Audio Research SP9 preamplifier SP9 Mk.II measurements

Sidebar 2: SP9 Mk.II measurements

Output impedances and line- and phono-stage gains measured pretty much according to the specification, with the CD input gain measuring a sensible 8dB rather than 20dB with the volume control wide open. Fig.1 shows the SP9's line-stage frequency response with the balance control centered. Flat within the audio band, there is a slight 0.5dB rolloff at 20kHz. Via the phono stage, the response was somewhat unflat (fig.2), with a slight treble shelf and a depressed bass region. This departure from strict neutrality doesn't correlate with Guy Lemcoe's impressions of the SP9's phono-stage sound. Distortion levels, however, were superbly low, being below 0.1% across the band via both phono and line stages.

Fig.1 Audio Research SP9 Mk.II, line-stage frequency response, volume control full, 100mV input (right channel dotted, 0.5dB/vertical div.)

Fig.2 Audio Research SP9 Mk.II, phono-stage RIAA error, measured at tape outputs, 10mV input at 1kHz (right channel dotted, 0.5dB/vertical div.)

The latter proved capable of swinging a huge signal: up to 50V! Fig.3 shows the distortion and noise across the band measured at 32V RMS output! Phono overload margins were also excellent, ranging from 36dB at 20Hz and 1kHz to 32dB at 20kHz (all assessed at 1% THD and referred, with the appropriate emphasis, to 5mV input at 1kHz).

Fig.3 Audio Research SP9 Mk.II, line-stage THD+noise (%) vs frequency at 32V output with volume control full (right channel dotted).

The only real negative measurement concerned channel separation via the line stage. As can be seen from fig.4, this echoed the SP14 in dropping to a marginal value at the top of the audio band. The shape of the curve suggests capacitive coupling between the channels, perhaps across the input switching. The intrinsic crosstalk of the phono stage, assessed at the tape-out sockets, was below 60dB at all frequencies.—John Atkinson

Fig.4 Audio Research SP9 Mk.II, line-stage channel separation: L-R (top); R-L (bottom) (10dB/vertial div.).