Thiel CS3.6 loudspeaker:
Measurements part 2
The cumulative spectral-decay, or "waterfall," plot reveals how a loudspeaker behaves over time after being excited by an impulse. The MLSSA system takes FFTs at discrete slices of time as the impulse decays. Each horizontal line is a separate FFT. When viewed in three dimensions (amplitude, frequency, and time), the waterfall plot shows resonances as vertical ridges. The CS3.6's waterfall plot (fig.5) was excellent, with a sudden decay (the white space between the top few traces and nearly all the subsequent traces) and few strong resonant modes. There is, however, a bit of treble hash, though this is well down in amplitude. There is also a mode at 4kHz, perhaps due to a resonance in the midrange cone. Note the generally flat response and lack of large peaks and dips in the response. This is exemplary performance.

Fig.5 Thiel CS3.6, cumulative spectral-decay plot on tweeter axis at 45" (0.15ms risetime).

Fig.6 is the CS3.6's FFT-derived amplitude response measured at the tweeter axis (36") and averaged over a 30 degrees lateral window. The nearfield woofer and passive radiator outputs, measured with the microphone nearly touching the diaphragms, have been vector-summed (this is the curve on the left side of fig.6). The exceptionally flat response is marred by a large suckout at the lower crossover frequency, caused by destructive interference on this axis between the woofer and midrange. This suckout can also be seen in the waterfall plot (fig.5). Despite the magnitude of the suckout, it was never audible as such during the auditioning. Although this dip does exist, it probably does so only at the 45" measurement distance.

Fig.6 Thiel CS3.6, anechoic response on tweeter axis at 45", averaged across 30 degrees horizontal window and corrected for microphone response, with overall nearfield loudspeaker LF rollout plotted below 300Hz.

Getting farther away from the loudspeaker will likely result in better integration of the individual drivers' outputs, making the suckout disappear. This is even more likely because the CS3.6 uses first-order slopes; the degree of overlap between drivers is much greater than with higher-order systems. Incidentally, the CS3.6 had a slightly lower HF energy output with the grille on—as Thiel recommends the loudspeaker be used.

The CS3.6's deep LF extension and generous bass output noted during the auditioning are apparent both in fig.6 and in the nearfield woofer and passive radiator responses (fig.7). The very steep rolloff below 30Hz in fig.6 is a result of the woofer and passive radiator being out of phase with each other below the latter's resonance, thus canceling the lowest frequencies. Fig.6 also shows a very slight downward tilt in the top octave, correlating to my listening impressions of a smooth top octave and also that the CS3.6 had less HF energy than the Hales Signature.

Fig.7 Thiel CS3.6, nearfield responses of midrange unit, woofer, and port