Network Audio Players/Servers

Every source that finds its way into the Caruso R is either digitised via a 48kHz ADC (analogue inputs), downsampled to 48kHz (LPCM via S/PDIF, network or USB) or upsampled to 48kHz (44.1kHz CD). The response is therefore limited to –0.2dB/22kHz and –1.35dB/23kHz before dropping to –21dB/24.3kHz as T+A's choice of minimum phase digital filter – offering a modest 44dB stopband rejection – bites. The Class D amplifiers demonstrate their own load sensitivity and so the response of the main output is further modified by speaker impedance, peaking at +0.8dB/20kHz into 8ohm, a 'flat' 0.0dB/20kHz/4ohm before the treble is progressively rolled off into lower impedances at –1.7dB/20kHz/2ohm and –4.7dB/20kHz/1ohm. The impact of the amplifier's inductive output filter is also reflected in its source impedance – increasing from 0.07ohm through bass and midrange to 0.2ohm/10kHz and 0.65ohm/20kHz.

Once again, distortion, particularly at high frequencies, is dominated by the Class D output stage. Via the DAC/preamp, distortion falls to as low as 0.0012-0.0025% (20Hz-20kHz) over the top 30dB of its dynamic range [see Graph 2] but this increases to 0.0065–1.09% at 10W/8ohm. Noise, too, is low via the DAC/preamp, yielding an A-wtd S/N ratio of 105.1dB even though the value of 80.5dB (re. 0dBW) via the Class D amp is, relatively speaking, below average. Power output is necessarily tightly regulated by the switchmode PSU and Class D output, reaching a hard clip at 2x50W/8ohm and 2x100W/4ohm under continuous conditions and 61W, 112W, 200W to 300W into 8, 4, 2 and 1ohm loads under dynamic conditions at <1% THD [see Graph 1, below]. So the Caruso R is compact but very punchy! PM