GamuT RS7 loudspeaker Measurements

Sidebar 3: Measurements

I used DRA Labs' MLSSA system and a calibrated DPA 4006 microphone to measure the GamuT RS7's frequency response in the farfield, and an Earthworks QTC-40 for the nearfield and spatially averaged room responses. Because the RS7 is large and heavy, I wasn't able to lift it onto a stand for the measurements; the reflection of its output from the floor thus occurred earlier than is usually the case with my published loudspeaker measurements, reducing the resolution in the midrange.

The RS7's voltage sensitivity is specified as 88dB/2.83V/m; my estimate was less than that, at 85.5dB(B)/2.83V/m. The impedance is specified as 4 ohms, with a minimum magnitude of 2.7 ohms at 120Hz. Fig.1 shows my measurement of the GamuT's impedance magnitude (solid trace) and electrical phase angle (dotted). The impedance remains below 5 ohms through the upper bass and the entire midrange, with a minimum value of 2.66 ohms at 120Hz, this agreeing with the specification. Though the phase angle remains low throughout most of the audioband, this is still a speaker that will work best with amplifiers that have no trouble delivering enough current into low impedances.

Fig.1 GamuT RS7, electrical impedance (solid) and phase (dashed) (2 ohms/vertical div.).

The traces in fig.1 are free from any small discontinuities that might hint at the presence of cabinet vibrational resonances of some kind. Logistical difficulties prevented me from performing my usual accelerometer tests; however, listening to the enclosure panels with a stethoscope indicated that the cabinet was relatively inert, with only some low-level modes audible between 500 and 650Hz at points on the sidewalls level with the woofers.

The RS7's low-frequency behavior is complex, with significant overlap between the midrange unit and the two woofers. This can be seen in fig.2, which shows the sum of the nearfield responses of the woofers (blue trace), the sum of the nearfield port responses (red), and the nearfield response of the midrange unit (green), all plotted in the ratio of the square root of the radiating area. As suggested by the saddle centered on 20Hz in the impedance-magnitude trace, the ports are tuned to this low frequency, which is indicated by the minimum-motion notch in the woofers' output at that frequency. The output of the ports rolls off above 50Hz, but is disturbed by a major resonance peak at 605Hz. I could hear this resonance from behind the speaker with pink noise; fortunately, the ports face away from the listener, minimizing the resonance's audibility at the listening position.

Fig.2 GamuT RS7, nearfield responses of midrange unit (green), woofers (blue), ports (red).

The woofers' response peaks broadly in the midbass, this partly an artifact of the nearfield measurement technique, and crosses over to the midrange unit close to the specified 250Hz. However, the midrange unit's output continues for two octaves below the nominal crossover frequency. Its cone was undergoing significant excursions even at 20Hz.

The black trace below 300Hz in fig.3 shows how these nearfield responses sum at a nominal farfield point. Again, the broad peak in the midbass will be due to the nearfield measurement technique, and the speaker rolls off in the low bass a little earlier than the low tuning frequency of the ports had led me to expect. Above 300Hz, the black trace shows the farfield response averaged across a 30° horizontal window centered on the tweeter axis, but with the speaker's front baffle tilted back, as Benno Meldgaard had set it up in my listening room. The midrange is flat, as is the treble, though the top octaves are slightly boosted. However, there is a well-defined suckout centered close to the upper crossover frequency of 2500Hz. Looking at the individual outputs of the midrange unit (green trace) and tweeter (blue), it appears that one or both of these drivers is rolled off a little too early for their outputs to add optimally in the crossover region, at least on this axis.

Fig.3 GamuT RS7, anechoic response on tweeter axis at 50", averaged across 30° horizontal window and corrected for microphone response, with: farfield tweeter response (blue), farfield midrange-unit response (green), complex sum of nearfield responses plotted below 300Hz (black).

The RS7's plot of horizontal dispersion (fig.4) does suggest that the effect of this suckout decreases to the speaker's sides, which is probably why Meldgaard set them up toed in only 5° to the listening position. The contour lines in this graph are respectably uniform, while the fact that the ring-radiator tweeter becomes relatively directional above 10kHz—something I've found with other speakers using this drive-unit—will offset the excess of energy on axis in this region. In the vertical plane (fig.5), again measured with the front baffle tilted back, the crossover-region suckout deepens above the tweeter axis, which is 36" from the floor with the speaker on its spiked plinth. As I found in my listening, increasing the backtilt of the baffle or slouching in my seat gave a slightly more immediate-sounding, better-integrated balance.

