (top) plotted below 300Hz, 1kHz, and 300Hz, respectively.

Higher in frequency, fig.2 shows that things fall apart. The low treble is marred by a massive suckout, while the tweeter region is both shelved up by up to 10dB and marred by an on-axis dip at 9kHz due to the horn-loading of the tweeter. (Both samples of the speaker measured identically.) In my own auditioning, I found the Troubador's tonal balance severely flawed, with both a hollow-sounding coloration and, as WP noted above, the feeling that the treble is disconnected from the lower frequencies.

As expected from the coincident drive-unit layout, the Troubador's horizontal dispersion (fig.3) is even, though the on-axis dip at 9kHz fills in to the sides. The vertical dispersion (not shown) is effectively identical.

Fig.3 Coincident Troubador, horizontal 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.

In the time domain, the step response (fig.4) reveals that despite its use of coincident drive-units, the Troubador is not time-coherent. The tweeter's output (the small, negative-going spike) arrives first, followed a fraction of a millisecond later by the positive-polarity woofer output. In the crossover region, they're acoustically out-of-phase and their outputs cancel, giving rise to the severe mid-treble suckout seen in fig.2.

Fig.4 Coincident Troubador, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).

Finally, the Troubador's waterfall plot (fig.5) continues the bad news. Overall, this is not a particularly well-engineered loudspeaker. I suspect that both reversing the tweeter polarity and reducing its sensitivity would give a better-balanced sound.—John Atkinson

Fig.5 Coincident Troubador, cumulative spectral-decay plot at 50" (0.15ms risetime).

Share | |

Enter your username.
Enter the password that accompanies your username.