Eminent Technology LFT-16 loudspeaker Measurements

Sidebar 3: Measurements

The very affordable, stand-mounted LFT-16 costs $950/pair and combines a conventional moving-coil woofer with a flat, magnetically driven planar midrange unit and tweeter, and impressed the heck out of JM: "There is much to admire and to enjoy in this idiosyncratically charming hybrid loudspeaker," he wrote. "The LFT-16 has more clarity in the midrange and treble [than the slightly more expensive Renaissance MLP-403.5], and overall a brighter, faster, slightly lighter sound."

The tweeter panel is to the side of the midrange panel, and both are mounted on an open subbaffle above the sealed woofer enclosure. This will affect the speaker's horizontal dispersion, so I measured the LFT-16 on the tweeter axis.

The Eminent Technology LFT-16's voltage sensitivity on the tweeter axis, with the grille removed, was both lower than specified and very low in asbolute terms, at 80.5dB(B)/2.83V/m. On the face of things, this speaker will need a powerful amplifier to play at acceptably loud levels. The speaker's impedance magnitude and electrical phase are shown in fig.1. The single peak centered on 50Hz suggests that this is the tuning frequency of the sealed enclosure. While the minimum value of 4 ohms at 150Hz, and the combination of 5 ohms magnitude and –45° phase angle at 95Hz, mean that the speaker will need to be partnered with a good amplifier designed to work well into 4 ohms, the much higher impedance in the treble means that the LFT-16 will sound tilted-up at high frequencies with tube amplifiers having a high source impedance.

Fig.1 Eminent Technology LFT-16, electrical impedance (solid) and phase (dashed). (2 ohms/vertical div.)

The traces in fig.1 are free from the small discontinuities that would imply the existence of enclosure resonances of various kinds. However, JM did write that when he played music with generous bass content at levels moderately though not extremely loud, he could feel "resonances on both the bass enclosure and, more strongly, on the baffle." Fig.2 is a cumulative spectral-decay plot calculated from the output of an accelerometer fastened to the center of the woofer enclosure's side panel. Two very strong resonant modes can be seen, at 100 and 345Hz, with a third, weaker mode around 150Hz; both of the first two resonances could be detected on all surfaces. JM did write that he found the LFT-16's bass "at times almost too big and at others a bit tubby." I'm sure the 100Hz resonance contributed to that perception.

Fig.2 Eminent Technology LFT-16, cumulative spectral-decay plot calculated from output of accelerometer fastened to center of side panel (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz).

Fig.3 shows the individual responses of the LFT-16's three drive-units, all measured in the farfield with the exception of the woofer' s output below 350Hz, which was measured in the nearfield; ie, with the microphone capsule almost touching the cone. The woofer's response (blue trace) peaks up significantly between 70 and 200Hz, and while half of this boost will be due to the nearfield measurement condition, it does look as if the Eminent Technology's woofer alignment is underdamped, which will also lead to the impression of low-frequency tubbiness.

Fig.3 Eminent Technology LFT-16, acoustic crossover on tweeter axis at 50" without front grille, corrected for microphone response, with nearfield response of woofer (blue) plotted below 350Hz.

Higher in frequency in fig.3, the woofer's output (blue trace) is pretty flat in its passband, but there is a significant overlap between its upper-midrange output and that of the midrange panel (red trace). The midrange unit is very flat within its passband, but with a notch in its response just before it crosses over to the tweeter panel at 7.5kHz. The tweeter output (with the level control set to High, green trace) is very flat within its passband. Other than the midrange unit's high-pass rollout—which is second-order, 12dB/octave in nature—the crossover filters all appear to be higher-order types, with approximately 18dB/octave slopes.

Fig.4 shows how these individual outputs sum in the farfield, averaged across a 30° horizontal window on the tweeter axis. The LFT-16's response is superbly flat from 1kHz up, but there's no escaping that horrendous, two-octave-wide suckout in the midrange. This suckout explains the low estimated voltage sensitivity on the tweeter axis, but peculiarly, JM didn't comment on any hollowness or lack of body in the speaker's balance. I jumped to the obvious conclusion: that when I'd hooked up the LFT-16, I had connected its woofer with the wrong electrical polarity, and this, with the large degree of overlap of the woofer and midrange outputs in the midrange, had caused the suckout.

Fig.4 Eminent Technology LFT-16, anechoic response on tweeter axis at 50" with woofer in manufacturer-recommended connection, averaged across 30° horizontal window and corrected for microphone response, with woofer nearfield response plotted below 300Hz.

