Mark Levinson No.431 power amplifier Measurements

Sidebar 3: Measurements

Before measuring an amplifier, I run it for an hour at one-third power into 8 ohms with both channels driven. This subjects an amp with a class-B output stage to the maximum thermal stress. The chassis of the Mark Levinson No.431 was very hot at the end of this period, and its internal heatsinks were way too hot to touch. That they are just adequately specified, however, was indicated by the fact that the amplifier shut itself down after 62 minutes of running at one-third power, the front-panel LED flashing red. After it cooled down, the No.431 allowed itself to be operated once more. The THD+noise percentage at 67W into 8 ohms was 0.0035% when cold; this had increased slightly, to 0.0045%, just prior to thermal shutdown, suggesting that the amplifier circuit's operating parameters are not modulated by operating temperature.

The No.431's voltage gain measured 26.7dB into 8 ohms from both its balanced and unbalanced inputs, and neither input inverted absolute polarity. (The XLR is wired with pin 2 hot.) The input impedance for both inputs was 51k ohms in the bass and midrange, dropping to 15k ohms at 20kHz. The Levinson's output impedance was a very low 0.08 ohm. Any variation of the amplifier's frequency response due to the interaction between this source impedance and the impedance of the speaker will be negligible (fig.1, top trace at 2kHz). This graph also reveals an extended HF response, the –3dB point lying at 141kHz, which correlates with the excellent 10kHz squarewave response (fig.2). The response was identical through both the unbalanced and balanced inputs. Channel separation (not shown) was better than 100dB below 500Hz, decreasing to 70dB at the top of the audioband.

Fig.1 Mark Levinson No.431, balanced frequency response at 2.83V into (from top to bottom at 2kHz): simulated loudspeaker load, 8 ohms, 4 ohms, 2 ohms (0.5dB/vertical div., right channel dashed).

Fig.2 Mark Levinson No.431, small-signal 10kHz squarewave into 8 ohms.

The No.431 exceeded its specified clipping power into 8 ohms at our usual definition of clipping of 1% THD+noise (fig.3). With both channels driven, it delivered 234Wpc (23.7dBW) into 8 ohms and 400Wpc into 4 ohms (23dBW). (The AC wall voltage dropped from 127V to 123.5V during this measurement.) Into 2 ohms, the No.431 delivered 700W with one channel driven, though there were some discontinuities in the plot of distortion percentage below that power level. Plotting the THD+noise percentage against frequency at 9V (fig.4) indicated that the distortion remained buried in the noise floor below 1kHz, but rose both into 2 ohms and at high frequencies, though not to any significant extent. The signal/noise ratio itself was an excellent 92.5dB (A-weighted ref. 1W into 8 ohms), decreasing to a still good 74.7dB wideband, unweighted.

Fig.3 Mark Levinson No.431, distortion (%)vs 1kHz continuous output power into (from bottom to top at 1W): 8 ohms, 4 ohms, 2 ohms.

Fig.4 Mark Levinson No.431, THD+N (%)vs frequency at 9V into (from bottom to top): 8 ohms, 4 ohms, 2 ohms (right channel dashed).

The distortion content was predominantly the subjectively inoffensive low-order type, but whether the second or the third harmonic was highest in level depended on the frequency, voltage, and current (figs.5 and 6). The second harmonic became dominant at high output currents (fig.7), but was still very low in absolute terms. Intermodulation distortion was also low, even just below visible waveform clipping into 4 ohms on the oscilloscope screen (fig.8).

Fig.5 Mark Levinson No.431, 1kHz waveform at 59W into 8 ohms (top), 0.0044% THD+N; distortion and noise waveform with fundamental notched out (bottom, not to scale).

Fig.6 Mark Levinson No.431, 1kHz waveform at 115W into 4 ohms (top), 0.0042% THD+N; distortion and noise waveform with fundamental notched out (bottom, not to scale).

Fig.7 Mark Levinson No.431, spectrum of 50Hz sinewave, DC–1kHz, at 200W into 4 ohms (linear frequency scale).

Fig.8 Mark Levinson No.431, HF intermodulation spectrum, DC–24kHz, 19+20kHz at 400W peak into 4 ohms (linear frequency scale).

The Mark Levinson No.431 is a well-engineered powerhouse of an amplifier offering very respectable measured performance. It does run warm, however, so make sure it is well ventilated.—John Atkinson

Mark Levinson
3 Oak Park
Bedford, MA 01730-1413
(781) 280-0300