Epos ELS-3 loudspeaker Measurements
Despite its small size, the Epos ELS-3 is of average sensitivity, at a calculated 87.5dB(B)/2.83V/m. In addition, with an impedance that remains above 6 ohms for most of the audioband and drops to a minimum value of 4.9 ohms at 250Hz (fig.1), it is fairly easy for an amplifier to drive (though the combination of 6 ohms and a -45 degrees phase angle at 160Hz will be a bit stressful for amplifiers not rated into 4 ohms). The saddle at 59Hz in the magnitude trace gives a clue to the tuning frequency of the rear-facing port.
Fig.1 Epos ELS-3, electrical impedance (solid) and phase (dashed). (2 ohms/vertical div.)
Other than a glitch in the fig.1 traces just below 1kHz, there are no signs of cabinet panel resonances in this graph. Even so, an accelerometer stuck to the center of the sidewall found a mild resonant mode at 176Hz, and a much stronger one at 605Hz (fig.2). The latter mode may be too high in frequency to introduce coloration, however.
Fig.2 Epos ELS-3, cumulative spectral-decay plot calculated from the output of an accelerometer fastened to the cabinet's side panel. (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz.)
The black trace in fig.3 shows the ELS-3's overall response on the tweeter axis, averaged across a 30 degrees horizontal window and spliced to the complex sum of the nearfield woofer and port responses. What would otherwise be a commendably smooth and even balance is broken by an excess of upper-midrange energy, which, all else being equal, I would have thought would make the speaker sound rather nasal. (It also contributes to the better-than-expected sensitivity.) The upside of a response like this is that, as BJR found, it makes the speaker sound very detailed. Another downside is that, again as BJR noted, it makes the speaker's upper midrange sound "glary" at high playback levels.
Fig.3 Epos ELS-3, anechoic response on tweeter axis at 50", averaged across 30 degrees horizontal window and corrected for microphone response, with nearfield responses of the woofer (red) and port (blue) and their complex sum, taking into account acoustic phase and distance from the nominal farfield point (black), plotted below 300Hz, 1kHz, and 600Hz, respectively.
The nearfield traces on the left of fig.3 reveal a conventionally tuned reflex design, the port's maximum output coinciding with the woofer's minimum-motion point at 59Hz—which, not coincidentally, is the -6dB point. If you ignore the usual 3dB nearfield boost in the upper bass, this is only modest bass extension in absolute terms. However, it is quite acceptable in a small speaker such as this. Other than a low-level resonant mode at 605Hz, the port's out-of-band output is clean. At the other end of the spectrum, the tweeter has a big resonance at 20kHz—a little lower in frequency than is usually found in aluminum-dome units. A second resonance is just apparent at 30kHz.
Turning to the ELS-3's radiation pattern, fig.4 reveals the speaker's horizontal dispersion to be pretty uniform, except for the usual beaminess above 10kHz and a bit of off-axis flare at the bottom of the tweeter's passband. (The latter may ameliorate the audibility of the upper-midrnage peak.) In the vertical plane (fig.5), a big suckout centered on the crossover frequency of 3kHz develops immediately above the tweeter axis. This suggests that high stands are to be preferred. As long as the listener sits with his ears between the tweeter and the top of the woofer, the balance should be optimal.
Fig.4 Epos ELS-3, lateral 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.
Fig.5 Epos ELS-3, vertical response family at 50", normalized to response on tweeter axis, from back to front: differences in response 45 degrees-5 degrees above axis, reference response, differences in response 5 degrees-45 degrees below axis.
In the time domain, the Epos's step response (fig.6) shows that both drive-units are connected in the same, positive acoustic polarity, but there is a suspicious-looking reflection about a millisecond after the first arrival. The cumulative spectral-decay, or waterfall, plot (fig.7) shows that this is associated with the upper-midrange peakiness. Again, I would have expected this to have made the speaker sound rather nasal.
Fig.6 Epos ELS-3, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).
Fig.7 Epos ELS-3, cumulative spectral-decay plot at 50" (0.15ms risetime).
All things considered—especially its price of $300/pair—the Epos ELS-3 is a nicely engineered little speaker that its owner needn't apologize for.—John Atkinson