Epos ES12 loudspeaker Measurements
My estimate of the ES12's B-weighted sensitivity was 85dB/2.83V/m—slightly lower than specified, but the difference is inconsequential. Its impedance (fig.1) doesn't drop below 7 ohms, meaning that it's basically an easy load for an amplifier to drive. The tuning of the reflex port is revealed by the "saddle" in the impedance magnitude plot at 58Hz, implying only moderate bass extension.
Fig.1 Epos ES12, electrical impedance (solid) and phase (dashed) (2 ohms/vertical div.).
Two discontinuities can be seen in the impedance traces: a slight one between 400Hz and 500Hz, and a larger peak at 650Hz. These generally indicate that there is some sort of resonant problem at these frequencies. Fig.2 shows a cumulative spectral-decay or waterfall plot calculated from the output of a simple plastic-tape accelerometer fastened to the Epos's front baffle beneath the woofer. (The speaker was standing on three upturned cones for this measurement.) The resonant mode at 465Hz could be found on all surfaces; it was also easily detected with a stethoscope using the semitone-spaced tones on Stereophile's Test CD 3. (The mode is almost exactly coincident in frequency with the B-flat in the center of the treble staff.) While this mode is relatively high in frequency, indicating a well-braced cabinet design, it is also high in level, meaning that the speaker will be more sensitive than most to stand and stand-interface issues—as indeed I found.
Fig.2 Epos ES12, cumulative spectral-decay plot of accelerometer output fastened to front baffle above tweeter. (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz.)
Fig.3 shows the individual responses of the woofer, port, and tweeter. The port's output peaks sharply at the tuning frequency, which is also the woofer's minimum-motion point. Note that the port's output has peaks at the frequencies of the anomalies in the impedance plot, but these are well down in level. In the December 1996 issue of the English magazine Hi-Fi News & Record Review, Martin Colloms ascribes the Epos ES12's 650Hz mode to the woofer cone. The woofer's response is a little ragged above this frequency—remember, there is no crossover filter at all on this drive-unit; what you see in this graph is its natural rollout behavior. The tweeter, fed by a series capacitor, comes in above 5.5kHz, about an octave higher than is usual for a two-way design with a 1" dome unit. Its ultrasonic resonance seems of lower Q than normal, with just a 9dB rise above the reference level.
Fig.3 Epos ES12, acoustic crossover on tweeter axis at 50", corrected for microphone response, with nearfield woofer and port responses plotted below 300Hz.
The ES12's overall response on the tweeter axis, averaged across a 30° horizontal window to minimize any specific microphone-position anomalies, is shown in fig.4, spliced to the complex sum of the nearfield woofer and port outputs. The speaker rolls off with a 24dB/octave slope below 55Hz, with a –6dB point of 45Hz—not bad extension for such a small speaker. The shape of the curve between 500Hz and 1kHz is usually associated with a nasal coloration. I did note this in my auditioning, but it was mild. Considering the lack of any formal crossover (other than the tweeter capacitor), the integration between the two drive-units in fig.4 is excellent. Other than slight suckouts at 2.5kHz and 8kHz, the measured treble balance is flat, giving no clue as to why I found the high treble a little mellow.
Fig.4 Epos ES12, anechoic response on tweeter axis at 50", averaged across 30° horizontal window and corrected for microphone response, with complex sum of nearfield woofer and port responses plotted below 300Hz.
The ES12's lateral dispersion is shown in fig.5. Despite its use of a "phase plug," taking the woofer up an octave higher than usual does result in slight beaming at the top of its passband, which contrasts with the "flare" in dispersion at the bottom of the tweeter's band. This is mild, however, and as the off-axis flare coincides with one of the on-axis suckouts, the perceived treble balance in a typical room will probably be fairly neutral.
Fig.5 Epos ES12, horizontal 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 shows the vertical dispersion, referenced to the response on the tweeter axis. A big suckout appears even if you sit just above the tweeter, whereas sitting 5° below it usefully pulls down the mid-treble. ES12 owners should use relatively high stands, therefore, if they are to get the optimal balance from this speaker.
Fig.6 Epos ES12, vertical response family at 50", normalized to response on tweeter axis, from back to front: differences in response 45°–5° above tweeter axis; reference response; differences in response 5°–45° below tweeter axis.
Epos has voiced the ES12 to give its most neutral lower-midrange balance when a degree of room-boundary reinforcement is present. With the speakers positioned where I liked them best for the bass quality and soundstaging, I found there was insufficient room support for the lower midrange. This can be seen in fig.7, the spatially averaged, 1/3-octave response for the pair of Eposes in my room. The lower-midrange and bass regions are shelved-down by 3–4dB; or, you could say that the upper midrange is shelved up by the same amount. Whether this character will be heard as leanness or forwardness will depend on both the listener and the program material. However, it did lend the speaker a wonderfully articulate character without any tendency toward brightness.
Fig.7 Epos ES12, spatially averaged, 1/3-octave response in JA's room.
The ES12's step response (fig.8) holds no surprises, both drive-units being connected with the same positive acoustic polarity, with the tweeter leading the woofer by 250;us or so. The corresponding waterfall plot (fig.9) reveals a very clean decay from the tweeter, with only a couple of low-level resonant modes present at the top of the woofer's passband.
Fig.8 Epos ES12, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).
Fig.9 Epos ES12, cumulative spectral-decay plot at 50" (0.15ms risetime).
Overall, this is excellent measured performance for what is basically an affordably priced loudspeaker.—John Atkinson