Meridian DSP8000 digital active loudspeaker Measurements
Performing acoustic measurements on a speaker that has only a digital input posed some problems, as the DRA Labs MLSSA system I use for all of my loudspeaker tests outputs only an analog signal. I therefore digitized the MLSSA stimulus at 24 bits/88.2kHz with a dCS 904 A/D converter, and converted the AES/EBU datastream to S/PDIF using a Z-Systems rdp-1, this set to do nothing but convert formats. The peak level of the digital signal, measured with a Dorrough digital-domain meter, was -6dBFS.
The second problem involved the sheer bulk of the DSP8000—it was impossible for me to lift it onto my Outline loudspeaker turntable for measurement. Fortunately, the '8000's top section is connected with an umbilical cord to the woofer bin. Meridian sent me a 10' extender cord, which allowed the tweeter/midrange section to be placed high in the air on the turntable, while the woofer section, with its digital circuitry and amplifiers, could be kept on the ground. As the crossover between the sections is set to 200Hz, below the frequency region where I can get accurate quasi-anechoic measurements, I felt this was an acceptable compromise.
The usual concept of "speaker sensitivity" is not applicable to the '8000. However, when I fed the Meridian pink noise at -6dBFS with its volume control set to "70," the B-weighted level at 1m was 84.8dB—this compared with my reference BBC LS3/5a, which gives 82.5dB(B) at 1m when fed with 2.83V RMS. Given that the Meridian's volume control goes up to "99" in steps of 1dB and that the digital level could be increased by another 6dB, this gives a maximum SPL for each speaker of 120dB (assuming the internal amplifiers don't clip)! That certainly makes the DSP8000 a loud speaker.
Fig.1 shows the Meridian's quasi-anechoic farfield response, averaged across a 30 degrees horizontal window on the tweeter axis and spliced to the nearfield responses of the midrange and woofers. The latter smoothly covers the passband between 30Hz and 150Hz, with the ultimate high-pass rollout apparently arranged to be 24dB/octave rather than the 12dB/octave typical of a sealed-box alignment. As a result, the '8000 will not benefit as much as usual from room reinforcement of the low-bass register. However, as I found in my auditioning, this was not a problem.
Fig.1 Meridian DSP8000, anechoic response on tweeter axis at 50", averaged across 30 degrees horizontal window and corrected for microphone response, with nearfield midrange and woofer responses plotted below 400Hz.
The crossover between the woofers and the midrange unit features fourth-order slopes, with the woofers' ultimate low-pass rollout even higher in slope, presumably due to the drive-unit's own acoustic behavior. The head unit's balance is basically smooth, though with the HF tilt control set to "0" there is a slightly elevated plateau apparent through the upper-midrange and presence regions. Although the tweeter's "oil-can" dome resonance results in a narrow spike 12dB above the reference level at 1kHz, this can be seen to occur at a high 26kHz. With CD material, this resonance will have no audible consequences, as it will not be excited by the band-limited source, even when upsampled by the Meridian 800. The resonance will be excited by wide-bandwidth SACD and DVD-Audio material, but I heard nothing untoward when using these sources.
The effect of the HF tilt control is shown in fig.2. Operating in 0.5dB steps, it basically hinges the speaker's response by up to ±5dB each side of a 1.3kHz center frequency, which is almost exactly what is needed to balance the '8000's perceived response. I say "almost" because, if you look again at fig.1, the hinge frequency is set about an octave lower than the top edge of the presence-region plateau.
Fig.2 Meridian DSP8000, effect on response of HF control set to "±10," normalized to response on tweeter axis at 50" (5dB/vertical div.).
A loudspeaker's perceived in-room balance will depend not only on its on-axis response but also on its dispersion—its radiation patterns in the horizontal and vertical planes. Fig.3 shows the vertical dispersion of the '8000's head unit. (As I said earlier, it had to be operated remotely from the woofer bin for the measurements, which means any effect of the lower crossover is absent from this graph.) In fig.3, the individual measurements are normalized to the tweeter axis; deep suckouts appear above and below the tweeter at exactly the specified 2.6kHz crossover frequency, but as I had inadvertently left the axis control set to "-1," the above-axis suckout appears earlier than the lower one. Other than that, the Meridian's off-axis behavior is well-controlled.
Fig.3 Meridian DSP8000, vertical response family at 50", normalized to response on tweeter axis, from back to front: differences in response 45 degrees-5 degrees above reference axis; reference response; differences in response 5 degrees-45 degrees below reference axis (not including baffle effect of woofer cabinet).
I have shown two graphs for the lateral dispersion: fig.4 shows the actual responses plotted out to 90 degrees away from the tweeter axis in 5 degrees increments, and fig.5 has the off-axis plots normalized to the tweeter-axis response to emphasize any differences that exist. The contour lines in these two graphs are evenly spaced, always a good indicator of well-defined stereo imaging. Particularly obvious in fig.5 are both the narrowing of the DSP8000's dispersion in the top octave of the midrange unit's passband and a flare at the bottom of the tweeter's passband.
Fig.4 Meridian DSP8000, horizontal response family at 50", from back to front: responses 90 degrees-5 degrees off-axis; reference response on tweeter axis; responses 5 degrees-90 degrees off-axis.
Fig.5 Meridian DSP8000, horizontal response family at 50", 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.