NHT SuperZero loudspeaker & SW2 subwoofer Measurements part 2
Fig.10 NHT SuperZero, cumulative spectral-decay plot of accelerometer output fastened to back of enclosure above terminal panel. (MLS driving voltage to speaker, 7.55V; measurement bandwidth 2kHz.)
Moving on to the SW2 subwoofer, fig.11 shows the impedance plot with the subwoofer's jumpers removed for use with the MA-1 crossover amplifier. With minima a hair over 4 ohms, the subwoofer will not be too difficult to drive, though I feel that Corey's use of two SW2s in parallel, which will give a 2 ohm impedance in the upper bass, is asking a lot from the MA-1. (The MA-1's subwoofer outputs are marked "min impedance 6 ohm.") The port tuning is revealed by the broad saddle in the magnitude trace centered on 25Hz. When the internal passive crossover is used, the SW2 becomes a more demanding load—fig.12, made with the high-pass outlets open-circuit—and its minimum impedance now drops to 2 ohms. With the passive low-pass filter, the SW2's overall output is a bandpass centered on 50-60Hz, as can be seen from fig.13, which also shows the individual nearfield outputs of the port and woofer. The -6dB point of their combined output lies at 35Hz, which will be extended in-room.
Fig.11 NHT SW2, active crossover mode (jumpers removed), electrical impedance (solid) and phase (dashed) (2 ohms/vertical div.).
Fig.12 NHT SW2, passive crossover mode (with jumpers), electrical impedance (solid) and phase (dashed) (2 ohms/vertical div.).
Fig.13 NHT SW2, low-pass filter in-circuit, individual nearfield responses of woofer (curve with minimum at 23Hz) and port (bottom curve at 60Hz), with complex sum of nearfield responses (top).
The NHT MA-1 amplifier's low-pass crossover functions, implemented using 4558 dual op-amp chips, are shown in fig.14. The flat curve is with the unit in "Bypass" mode. The passive high-pass feed from the MA-1 to the satellite amplifier has a -3dB point of 120Hz and a rollout slope of 6dB/octave, while the three low-pass settings for the subwoofer have -3dB points of 51Hz, 82Hz, and 110Hz, each with an ultimate slope of 12dB/octave.
Fig.14 NHT MA-1, Bypass, high-pass, and low-pass frequency responses (2dB/vertical div.).
If there's one easy job for an amplifier to do, it's driving a subwoofer. With a bandwidth limited to three or four octaves at most, a low-pass filter that suppresses the higher-distortion harmonics, and a speaker whose impedance is generally benignly behaved, it might be thought that a subwoofer's designer could relax the specs a bit. Yet Ken Kantor's design appears to be a respectable little amp. (Though it did get very hot after the one-hour, 1/3-power preconditioning period.) It uses a single output pair of devices and a series inductor, yet its output impedance was low, at 0.09 ohms between 20Hz and 1kHz; its input impedance was a reasonable 22k ohms. The maximum gain in Bypass mode, with the volume control wide open and an 8 ohm load, measured a rather high 48.6dB. The unweighted, 22Hz-22kHz S/N ratio was 79dB (ref. 1W into 8 ohms), with some very low-level 60, 120, 180, and 240Hz power-supply harmonics noticeable on an FFT-derived spectrum (not shown).
Overall distortion was very low, generally lying below 0.01%. Though it did rise at higher frequencies, this will be of no concern to its subwoofer-driving owner. The distortion spectrum itself was dominated by third harmonic (fig.15), though "dominated" is perhaps hardly the correct term when this only reached -69.6dB with respect to the level of the 50Hz fundamental (0.03%). The fifth harmonic was the next highest in level, with the seventh lying below the second.
Fig.15 NHT MA-1, spectrum of 50Hz sinewave, DC-1kHz, at 80W into 4 ohms (linear frequency scale). Note that the third harmonic at 150Hz is the highest in level, 69.6dB below the level of the 50Hz fundamental (0.03%).
Finally, the manner in which the distortion/noise level varied with output power for a 117V AC source voltage is shown in fig.16. The amplifier gives its lowest distortion into 8 ohms (bottom trace), with the approximate clipping power into that load (defined as 1% THD+noise) being 72W (18.6dBW). Into 4 ohms, 96W was available (16.8dBW), and into 2 ohms, 110W (14.4dBW), though the amplifier's distortion into this low impedance was significantly higher than into 8 or 4 ohms. These figures are a little below spec, perhaps due to my use of a 100Hz test signal rather than the more traditional 1kHz. (I think that 100Hz is a more appropriate signal for a subwoofer amplifier.)
Fig.16 NHT MA-1, 100Hz distortion vs output power into (from bottom to top at 10W): 8 ohms, 4 ohms, and 2 ohms.
Regarding the sound of the MA-1/ SW2 combination: In absolute terms I would have liked a little more low-bass extension. It could also add a rather plodding, one-note signature to the sound unless you were painstakingly careful in choosing the woofer's placement, polarity, and crossover settings. (I'm always nervous about blanket recommendations for subwoofer/satellite systems, given the difficulty even experienced audiophiles have in achieving the optimal blend in-room between sub and sats.) This was much less noticeable with orchestral music than with rock, however, and—its slight treble wiriness apart—the combination of the SuperZeros and the SW2 reproduced large-scale classical music with aplomb.
I agree with CG that the NHT subwoofer system is an excellent value at $650. Using Y-adaptors with passive high-pass filters in series with the main amplifiers, I got a well-integrated, almost full-range sound that allowed the SuperZeros to sing in a most effective manner. At $880, the combination of the SuperZeros with the SW2/MA-1 is a shoo-in recommendation.—John Atkinson