Silverline Audio Technology Prelude loudspeaker Measurements

Sidebar 3: Measurements

All my measurements were performed on samples of the "Revised" version of the Prelude. The terminal posts were loose on both samples and one of the samples (serial no.2041A) appeared to be miswired: the drive-units were silent when driven individually, the speaker giving out sound only when both sets of terminals were connected to the amplifier. Although its farfield response and sensitivity then matched that of the second sample (serial no.2041B), I put it back in the box and performed all the measurements on the second speaker. (I left the grille off.)

Though its estimated voltage sensitivity on the tweeter axis is very slightly higher than average, at 87.7dB(B)/2.83V/m, the Prelude is significantly less sensitive than the specified 91dB. Its impedance remains between 4 and 6 ohms in the midrange, with a minimum value of 4.2 ohms just above 1kHz (fig.1), but the speaker is otherwise relatively easy to drive. There is a major discontinuity around 180Hz in the impedance traces, indicating the presence of some sort of cabinet resonance at this frequency. I investigated the panel's vibrational behavior with a plastic-tape accelerometer and found a very strong resonance present on all surfaces, not at 180Hz but at 422Hz (fig.2). This might have been thought too high in frequency for its effect to be heard as coloration, but as Robert Deutsch did comment on box coloration at high playback levels, I assume that this strong mode was the cause.

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

Fig.2 Silverline Prelude, cumulative spectral-decay plot calculated from the output of an accelerometer fastened to the center of the cabinet's rear panel (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz).

I'm always suspicious of speaker cabinets that are narrow but tall, as such a shape increases the likelihood of the cabinet starting to act as a tuned pipe. Looking at the nearfield responses of the woofers and port (fig.3, left-hand traces), it can be seen that the port does have a peak in its output centered on 180Hz, the frequency of the impedance anomaly, and that the woofer's output is notched at the same frequency. The relative levels of the port and woofers in this graph were calculated in the ratio of the square roots of the total radiating areas, and appear to indicate that the port doesn't fully extend the Prelude's low-frequency response. However, the port's proximity to the floor will boost its output in-room.

Fig.3 Silverline Prelude, acoustic crossover on tweeter axis at 50", corrected for microphone response, with the nearfield responses of the port and woofers plotted below 300Hz.

Higher in frequency, the response of the woofers rises to a peak at 1100Hz before rolling off with an approximate third-order slope (both drivers had an identical output). Unusually, even though the tweeter was not driven for the measurement of the woofers' output, it was putting out some sound, revealed by the presence of its ultrasonic resonance peak about 30dB down from the nominal midband level. The tweeter appears to be rolled in with a steep fourth-order slope, and though a narrow suckout is visible at 5.2kHz, this is probably a position-specific problem. As is often the case, the unit's top-octave response is shelved down a little before rising again to a peak at 25kHz, this due to the inevitable but inaudible metal-dome resonance.

Fig.4 shows the Prelude's overall frequency response, averaged across a 30° horizontal window on the tweeter axis and spliced at 300Hz to the sum of the nearfield responses (taking into account acoustic phase and the different distances of the radiators from the nominal farfield measurement point). The speaker is quite well balanced overall, but with a prominence in the upper midrange and a slight lack of energy in the crossover region. The excessive upper-midrange energy will be heard as extra detail, and, I suspect, is the reason RD remarked on the revised Prelude's having a slightly forward sound. It might also add a touch of nasal coloration, but I note that RD heard nothing untoward of this nature. Other than the notch at the port resonance frequency, the Prelude's lower frequencies are smooth and gently shelve down, suggesting a rather overdamped alignment. But if you take the usual amount of room gain present at low frequencies, the Silverline's bass output will usefully extend down to the mid-40Hz region. Its overall balance, however, might still be on the lightweight side.

Fig.4 Silverline Prelude, anechoic response on tweeter axis at 50", averaged across 30° horizontal window and corrected for microphone response, with the complex sum of the port and woofer nearfield responses plotted below 300Hz.

In the Prelude's plot of lateral dispersion (fig.5), what looks like a slight flare at the bottom of the tweeter's passband is actually due to the on-axis suckout at 3.5kHz tending to fill in to the speaker's sides, though the depression a half-octave lower actually deepens off axis. Whether or not the Silverline sounds neutral in the low treble, and whether or not the combination of upper-midrange prominence and low-treble depression makes the speaker sound nasal, will depend on the size of the listening room. The contour lines in this graph are generally even, with no "hot spots," which correlates with the stable stereo imaging RD found in his auditioning. In the vertical plane (fig.6), the Prelude's balance doesn't change significantly for listening axes between the base of the lower woofer and the top of the upper woofer. This is optimal behavior given the 36" height of the typical seated listener's ears.

Fig.5 Silverline Prelude, lateral 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.

Fig.6 Silverline Prelude, vertical response family at 50", normalized to response on tweeter axis, from back to front: differences in response 20–5° above axis, reference response, differences in response 5–10° below axis.

In the time domain, the Prelude's step response (fig.7) indicates that despite Silverline specifying the tweeter to be connected in inverted acoustic polarity, all three drive-units are actually connected in positive polarity. The ripple just before the 5ms mark ties in with the upper-midrange prominence. Other than that, the Prelude's cumulative spectral-decay plot (fig.8) is superbly clean.

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

Fig.8 Silverline Prelude, cumulative spectral-decay plot at 50" (0.15ms risetime).

Whether or not its tonal balance is perceived as a heightened sense of musical detail or as coloration will depend on the room's furnishings and size. But as RD found, in a fairly small room, the Prelude seems to achieve a balanced performance.—John Atkinson

Silverline Audio Technology, Inc.
P.O. Box 30574
Walnut Creek, CA 94598
(925) 825-3682

Gerald Beato's picture

the speaker was so beautiful.