NHS XdS Measurements and Analysis
Figure 1a & b: Impedance & Phase. At Left: XdS midrange (blue) and tweeter (red). At right: XdW
A series of measurements were undertaken to develop an objective assessment of the Xd system.
In figure 1a & b we see the impedance scans derived from measuring the various drivers. These measurements were done with the drivers in their respective cabinets. At left are the midrange (blue) and tweeter (red) impedance curves for one of the XdS. Both are clear of any of the sort of anomalies that would arise owing to mechanical resonances and so forth. The design of the cabinet, the mechanical properties of the materials used in its construction, as well as the substantial amount of stuffing packed into the cabinet’s internal volume all contribute to this particularly clean looking plot. (As a side note, the slight glitches that show, for example, at 100 Hz, 1kHz, and 10kHz on the tweeters impedance curve are measurement artifacts generated by the LMS card that was used in capturing the data and are not symptoms of any sort of underlying driver/system pathology). The blue curve seen at left is typical for a totally enclosed cabinet, with its single Z-peak seen at left. The peak itself, appearing as it does at about 72 Hz, indicates the system is tuned lower than the high-pass knee of XdS’ intended acoustical response, thus helping to keep cone excursion within well controlled limits The tweeter’s curve (red) indicates an in-cabinet resonance frequency of about 1.6 kHz and the general appearance of the curves that lie just to the left & right of the peak are typical for fluid-cooled tweeters, such as those used in the XdS. Like its midrange counterpart, it too sports an in-cabinet Z-peak lower than the high-pass knee of its intended acoustical response, once again helping to keep cone excursion within well controlled limits, in turn helping keep distortion down as well. At the upper end of the tweeter’s z-curve, in the 27 – 28 kHz region, are a few barely discernible glitches indicating the tweeter’s metal cone entering its primary breakup mode.
The green curve seen at right is the impedance curve for the XdW. For this measurement, the built in power amp was removed from the XdW, exposing the leads connected to the two 10” drivers. As before, impedance data were taken with the drivers in place and in the cabinet. The curve presented here, like that showing for the XdS’ midrange driver, is typical for that of a system featuring a totally enclosed box. The location of the peak indicates the system is tuned to about 55 Hz. The only noticeable glitch is seen at ~ 250 Hz. Whatever the underlying cause, there was no parallel anomaly seen in the amplitude response plot of the XdW.
Figure 2: System response, dB spl. Dark blue: XdW; Black: XdS combined tweeter (red) & Midrange (blue) Response. 1/12th Octave smoothing employed for improved visual clarity
In figure 2 the system amplitude response plot curves, in dB spl, are shown for the XdS tweeter (red), midrange driver (blue), their combined response (black) and that of the XdW (dark blue). A variety of measurement approaches were employed and the results were scaled to 1m. The XdA was driven by the LMS card at a voltage just above that needed to produce amplitude response plots consistent with the sensitivity of the system, as determined by deriving the midrange driver’s Thiele/Small parameters. Another words, the plots are a dB or two above actual system sensitivity.
Of particular note here are the remarkably steep (acoustic) crossover slopes seen in Figure 2 as well as the tweeter’s response affected by the metal dome entering its first breakup mode above 20 kHz. The general appearance of the system’s amplitude response in the ~ 1 kHz to 10 kHz range illustrates the electrostatic-like subjective impression mentioned earlier, made by the XdS system. The XdW bass module supplied for this review has useful response down to around 25 Hz and consistently surprised and impressed with the quality of its performance.
Figure 3a, b: A: XdA Out; B: XdW Amp Out
Curious to see just what the XdA’s output looked like, the voltage vs. frequency curves seen in Figure 3a were generated. At left (in green) in Figure 3a is the XdA balanced out to XdW response. The collection of curves seen in the middle is the XdA’s output to the XdS midrange driver as presented when each of the 4 output modes are activated. The collection of curves showing at right in Figure 3a is, of course, for the XdS tweeter. Note the minimal effect seen by switching modes on the tweeter’s magnitude response plots. Taking into account the complexity of the individual curves its easy to appreciate the complexity of the task handed the XdA’s DSP functions.
Curious to see what, if any, role in contouring the XdW’s amplitude response plot was played by the XdW’s built-in power amplifier, the voltage vs. frequency curves seen at right in Figure 3b were generated. The frequency sweeps were done at the “More” (red), “Just Right” (blue) and “Less (green) levels. As can be seen, there is indeed some low-pass contouring going on, varying with however the XdW’s trim is set.
