Peak Consult El Diablo loudspeaker Measurements
Although Peak Consult claims a very high sensitivity of 94dB for the Diablo, my estimate of the speaker's sensitivity was somewhat lower, at just under 90dB(B)/2.83V/m. This is still usefully higher than average, however. Its impedance (fig.1) resembles a 4–5 ohm resistor over much of the audioband, with a very small electrical phase angle, though this is significantly lower than the 7 ohms specified. There is a combination of 3.8 ohms magnitude and –34° capacitive phase angle at 27Hz, but given the low frequency, where music will rarely have high energy, this will not give rise to any current-delivery problems with good amplifiers.
Fig.1 Peak Consult El Diablo, electrical impedance (solid) and phase (dashed). (2 ohms/vertical div.)
The traces in fig.1 are free from any of the small wrinkles that would indicate the presence of cabinet resonance problems, and indeed, investigating the panels' vibrational behavior, I found almost nothing of interest. Fig.2, for example, is a cumulative spectral-decay plot calculated from the output of a simple accelerometer fastened to the cabinet sidewall level with the midrange drive-unit. Only two modes can be seen, and even the lower one, at 480Hz, is high enough in frequency and low enough in level to be subjectively irrelevant.
Fig.2 Peak Consult El Diablo, cumulative spectral-decay plot calculated from the output of an accelerometer fastened to the cabinet's side panel level with the midrange unit (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz).
The small saddle centered on 32Hz in the impedance-magnitude plot indicates that this is the tuning frequency of the two large-diameter ports on the Diablo's rear panel. The ports appear to behave very similarly; the red trace in fig.3 is the sum of their outputs, scaled with respect to the outputs of the woofers (blue trace) and midrange unit (green) in the ratio of the square root of the radiating areas. The ports' output does indeed peak in the region of 30Hz, though the corresponding minimum-motion notch in the summed woofer output (where the back pressure from the port resonance holds the woofer cones stationary) occurs a little higher in frequency, at 34Hz. Of more concern is the high-Q peak seen at 480Hz in the ports' response, coincidentally the frequency of the panel vibrational resonance seen in fig.2. This might be high enough in level to lead to coloration, though working against its audibility will be the fact that the ports face away from the listener. I was bothered by a slight "cupped hands" coloration in the midrange when I auditioned the Diablos in Mikey's room. Perhaps I was hearing this port behavior.
Fig.3 Peak Consult El Diablo, anechoic response on tweeter axis at 50", averaged across 30° horizontal window and corrected for microphone response, with the nearfield responses of the midrange unit (green), ports (blue), and woofers (red) plotted below 1kHz, 1kHz, and 2kHz, respectively, along with their complex sum plotted below 300Hz (black).
The woofers (fig.3, blue trace) cross over to the midrange unit at around 300Hz, and are well behaved above their passband, rolling off with what appears to be a 12dB/octave slope. The Diablo extends quite low in frequency: –6dB at 30Hz in this graph (the slight boost between 45 and 400Hz is probably due to the nearfield measurement technique). The midrange rolls in with a shallow slope and has a shallow suckout in its farfield output around 1200Hz. The overall response is otherwise smooth and even through the bass and midrange, up to the crossover to the tweeter, where there is a sharp discontinuity in the speaker's output followed by a slightly shelved-down high treble. As MF noted in his auditioning, El Diablo's "personality" is on the slightly mellow side, which I also felt to be the case.
Michael mentions the danger of taking a relatively large-diameter midrange unit too high in frequency, mainly due to the fact that it will "beam" where its diameter approaches the wavelengths of the frequencies it is emitting. Yet looking at Peak Consult's plot of lateral dispersion (fig.4), it can be seen that the Diablo's output off axis is relatively uniform below 3kHz. However, there is a distinct step in its radiation pattern just above 3kHz, and it's possible that this, rather than the port resonance, is responsible for the slight coloration I heard. The tweeter is slightly more directional in its passband than is usual for a 1" dome, which will add to the speaker's mellow character in all but very small rooms.
Fig.4 Peak Consult El Diablo, 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.
In the vertical plane (fig.5), the Diablo has a sharp suckout centered on 5.6kHz for the listening axis above the tweeter, which suggests that the actual crossover between the midrange unit and tweeter occurs here rather than at the specified 4.8kHz. To provide the flattest perceived treble balance, the Diablos should be auditioned with the listener's ears on a level between the tweeter and the midrange unit. Fig.6 is the Diablo's in-room response, averaged across a grid centered on the position of MF's ears in his listening chair. Other than residual room effects that have not been eliminated by the spatial averaging, the pair of speakers produce a remarkably even and smooth balance at the listening seat from the upper bass through the mid-treble. The region covered by the tweeter definitely shelves down more than would be expected from the usual increase in the room's sound absorption in this region.
Fig.5 Peak Consult El Diablo, vertical-response family at 50", normalized to response on tweeter axis, from back to front: differences in response 15–5° above axis, reference response, differences in response 5–10° below axis.
Fig.6 Peak Consult El Diablo, spatially averaged, 1/6-octave response in MF's listening room.
In the time domain, the Diablo's step response on the tweeter axis (fig.7) indicates that its tweeter and woofers are connected in positive acoustic polarity, its midrange unit in inverted polarity, this confirmed by looking at the step responses of the individual units (not shown). However, the fact that the step of each driver smoothly hands over to that of the next lower in frequency correlates with the good frequency-domain integration seen in fig.3. The relatively small height of the tweeter's step (the initial spike in fig.7) ties in with this unit's more than usually restricted bandwidth.
Fig.7 Peak Consult El Diablo, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).
The Peak Consult's farfield cumulative spectral-decay plot (fig.8) is generally clean, but is marred by a ridge of delayed energy at the frequency of the on-axis response discontinuity. This is probably a cone phenomenon in the midrange unit, and could also be associated with the slight "cupped hands" coloration MF and I noted in our auditioning, though it is fair to point out that he felt this could be accommodated to relatively quickly.
Fig.8 Peak Consult El Diablo, cumulative spectral-decay plot at 50" (0.15ms risetime).
The Peak Consult El Diablo's measured performance is very respectable, though its balance appears to have been tailored to achieve a specific end result. It also looks drop-dead gorgeous.—John Atkinson