Spica TC-60 loudspeaker Measurements

Sidebar 3: Measurements

The TC-60 is significantly less sensitive than the Dunlavy SC-I, its B-weighted figure being around 85dB/2.83V/1m. It's easier to drive, however, its impedance (fig.1) remaining above 7 ohms other than in the high treble, where it drops to a still-kind minimum of 6.2 ohms. Note the slight wrinkle in the traces just under 300Hz—this implies the existence of a cabinet resonance at this frequency. The port tuning is indicated by the saddle in the magnitude trace at 29Hz. This is very low for the size of the speaker; it's difficult to predict what the actual bass extension will be.

Fig.1 Spica TC-60, electrical impedance (solid) and phase (dashed) (2 ohms/vertical div.).

The actual acoustic outputs of the drive-units and the port can be seen in fig.2. The woofer has a very slight rise in its response in the upper bass before rolling out gently to its minimum-motion point at the port tuning frequency of 29Hz. The port's output is centered in the same region, but is way down in level—you can't expect much from a 1.5" port tuned to such a low frequency. Two higher-frequency resonant modes can be seen in the port's output—at 300Hz and 500Hz—but these will be innocuous considering that the port faces to the speaker's rear.

Fig.2 Spica TC-60, acoustic crossover on listening axis at 50" corrected for microphone response, with nearfield woofer and port responses below 300Hz and 600Hz, respectively.

The woofer is well-behaved in its upper frequency range before it rolls out sharply above the crossover frequency of 3.5kHz. There is a slight notch/peak in the 1kHz region, perhaps due to the cavity formed by the felt blanket. The slow roll-in of the tweeter can be seen in this graph, but apart from a premature rolloff above 12kHz, it too appears to be well-behaved.

How these individual responses sum on the listening axis level with the top of the woofer is shown in fig.3, which shows the overall response averaged across a 30° window. Impressively smooth for such an inexpensive speaker, the upper bass is slightly boosted, the presence region slightly depressed—which is what I expected from my auditioning. Though it's down in absolute level, the port does usefully reinforce the TC-60's bass output to give a gentle rollout, down 6dB at 40Hz. (This is just below the lowest note of the 4-string bass guitar and double-bass.) More importantly, it's only 12dB down at the port tuning frequency of 29Hz. In effect, the QB3 alignment chosen by John Bau cheats, or at least bends the laws of physics, meaning that as long as you don't play the speaker too loud—it's only a small port—some low-bass frequencies will be audible. As I found.

Fig.3 Spica TC-60, anechoic response on listening axis at 50" averaged across 30° horizontal window and corrected for microphone response, with complex sum of nearfield woofer and port responses below 300Hz.

The top octave does roll off a little early, tying in with the speaker sound lacking a bit of "air" on this axis. Often such a rolloff is associated with limited horizontal dispersion in the high treble. The TC-60's tweeter, however, does maintain its output moderately well to the sides, as can be seen in fig.4. A notch at 12kHz does develop at off-axis angles greater than 15°, however, implying that the speakers be toed-in to the listening position. (Ignore the apparent peak in this graph at 26kHz: this is due to the notch at this frequency in the on-axis response filling in to the speaker's sides.)

Fig.4 Spica TC-60, horizontal response family at 50", normalized to response on listening axis, from back to front: differences in response 90°–5° off-axis on other side of baffle; reference response; differences in response 5°–90° off-axis on tweeter side of baffle.

Fig.4 shows only the off-axis differences in response. It also reveals that the TC-60's wide baffle results in the woofer's dispersion becoming limited lower in frequency than might be expected from its small radiating diameter. At off-axis angles greater than 20°, the speaker lacks energy in the top octave of its passband, meaning that the Spica's sound might lack immediacy in lively rooms. This was not a factor in my relatively well-damped room.

Vertically, the Spica's unique combination of a fourth-order low-pass filter for the woofer and a first-order high-pass for the tweeter results in quite a critical listening window. Fig.5 reveals that, had I measured a little lower than the top of the woofer, the presence-region depression in fig.3 would have filled in nicely, though the price to be paid is a reduced amount of top-octave energy. Conversely, moving even 5° above this axis results in a severely compromised response, the presence region dropping and the top octave becoming boosted. As I always seem to be saying—don't listen to this speaker standing up. (That Spica is aware of this fussiness is indicated in the TC-60's excellent owner's manual; they even recommend tilting the speaker back slightly to ensure that you are sitting below the tweeter axis.)

Fig.5 Spica TC-60, vertical response family at 50", normalized to response on tweeter axis, from back to front: differences in response 45°–5° above reference axis; reference response; differences in response 5°–90° below reference axis.

In my listening room (fig.6), the result of this excellent on-axis performance but more idiosyncratic off-axis behavior was a superbly flat balance throughout the audioband. With the exception of the presence region, where the contribution of the above-axis response resulted in a slight lack of energy in the speaker's room sound, the Spica's 1/3-octave response met superb ±0.7dB limits from the lower midrange up to 10kHz. Few speakers at any price have done this well in my listening room. That the $800 Spica can do it is testament to the talent of its designer.

Fig.6 Spica TC-60, spatially averaged 1/3-octave response in JA's listening room.

A flat perceived balance is paramount in a speaker's performance; good time-domain performance is a nice bonus if you can get it. The TC-60's impulse response (fig.7) is indeed pretty time-coherent, as is its step response (fig.8), which possesses the desired right-triangle shape, broken only by a leading-edge peak which could well be due to my measuring microphone. (The ripple in the tail between 6ms and 7ms is due to the reflection of the bass sound from the floor being magnified by the mathematical integration employed to produce this graph.)

Fig.7 Spica TC-60, impulse response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).

Fig.8 Spica TC-60, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).

With such good time-domain performance, you'd expect the TC-60 to have very little excess phase—which it doesn't (fig.9). This meets ±15° limits from 700Hz to 20kHz, which is superb performance. The small positive-phase angle for the tweeter indicates that it's placed behind the woofer by a millimeter or so on this axis. As shown in the vertical-response–family graph (fig.5), moving down slightly would have resulted in the flattest midrange and treble balance. But this would move the tweeter back in time even more, suggesting that there isn't actually a listening axis on which you get both the flattest amplitude response through the crossover region and the optimum time-domain performance. This is a pedantic point, however, as, in either case, the Spica's departure from the ideal is very small.

Fig.9 Spica TC-60, excess phase on tweeter axis at 50" (0.147ms delay).

The TC-60's waterfall plot (fig.10) reveals a very clean decay, with only a small degree of overhang in the upper midrange. However, a similar waterfall graph plotted for the output of a simple accelerometer attached to the center of the speaker's back panel (fig.11) reveals the existence of a strong cabinet resonance at 273Hz, and another just below it. These modes could also be detected on the side panels and, to a lesser degree, on the small top panel. That the front baffle is covered with felt will damp their effects, which is probably why I only occasionally noted any lower-midrange congestion.—John Atkinson

Fig.10 Spica TC-60, cumulative spectral-decay plot at 50" (0.01ms risetime).

Fig.11 Spica TC-60, cumulative spectral-decay plot of accelerometer output fastened to center of enclosure back panel. (MLS driving voltage to speaker, 7.55V; measurement bandwidth 2kHz.)

Spica Loudspeakers Division of Parasound Products Inc.