Solid State Power Amp Reviews

Usually, when I measure an amplification or digital product, I carry the box down the stairs to my basement, where I have my listening room and test lab. However, the Accustic Arts AMP V was too heavy for me to do this. Instead, I rolled it in its wheeled flightcase into the vestibule outside the listening room, removed the flightcase cover, and left the amplifier sitting in the flightcase (footnote 1) at the top of the small flight of stairs. I then set up my Audio Precision SYS2722 system (see the January 2008 As We See It) with its dedicated PC, my analog and digital oscilloscopes, test loads, etc, at the foot of the stairs. The amplifier's two AC leads were plugged into the closest wall socket, along with those for the test system. (This is to avoid ground loops.)

Before I performed the testing, I preconditioned the amplifier by running it at one-eighth its specified power into 8 ohms for 30 minutes, as recommended by the CEA. After the preconditioning, the heatsinks on the amplifier's sides were hot, at 113.4°F (45.3°C). The top panel was warm, at 105.0°F (40.6°C).

Soon after I began the testing—after the preconditioning period—the 15A circuit breaker serving the wall socket into which everything was plugged tripped. (At the time I was measuring distortion at a high power into 4 ohms.) I therefore ran two long extension cords from the 20A wall sockets in the test lab, one for the amplifier, the other for the test gear, and started where I had left off.

The AMP V's specified voltage gain is 30dB. I measured 30.9dB into 8 ohms for the balanced inputs and 30.75dB for the single-ended inputs. All the inputs and outputs preserved absolute polarity, ie, were noninverting. The input impedance was lower than the specified values: 10k ohms balanced, across the audioband, rather than the specified 32k ohms; the unbalanced input impedance was 6.7k ohms compared with the specified 15k ohms. Still, both values are high enough that compatibility problems are unlikely. With the Damping Factor correction switched off, the output impedance (including the series impedance of 6' of spaced-pair cable) was low, at 0.14 ohm from 20Hz to 20kHz. Turning on the Damping Factor increased the output impedance slightly, to 0.2 ohm. Even so, the response with our standard simulated loudspeaker varied by less than ±0.2dB (fig.1, gray trace). The response into 8 ohms (blue and red traces) was down by 1.5dB at 200kHz, and the AMP V's reproduction of a 10kHz squarewave (fig.2) featured short risetimes with no overshoot or ringing. Figs.1 and 2 were taken with the balanced inputs. The frequency and squarewave responses were the same for the unbalanced inputs and with Damping Factor on and off.

Likely as a result of its dual-mono construction, the amplifier's channel separation was a superb 110dB in both directions below 1kHz, falling to 80dB at the top of the audioband, presumably due to capacitive coupling between channels. Measured with the single-ended inputs shorted to ground, the Accustic's wideband, unweighted signal/noise ratio was a good 72.3dB ref. 1W into 8 ohms (average of both channels). This ratio improved to 80.5dB when the measurement bandwidth was restricted to 22Hz–22kHz, and to 83.3dB when the reading was A-weighted. The spectra shown in fig.4 were taken at 1Wpc into 8 ohms. Spuriae can be seen at 60Hz and its odd-order harmonics, which will be due to magnetic coupling from the two massive power transformers. However, these spuriae are very low in level.

I examined how the percentage of THD+N varied with output power, with clipping defined as when the THD+noise in the amplifier's output reached 1%. With both channels driven into 8 ohms, the AMP V didn't meet its specified maximum power of 900Wpc (29.54dBW) at 1% THD+N, clipping at 665Wpc (28.23dBW, fig.4); note, however, that I don't hold the wall voltage constant during the testing, and the wall voltage dropped from 119.2V with the amplifier idling to 114.2V with both channels of the amplifier clipping into 8 ohms. Similarly, into 4 ohms, again with both channels driven, I measured clipping power of 890Wpc (26.5dBW, fig.5); the specified value is 1360Wpc (28.325dBW). This time, the wall voltage dropped to 111.5V.

As I always do, I saved the measurement of the AMP V's clipping power into 2 ohms until the very end. However, when I performed my penultimate test, which was to measure how THD+N varied with frequency at 20V (equivalent to 50W into 8 ohms, 100W into 4 ohms, and 200W into 2 ohms), the amplifier turned itself off after 10 seconds of driving 2 ohms at this level (fig.6, gray trace). I had assumed that this would be due to protection circuitry doing its job; however, after I removed the power cables, let the amplifier cool down, and plugged it back in, it wouldn't turn on again. Each of the two AC power sockets has two fuses; all four were okay. The AMP V also has internal fuses, which must have gone open-circuit. I removed the 10 bolts that hold the top panel in place, but I wasn't able to remove it. I tried everything I could think of that did not present a substantial risk of damaging the finish.

Returning to the measurements: Fig.6 shows that the distortion is very low over most of the audioband into both 8 and 4 ohms, though a little higher in the right channel (red and magenta traces) than the left (blue, cyan traces). The THD rises in the top octave—this will be due to a restricted open-loop bandwidth reducing the amount of negative feedback available—but it remains below 0.17% at 20kHz into 4 ohms.

The waveform of the THD+N at 100Wpc into 8 ohms with the 1kHz fundamental notched out (fig.7) suggests that the distortion signature consists primarily of low-order harmonics. The spectrum of the AMP V's output with a 50Hz signal at that level into 8 ohms (fig.8) shows that the second harmonic lay at –106dB (0.0005%) in both channels, the third harmonic at –96dB in the left channel (0.0015%, blue trace) and at –88dB in the right channel (0.004%, red trace). Supply-related spuriae are visible, but all lie at or below –96dB. Fig.9 shows the spectrum of the AMP V's output as it drove an equal mix of 19 and 20kHz tones into 8 ohms at a peak level of 100Wpc. The second-order difference product at 1kHz lies at a negligible –100dB (0.001%), and the higher-order intermodulation products are all at or below –86dB (0.005%).

The Accustic Arts AMP V offers extremely high power coupled with extremely low distortion. But why did those internal fuses blow—assuming that's what happened—after a short while driving 200Wpc into 2 ohms with high-frequency test tones? I don't know what caused it, but I can confidently say that this is unlikely to happen with a music signal.—John Atkinson