I therefore inverted the woofer connection (all drive-unit terminals are accessible behind the LFT-16's flat upper-frequency baffle) and repeated the response measurements. A large peak now appeared in the midrange (fig.5), which means that this is also not the correct woofer polarity. I checked the other speaker of the pair, and it appears that the manufacturer-recommended connection is that used to generate fig.4, which produces the on-axis suckout. A quick check of the ET's in-room response with pink noise (not shown) didn't show the midrange suckout, meaning that the missing on-axis energy did exist somewhere in the room.

Fig.5 Eminent Technology LFT-16, anechoic response on tweeter axis at 50" with woofer in opposite-polarity connection, averaged across 30° horizontal window and corrected for microphone response, with woofer nearfield response plotted below 300Hz.

So I turned to the ET's dispersion measurements. Figs. 6 and 7 show the speaker's horizontal dispersion, referenced to the response on the tweeter axis. (Fig.6 shows the actual off-axis responses; fig.7 normalizes the behavior off-axis to that on the tweeter axis; in both cases, the behavior on the tweeter side of the baffle is shown to the graph's rear.) The on-axis suckout does tend to fill in to the speaker's sides, as it does below the tweeter axis (fig.8), which is why the reverberant energy in the listening room is smoothly balanced through the midrange, and why JM was not bothered by the missing midrange energy in the first-arrival sound. (Though I suspect that it does contribute to his finding the LFT-16's balance "slightly light.")

Fig.6 Eminent Technology LFT-16, lateral response family at 50", from back to front: response 90–5° off axis on tweeter side of baffle, reference response on HF axis, response 5–90° off axis on midrange side of baffle.

Fig.7 Eminent Technology LFT-16, lateral response family at 50", normalized to response on tweeter axis, from back to front: differences in response 90–5° off axis on tweeter side of baffle, reference response, differences in response 5–90° off axis on midrange side of baffle.

Fig.8 Eminent Technology LFT-16, vertical response family at 50", from back to front: response 45–5° above axis, reference response on HF axis, response 5–45° below axis.

The LFT-16 offers three level settings for its tweeter; JM preferred the Low (–6dB) to the Mid (–3dB), never the High (0dB). "A hotter tweeter setting," he wrote, "seemed to make the sound less coherent and less timbrally accurate." Fig.9 shows the effects of the Low and Mid settings, subtracting the response with the High setting from both to reveal only the changes in output. The Mid setting drops the output at 20kHz by the expected 3dB, which will be mildly audible; the Low setting drops it by closer to 7dB, which might be perceived as a lack of top-octave air.

Fig.9 Eminent Technology LFT-16, effect of tweeter level control on HF-axis response set (from top to bottom) to Mid, Low (2dB/vertical div.).

Turning to the time domain, the LFT-16's step response (fig.10) is instructive. All three drivers are connected with positive acoustic polarity. The small up/down spike at the 3.6ms mark is the tweeter's output; the decay of its step feeds smoothly into the initial rise of the midrange unit's step, which is what is required for good frequency-domain integration of their outputs. However, because the flat upper-range units are mounted on an open subbaffle at the front of the woofer enclosure, the woofer's step (the slow rise in output after 4.3ms) follows the midrange unit's too late for optimal integration. Even though both are connected with positive polarity, the relative time delay of the woofer's output means that the rise of its step works against the negative-going decay of the midrange unit's step response. The result in the frequency domain is the suckout in the region where their responses overlap.

Fig.10 Eminent Technology LFT-16, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).

The Eminent Technology's farfield cumulative spectral-decay plot (fig.11) shows that there is delayed energy associated with the on-axis suckout, and that there is also a ridge of resonant energy in the midrange unit's output just below its crossover to the tweeter.

Fig.11 Eminent Technology LFT-16, cumulative spectral-decay plot on tweeter axis at 50" (0.15ms risetime).

Although the Eminent Technology LFT-16 is affordably priced and beautifully finished, there is no doubt in my mind that the woofer/midrange crossover is suboptimal, in that the woofer's output extends too high in frequency and the crossover topology does not take into account the fact that the woofer's acoustic center is farther away from the listener than that of the flat midrange unit. I also feel that the decision to mount the two planar units side by side rather than with the tweeter above the midrange units leads to unnecessary complication in setup and use. John Marks may have been impressed by the LFT-16's overall sound but I can't help thinking that a better-performing speaker could be designed using the same elements. Audition with more than the usual care.—John Atkinson

Eminent Technology Inc.
225 East Palmer Street
Tallahassee, FL 32301
(850) 575-5655