Integrated Amp Reviews

As usual, I preconditioned the NAD M3 by running it at one-third its specified power into 8 ohms for one hour with both channels driven. While the measured percentage of THD+noise remained constant throughout this period, at a low 0.0035%, the chassis got almost too hot to touch, and the side-mounted heatsinks definitely so. Measured at its speaker terminals, the M3 offered a maximum voltage gain of 39.2dB into 8 ohms for both balanced and unbalanced input signals. With the volume control set to "0.0dB," the overall voltage gain was 29.2dB. This was also the gain of the power-amplifier section on its own. Looking at the preamplifier section on its own, this offered a maximum gain at the preamp outlet jacks of 10dB, as expected, with the unity-gain setting of the volume control "0.0dB," also as expected. All inputs and outputs preserved absolute polarity.

The M3's input impedance was high, at well over 100k ohms unbalanced and twice as high balanced. The power-amplifier input impedance was much lower, at 22k ohms across most of the audioband, dropping slightly to 14.5k ohms at 20kHz. The preamplifier section's output impedance was a low 100 ohms across the band; the power-amplifier section's output section was a low 0.08 ohm at low and midrange frequencies for both channels, rising inconsequentially to 0.11 ohm at 20kHz.

With its low source impedance, the M3 offered only ±0.1dB of response variation with our simulated loudspeaker load, and only minor reductions in level into lower impedances (fig.1). The small-signal bandwidth was wide, with a –3dB point of 180kHz, this higher than the specified 80kHz. As a result, the M3's reproduction of a 10kHz squarewave demonstrated very short risetimes, with a complete absence of overshoot and ringing (fig.2).

Fig.1 NAD Master Series M3, volume control at "0.0dB," frequency response at 2.83V into (from top to bottom at 2kHz): simulated loudspeaker load, 8, 4, 2 ohms (0.5dB/vertical div., right channel dashed).

Fig.2 NAD Master Series M3, small-signal 10kHz squarewave into 8 ohms.

Measuring the effect of the Tone controls was initially problematic: Instead of the specified ±5dB of response variation, I obtained just a tenth of that figure. It turned out that the effect of these controls is affected by the volume-control setting. At "–10.0dB," you get the full effect of the controls when set to their maximum and minimum positions (fig.3, top and bottom pairs of traces). But with the control set to "0.0dB," the maximum boost and cut drops to 0.5dB (fig.3, middle pairs of traces). And with the volume control set to its maximum, "+10.0dB," the Tone controls have no effect at all. The situation was similar with the Tilt control, though the maximum amount of response modification is ±3dB. Contrary to what I expected, setting the Tilt positive, to "+3dB," rolled off the highs and boosted the bass, and vice versa (fig.4).

Fig.3 NAD Master Series M3, effect of tone controls set to maximum and minimum positions with volume control set to "0.0dB" (middle traces) and to "–10.0dB" (top and bottom traces). (Right channel dashed, 0.5dB/vertical div.)

Fig.4 NAD Master Series M3, effect of Tilt control set to maximum (top below 500Hz) and minimum (top above 500Hz) positions (right channel dashed, 0.5dB/vertical div.).

Channel separation (not shown) was better than 100dB in the treble for the preamplifier section alone, 95dB below 1kHz for the power-amplifier section alone. The power amplifier's noise levels were very low, at 86.3dB, wideband, unweighted, ref. 2.83V into 8 ohms, with the input short-circuited, this increasing to 106dB when A-weighted.

Plotting the THD+noise percentage against output power with both channels driven gave a somewhat surprising result, as I obtained the same power at 1% THD+N into 4 ohms as I did into 8 ohms: 235W (fig.5). This does conform with the M3's specification, however. With one channel driven, I obtained 335W into 2 ohms. Note the low levels of distortion prior to clipping. Fig.6 plots the percentage of THD+N at 10V RMS into 8, 4, and 2 ohms. While THD does increase slightly into lower impedances and in the top audio octave, it remains low at all frequencies, and its content is heavily second harmonic (fig.7).

Fig.5 NAD Master Series M3, distortion (%)vs 1kHz continuous output power into (from bottom to top at 100W): 8 and 4 ohms, both channels driven; and 2 ohms, one channel driven.

Fig.6 NAD Master Series M3, THD+N (%)vs frequency at 10V into (from bottom to top): 8, 4, 2 ohms (right channel dashed).

Fig.7 NAD Master Series M3, 1kHz waveform at 19W into 4 ohms (top), 0.00755% THD+N; distortion and noise waveform with fundamental notched out (bottom, not to scale).

The third harmonic makes an appearance at high powers into 8 ohms (fig.8) and 4 ohms (fig.9), but its level remains very low in absolute terms. And even at a level close to waveform clipping into 4 ohms, intermodulation distortion was virtually absent (fig.10).

Fig.8 NAD Master Series M3, spectrum of 50Hz sinewave, DC–1kHz, at 132W into 8 ohms (linear frequency scale).

Fig.9 NAD Master Series M3, spectrum of 50Hz sinewave, DC–1kHz, at 200W into 4 ohms (linear frequency scale).

Fig.10 NAD Master Series M3, HF intermodulation spectrum, DC–24kHz, 19+20kHz at 200W peak into 4 ohms (linear frequency scale).

The NAD M3 offers excellent measured performance, and in my auditioning I was struck by how smooth it sounded: "as smooth as silk," according to my listening notes.—John Atkinson