Morel Octwin 5.2M loudspeaker Measurements
The Octwin 5.2M's voltage sensitivity was just below average, at 86.5dB(B)/2.83V/m. (A single Octave 5.2 will be 3dB lower than this, of course.) And with its paralleled drive-units, the dual speaker's impedance (fig.1) will be half that of a single unit. It remains at or below 4 ohms for most of the audioband, with a minimum magnitude of 3.38 ohms at 220Hz. However, the fact that the electrical phase angle is very low over almost all this range will ameliorate any drive difficulty. The traces in fig.1 are free from the wrinkles that would imply the existence of cabinet resonances. However, I did find low-level modes present at 156Hz and in the 600Hz region (fig.2).
Fig.1 Morel Octwin 5.2M, electrical impedance (solid) and phase (dashed). (2 ohms/vertical div.)
Fig.2 Morel Octwin 5.2M, 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 saddle centered on 50Hz in the impedance magnitude trace in fig.1 suggests that this is the tuning frequency of the triangular ports, which is confirmed by the nearfield traces in fig.3 (red is the response of the woofers, blue that of the ports). The upper-bass region in this graph is shelved up by 5-6dB. This is mainly an illusion, however, resulting from the nearfield measurement technique.
Fig.3 Morel Octwin 5.2M, anechoic response on tweeter axis at 50", averaged across 30 degrees horizontal window and corrected for microphone response, with the nearfield responses of the woofer and port and their complex sum plotted below 300Hz, 1kHz, and 300Hz, respectively.
Moving higher in frequency, the peak at just above 600Hz in the output of the ports alarmed me (footnote 1), as did the strong peak in the speaker's farfield response at the same frequency. With my ear pressed up to one of the rear-panel ports, I could easily hear that pink noise was colored by this peak, which is presumably due to some sort of internal air-space resonance. However, I was surprised to find that the peak was fairly difficult to hear from the listening chair on many kinds of music, even after I had looked at the measurements and knew it existed. So much for self-fulfilling prophecies. The Octwin's response is basically flat In the treble, though shelved-down by a couple of dB compared with the midrange and below. Presumably, this contributes to the rather warm balance I heard in my auditioning.
The Morel 5.2M's lateral dispersion is basically wide and even, though some small off-axis ridges and troughs can be seen (fig.4). These are presumably due to interference between the vertically separated woofers. In the vertical plane, large suckouts develop when the listener is well above or below the tweeters (fig.5). The double-speaker arrangement makes listening axis quite critical: the listener needs to sit with his ears between 30" and 33" from the floor, which is a little low for someone of normal height sitting in a typical sofa or easy chair. And forget director's chairs, which place people's ears around 43" from the floor.
Fig.4 Morel Octwin 5.2M, lateral response family at 50", normalized to response on central 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 Morel Octwin 5.2M, vertical response family at 50", normalized to response on central tweeter axis, from back to front: differences in response 20 degrees-5 degrees above axis, reference response, differences in response 5 degrees-10 degrees below axis.
Fig.6 shows the Octwin 5.2M's in-room response, averaged across a window approximately 4' wide by 18" high and centered on my ear position in my listening chair. The expected prominence in the 630Hz 1/3-octave band is evident, as is a relative lack of energy in the subsequent octave, but the overall response is fairly well-balanced. The dual tweeters' restricted dispersion above 10kHz will lend the speaker's tonal balance a mellow quality in all but very small rooms. The tweeters' wide lateral dispersion at the base of their passband, however, compensates for the shelved-down on-axis behavior in the same region, though overall, the 5.2M's output through the upper bass to the middle of the midrange is still a little higher in level than the treble. The measured low frequencies extend in-room down to the 32Hz band, confirming the good bass weight I heard on classical piano.
Fig.6 Morel Octwin 5.2M, spatially averaged, 1/3-octave response in JA's listening room.
In the time domain, the Morel's impulse response (not shown) was disturbed by some high-frequency ringing and a strong reflection about 2ms after the initial arrival. This reflection is also evident in the speaker's step response (fig.7); its timing suggests it comes either from the back panel of the speaker (most likely) or from the "cup" on the intermediate stand (much less likely). Either way, while it arrives at the ear much too soon to be perceived as a discrete arrival, it does give rise to the ripples seen in the on-axis frequency response (fig.3).
Fig.7 Morel Octwin 5.2M, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).
As well as introducing ripples, this reflection also interferes with the speaker's cumulative spectral-decay plot, obscuring the decay of the driving impulse with low-level treble hash (fig.8). Windowing out the reflection drastically curtails the resulting plot's midrange resolution (fig.9), but now the speaker's true decay appears very clean, with only a very-low-level mode visible just under 12kHz, this correlating with the ripples seen in the impulse response. The clean treble decay correlates nicely with what I heard in my auditioning.
Fig.8 Morel Octwin 5.2M, cumulative spectral-decay plot at 50" (0.15ms risetime).
Fig.9 Morel Octwin 5.2M, cumulative spectral-decay plot at 50" (0.15ms risetime, 2ms time window).
Though there is much to admire in the Morel Octwin 5.2M's measured performance, I was bothered both by the 600Hz peak in its response and the strong early reflection. Though neither impinged on my auditioning to any great extent, given my druthers I would much rather they didn't exist in a speaker at this price.—John Atkinson
Footnote 1: Using the Aurora MLSSA plug-in for Cool Edit Pro, I have prepared a 32kHz WAV file of the impulse response from which this waterfall plot was calculated; the 600Hz resonance is clearly audible. If anyone is interested in hearing it, e-mail me with "Port Impulse Response" in your message's subject field or hit play on the embedded file below.—John Atkinson