Integrated Amp Reviews

The I/50's channel separation (not shown) was very good in both directions in the midrange, at >80dB, though it decreased to 54dB at 20Hz and 60dB at 20kHz. With the volume control set to the maximum, the unweighted, wideband signal/noise ratio from the 8 ohm tap (ref. 1W into 8 ohms and measured with the single-ended input shorted to ground) was excellent, at 81.6dB (average of the two channels). This ratio improved to 84.7dB, left, and 86.8dB, right, when the measurement bandwidth was restricted to 22Hz–22kHz, and to 100dB in both channels when A-weighted. The S/N ratios from the 4 ohm tap were all 3dB greater, correlating with the 3dB lower gain from this output. Spectral analysis of the low-frequency noisefloor while the Audio Research amplifier drove a 1kHz tone at 1Wpc into 8 ohms (fig.4) revealed that AC-related spuriae were present at 60Hz and its harmonics but were all low in level. This graph was taken with the volume control set to its maximum; the AC spuriae remained at the same level at lower settings of the control, implying that they occurred in the driver and output stages after the volume control.

The I/50's output power is specified as 50Wpc, which is equivalent to 17dBW into 8 ohms and 14dBW into 4 ohms. With our usual definition of clipping as being when the THD+noise reaches 1%, I measured a clipping power of 44Wpc from the 8 ohm tap into 8 ohms (16.43dBW, fig.5) and 42.5Wpc from the 4 ohm tap into 4 ohms (13.26dBW, fig.6). Relaxing the definition of clipping to 3% THD+N meant that the I/50 met its specified maximum power from the 8 ohm tap but was still 0.5dB short of the specified figure from the 4 ohm tap. The THD+N percentage rises in a linear manner with increasing power in figs.5 and 6, which suggests that the circuit uses only a modest amount of loop negative feedback, but the distortion lies below 0.5% at powers below a few watts. The distortion was higher at low powers with the 8 ohm tap driving 4 ohms (not shown) but considerably lower below actual clipping with the 4 ohm tap driving 8 ohms (fig.7).

This can be seen in fig.8, which shows how the percentage of THD+noise varies with frequency into 8 ohms (blue and red traces) and 4 ohms (green, gray) at 6.33V (equivalent to 5W into 8 ohms and 10W into 4 ohms). The distortion rose at low and high frequencies into both impedances—the former is probably due to the onset of core saturation in the output transformer; the latter is most likely due to limited open-loop bandwidth, which means that less negative feedback is available in this region.

The distortion harmonic waveform (fig.9, bottom trace) primarily consists of the second harmonic, though spectral analysis (fig.10) indicates that higher-order harmonics are also present, with their levels decreasing as their order increases. With the I/50's 4 ohm tap driving an equal mix of 19 and 20kHz tones at 5Wpc peak into 8 ohms (fig.11), the 1kHz difference product lay at a fairly high –53dB (0.2%), though the higher-order products at 18 and 21kHz were almost 20dB lower in level.

The Audio Research I/50's measured performance indicates that the amplifier is happiest when it is driving a higher impedance than the nominal values of the output transformer taps. While the distortion signature is primary the subjectively benign second harmonic, the I/50 offers better linearity from its 4 ohm tap.—John Atkinson