JMlab Micron & Micron Carat loudspeaker 1996 Measurements

Sidebar 2: Measurements

I estimated the Micron Carat's B-weighted sensitivity to be around 87dB/2.83V/m, which is pretty good for such a small speaker. Its impedance traces (fig.1) show a drop to just below 4 ohms in the lower midrange, coupled with a moderately high phase angle in the region below. A good 4 ohm-rated amplifier will be called for, as there is quite a lot of musical energy in this region. But as I found, the single-ended Cary amplifier proved to be well-matched to the JMlab. The saddle in the amplitude trace at 60Hz reveals the port tuning: relatively high in frequency, as you might expect.

Fig.1 JMlab Micron Carat, electrical impedance (solid) and phase (dashed) (2 ohms/vertical div.).

The port's output is shown in fig.2—it's the bandpass output, with its maximum output frequency coincident with the woofer's minimum-motion point at 60Hz. Note that there are no midrange peaks present in its output. The woofer rolls off steeply above 3kHz, with a degree of raggedness present in the top octave of its passband. The tweeter also rolls off quite steeply below its operating region, but is quite flat overall, other than a degree of peakiness above 14kHz or so.

Fig.2 JMlab Micron Carat, acoustic crossover on tweeter axis at 50", corrected for microphone response, with nearfield woofer and port responses plotted below 300Hz.

The Micron Carat's overall response, averaged across a 30 degrees lateral window, is shown in fig.3. The highish-Q reflex tuning results in a bass that is flat down to the port tuning frequency but then dies rapidly away. The midrange is a little recessed, but the overall treble response trend is flat.

Fig.3 JMlab Micron Carat, anechoic response on tweeter axis at 50", averaged across 30 degrees horizontal window and corrected for microphone response, with complex sum of nearfield woofer and port responses plotted below 300Hz.

Even though the HF peakiness in fig.3 is above my hearing range, I did find the speaker's balance to be "wispy" in the highs. The lateral dispersion plot (fig.4) suggests that this is perhaps a function of the Carat's wide dispersion throughout the treble, with some slight energy peaks apparent off-axis at 5kHz and 7kHz in this graph, coupled with the restricted LF extension. But the speaker's horizontal dispersion is otherwise textbook, a function of the narrow baffle width and a well-engineered crossover.

Fig.4 JMlab Micron Carat, 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.

In the vertical plane (fig.5), though the Micron's balance is relatively unchanged as long as the listener keeps his ears on or below the tweeter axis, a big crossover-region suckout appears above the tweeter axis. This is the opposite to the B&W CDM 1: with this speaker, make sure you don't use stands that are too short.

Fig.5 JMlab Micron Carat, vertical response family at 50", normalized to response on tweeter axis, from back to front: differences in response 90 degrees-5 degrees above tweeter axis; reference response; differences in response 5 degrees-90 degrees below tweeter axis.

In my listening room, the spatially averaged, 1/3-octave response (fig.6), taken with the Mark Levinson amplifier, correlated well with my listening impressions. A slight excess of energy in the mid-treble probably gave rise to my feeling that the sound tilted-up in the highs when the speaker was driven by the solid-state amplifier. The tubed amplifier, with its high source impedance, brings the low treble up to better balance the mid-treble, as well as pulling the latter region down a little. The bass, as expected, drops quite rapidly below the port-tuning region.

Fig.6 JMlab Micron Carat, spatially averaged 1/3-octave response in JA's room.

In the time domain, the Carat's step response (fig.7) indicates that both drive-units are connected with positive acoustic polarity but that the tweeter's output leads the woofers by a quarter of a millisecond or so. The associated waterfall plot (fig.8) reveals a clean initial decay, but with a residual woofer-cone mode apparent at 3.1kHz. I didn't note any nasality in my auditioning, so I assume that this mode is low enough in level not to have much of a negative effect. There is some resonant behavior noticeable in the 1kHz region; I assume that these helped give rise to the upper-midrange congestion I noted in the treble at high playback levels.

Fig.7 JMlab Micron Carat, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).

Fig.8 JMlab Micron Carat, cumulative spectral-decay plot at 50" (0.15ms risetime).

Though no wrinkles were apparent in the Carat's impedance plot, a waterfall plot calculated from the output of a PVDF accelerometer strip fastened to the center of the cabinet sidewall (fig.9) did indicate the presence of a reasonably strong panel resonance at 390Hz. This was present on all the enclosure surfaces.

Fig.9 JMlab Micron Carat, cumulative spectral-decay plot of accelerometer output fastened to side of enclosure (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz).

One final note: While measuring one of the Micron Carats after the auditioning was all but finished, it accidentally fell from the 45" Outline turntable stand that I use. Other than a dented corner and a couple of cosmetic blemishes, the speaker still worked fine after its tumble. Obviously they make 'em tough in Saint-Etienne!—John Atkinson