Fig.4 GamuT RS7, 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.

Fig.5 GamuT RS7, vertical response family at 50", normalized to response on tweeter axis, from back to front: differences in response 20–5° above axis, reference response, differences in response 5–10° below axis.

According to Meldgaard, GamuT does its measurements at the seated listening position, between two speakers—not in front of just a single speaker. In my listening room, the RS7s' spatially averaged response—measured for each speaker individually in a rectangular grid 36" wide by 18" high and centered on the positions of my ears, using SMUGSoftware's FuzzMeasure 3.0 program and a 96kHz sample rate—is shown as the red trace in fig.6. For comparison, the blue trace in this graph is the in-room response of KEF's Blade Two, taken under identical conditions when I reviewed it for the June 2015 issue. Overall, the KEF's room response is more even from the upper bass through the top octaves, but it's fair to say that the GamuT's balance would have been close to the English speaker's behavior had it not been for slight lacks of energy in the RS7's two crossover regions. The lack between 1.5 and 3kHz is due to the crossover issue noted earlier; that between 150 and 600Hz is due to boundary interference effects, which are always difficult to minimize with a speaker having widely spaced lower-frequency drivers. (The KEF also had problems in this region, though milder.)

Fig.6 GamuT RS7, spatially averaged, 1/6-octave response in JA's listening room (red); and of KEF Blade Two (blue).

Both the GamuT and the KEF are boosted in the low bass by the lowest-frequency resonant mode in my room, but with its low port-tuning frequency, the GamuT has more output apparent below 30Hz. Listening to the RS7s while I performed the measurements suggested that the speakers don't handle infrasonic signals gracefully, the low port-tuning frequency meaning that the unloaded woofer cones undergo high excursions below 20Hz. This will not be a problem with digital sources, but it might make the RS7 oversensitive to warp wow signals from poorly optimized LP players. The individual measurements taken to generate the red trace in fig.6 revealed that while the two speakers' midrange units were well matched, the left RS7's tweeter was about 1dB higher in output than the right's.

Turning to the time domain, the RS7's step response on the tweeter axis (fig.7) indicates that all four drive-units are connected in positive acoustic polarity. The cumulative spectral-decay plot on the tweeter axis (fig.8) is impressively clean throughout the midrange and treble.

Fig.7 GamuT RS7, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).

Fig.8 GamuT RS7, cumulative spectral-decay plot on tweeter axis at 50" (0.15ms risetime).

The GamuT RS7's measured behavior is, in general, quite good: a clean waterfall plot, excellent low-frequency extension, well-organized lateral dispersion, and well-balanced in-room behavior. However, I must admit to being perplexed by the fact that the midrange unit doesn't appear to be high-pass filtered to any significant extent, and to being concerned about the less-than-optimal implementation of the upper-frequency crossover.—John Atkinson

GamuT Audio
US distributor: GamuT USA
(888) 252-2499

jmsent's picture

"The soft diaphragm is terminated with rubber roll surrounds not only at its circumference, as usual, but also at a central, stationary, sharp-pointed phase plug."

I don't understand this. Tweeters are generally not terminated with rubber rolls and this one is no exception. The picture shows a one piece coated textile dome assembly, just like every other ScanSpeak and Vifa ring radiator tweeter I've ever seen. Since the large rolls are designed to be the actual radiating surfaces , I think rubber would be the last material you'd want to use for this purpose.
As for the performance...I guess you're being "diplomatic" here..but I see a system with serious flaws in its design. The specs describe this as a 3 way speaker, but the individual driver curves reveal it to actually be a 2-1/2 way system. The so called "midrange driver"isn't really a midrange at all. It's a woofer/midrange, since it is being fed all the spectral energy right up to the tweeter crossover point, including all the bass. At least, the impedance curve gives no indication of an electrical crossover at 250Hz And what exactly is the deal with the huge response dip at the tweeter crossover? Did they run into the well known problem of using a ring radiator tweeter at too low a frequency? It's no secret that 2nd order distortion in this type of tweeter rises quickly under 2.5 kHz, which is why this tweeter performs best when crossed over at or above 3kHz. But then, there's the problem of trying to push a 7" bass/midrange beyond 3kHz. It seems to me this all could have been avoided by using a real midrange driver; e.g., 5" with a true bandpass filter, or choosing a different tweeter that could be crossed in at a lower frequency. To me, this speaker has major flaws, inexcusable at $39,000.

dcolak's picture

That can't be normal, specially for that kind of money?