Solid State Power Amp Reviews

Measurements
After a one-hour preconditioning period driving 1kHz at one-third power into 8 ohms (which maximally thermally stresses an amplifier with a class-B output stage), the No.33H's vertical heatsink arrays were warm, but not excessively so. The amplifier was non-inverting via the RCA jack or via the XLR with pin 2 wired "hot." Its input impedance was to specification at 49.6k ohms (unbalanced) or 104k ohms (balanced). The sensitivity was also as specified, 130mV resulting in an output voltage of 2.838V into 8 ohms (a voltage gain of 26.8dB). The measured output impedance was extremely low at 0.02 ohms at 20Hz and 1kHz, rising slightly to 0.026 ohms at 20kHz. This is an amplifier that won't let itself be bossed around by any loudspeaker out there!

This can be seen in its small-signal frequency-response plot (fig.1). Even driving our dummy loudspeaker load, the change in measured response is negligible. At the high end, the No.33H is 0.5dB down at a sensible 62kHz, this resulting in a slight leading-edge droop in the 10kHz squarewave waveform (fig.2). There is also just the faintest hint of a single-cycle overshoot in this plot, but the waveform is otherwise textbook in appearance.

Fig.1 Mark Levinson No.33H, frequency response at (from top to bottom at 20kHz): 1W into 8 ohms, 2W into 4 ohms, and 2.83V into simulated speaker load (0.5dB/vertical div.).

Fig.2 Mark Levinson No.33H, small-signal 10kHz squarewave into 8 ohms.

Noise levels were pretty low, the A-weighted S/N ratio measuring 87dB ref. 1W/8 ohms. Without weighting, this figure worsened slightly to 84.3dB in the audioband, and 67.2dB with a wide 10Hz-500kHz measurement bandwidth. The big Levinson is superbly linear—its distortion signature is more typical of a high-quality preamplifier! Fig.3 shows the manner in which its THD+noise percentage changes with frequency. We normally perform this measurement at 2.83V, corresponding to an output power of 1W into 8 ohms, but at this level, the measurement was dominated by noise. Accordingly, I ran this test at much higher levels: 15W into 8 ohms, 30W into 4 ohms, and 60W into 2 ohms. In all cases, the distortion remains below 0.01% below 2kHz, with a slight rise above that frequency that is greater into the lower impedances. Our dummy loudspeaker load is not intended to be driven at this sort of level, so I ran that sweep at 5V. Any change in distortion with frequency into that load was negligible.

Fig.3 Mark Levinson No.33H, THD+noise vs frequency at (from top to bottom at 1kHz): 60W into 2 ohms; 30W into 4 ohms; 15W into 8 ohms; and 5V into simulated speaker load.