Mark Levinson No.33H monoblock power amplifier:
Measurements part 2
At a higher 50W level into 8 ohms, the residual distortion (fig.4) was primarily third-harmonic; though at 0.0045%, this is probably of academic interest only. Lowering the load impedance to 2 ohms and keeping the output voltage constant, resulting in 200W dissipation in the load, brought up the level of second-harmonic distortion, as is shown in fig.5.

Fig.4 Mark Levinson No.33H, 1kHz waveform at 50W into 8 ohms (top); distortion and noise waveform with fundamental notched out (bottom, not to scale).

Fig.5 Mark Levinson No.33H, 1kHz waveform at 200W into 2 ohms (top); distortion and noise waveform with fundamental notched out (bottom, not to scale).

The mainly third-harmonic nature of the No.33H's distortion can also be seen in fig.6, a spectral analysis of the amplifier's output while it drives a low-frequency sinewave (50Hz) at two-thirds its nominal output power into 4 ohms. The only significant distortion component is the third harmonic at 150Hz, this a low -80dB (0.01%). A power-supply artifact can just be seen at 180Hz, but this is 100dB down! The high-level, high-frequency intermodulation plot (fig.7) was also very clean, the 1kHz difference component lying at -95dB (0.002%), and the 18kHz and 21kHz components still better than 80dB down from peak level.

Fig.6 Mark Levinson No.33H, spectrum of 50Hz sinewave, DC-1kHz, at 200W into 4 ohms (linear frequency scale).

Fig.7 Mark Levinson No.33H, HF intermodulation spectrum, DC-22kHz, 19+20kHz at 25V into 4 ohms (linear frequency scale).

Finally, the big Levinson is a powerhouse of an amplifier, comfortably exceeding its rated power. Specified at 150W into 8 ohms, it actually didn't clip (defined as 1% THD+N) into that load until 265W (24.2dBW)! And the wall AC supply, at 114.5V, was starting to droop—this means that, with its own dedicated 30A line, this amplifier will probably put out 300W into 8 ohms.

Fig.8 Mark Levinson No.33H, distortion (%) vs output power into (from bottom to top): 8 ohms, 4 ohms, and 2 ohms.

Into 4 ohms, the maximum output power almost doubled, to 500W (24dBW); into 2 ohms, 900W was available (23.4dBW). (The wall AC voltages for these power figures were 113.3V and 112.5V, respectively.) As we don't have a dummy 1 ohm resistive load capable of sinking the almost 2kW that the Mark Levinson is presumably capable of putting out into this load, I wasn't able to check its clipping power into 1 ohms. However, the fractional decibel drop in dBW each time the load is halved suggests that this amplifier behaves as an almost perfect voltage source.—John Atkinson