Solid State Preamp Reviews

Sidebar 3: Measurements

I measured the Sugden Masterclass LA-4's performance with my Audio Precision SYS2722 system (see the January 2008 "As We See It"). The maximum gain for the balanced inputs to the balanced outputs was 20.75dB, and a little higher from the unbalanced input to the unbalanced output: 21.4dB. The preamplifier preserved absolute polarity (ie, was non-inverting) with both balanced and unbalanced inputs and outputs. (Its XLR jacks are wired with pin 2 hot, the AES convention.)

The LA-4's unbalanced input impedance was lower than specified, at 12.5k ohms at 20Hz and 20kHz, dropping slightly and inconsequentially to 11.6k ohms at 20kHz. The balanced input impedance was 37.5k ohms from 20Hz to 20kHz. The unbalanced output impedance was extremely low, at about 0.6 ohm from 20Hz to 20kHz. (As this is much lower than the specified value, I checked this using two different methods, with the same result.) The Sugden's balanced impedance was higher, at 119 ohms at 20Hz, and 94 ohms at 1kHz and 20kHz.

The Masterclass LA-4's balanced frequency response into high impedances was flat from 100Hz to 200kHz, though the low frequencies rolled off slightly, reaching –0.75dB at 20Hz (fig.1, blue and red traces). The increase in output impedance at very low frequencies mean that this rolloff reached –1dB at 20Hz into the punishing 600 ohm load (fig.1, cyan and magenta traces). Fig.1 was taken with the LA-4's volume control at its maximum setting. At lower settings, the bass was still down by 0.75dB at 20Hz, but the ultrasonic output rolled off slightly, reaching –0.5dB around 150kHz. In unbalanced mode, with the volume control set to 12 o'clock, the output was again down by 0.5dB at 150kHz, but the low-frequency response was now flat down to 20Hz (fig.2). I looked at the circuit board, but could see no obvious reason for the differences in output impedance and frequency response. Both the balanced and unbalanced outputs appear to be based on high-performance Analog Devices OP275 op-amp chips, followed by complementary pairs of discrete devices.

I was also puzzled by the measurements of channel separation. Measured with the volume control at its maximum setting, the separation was good below 1kHz, at >80dB in both directions, decreasing to 60dB at 20kHz for both balanced and unbalanced operation (fig.1, blue and red traces). However, when I repeated the test at the same output level with the volume control set to 12 o'clock, the R/L crosstalk rose to –53dB at 20kHz (fig.3, gray traces), with the L/R crosstalk almost 20dB higher (fig.3, green trace). This is very unusual.

The Masterclass LA-4's wideband, unweighted signal/noise ratio, measured with the input shorted to ground but the volume control set to its maximum, was disappointing, at 56dB ref. 1V output (average of both channels, which were very similar). Restricting the measurement bandwidth to the audioband increased the S/N to an excellent 81dB, while switching an A-weighting filter into circuit further improved this ratio, to 92dB. Spectral analysis of the Sugden's low-frequency noise floor in balanced mode revealed a spurious tone at 60Hz in both channels. The power-supply transformer has a steel shield around it, but the 60Hz leakage will still be due to magnetic interference from the transformer. However, it was very low in level at –97dB (0.0014%). (It was 10dB higher in level in unbalanced mode.) A 120Hz component can be seen in the left channel (fig.4, blue trace), though at –110dB (0.0003%) this will be inconsequential and inaudible. As the power-supply–related spuriae in the LA-4's output are very low in level, all I can surmise about the disappointing wideband S/N ratio is that, with its extremely extended ultrasonic response, the LA-4 is susceptible to RF interference.

Fig.5 plots the percentage of THD+noise in the Sugden's balanced output into 100k ohms. The THD+N rises below 2V output due to the fixed level of noise becoming an increasing percentage of the signal level, and the LA-4 clips (ie, when the THD+N reaches 1%) at 8V (fig.4), which is well above the maximum level it will be asked to deliver with typical power amplifiers. The balanced output clipped at a still-high 6.5V into 600 ohms, and for unbalanced input to unbalanced output, the LA-4 clipped at 9V into 100k ohms. I measured how the Masterclass LA-4's distortion changed with frequency at a very high level, to be sure that the reading was not dominated by noise. It remained consistently low throughout the audioband into both 100k ohm and 600 ohm loads, with no rise in the top octave (fig.7), though the THD was a little higher from the balanced outputs than the unbalanced, as suggested by figs. 5 and 6.

The spectrum of the distortion in both balanced and unbalanced modes comprised almost pure second harmonic (fig.8), but this was very low in level. Intermodulation distortion was also very low, the second-order difference product at 1kHz with an equal mix of 19 and 20kHz tones lying at just –97dB (0.0014%) (not shown).

I was puzzled by the difference between its balanced and unbalanced operations, but otherwise, the Sugden Masterclass LA-4 measured well.—John Atkinson