Cary Audio Design CD 306 SACD Professional Version SACD/CD player Measurements

Sidebar 3: Measurements

I examined the Cary CD 306 SACD Professional Version's measured behavior using Audio Precision's top-model SYS2722 system (see www.ap.com and "As We See It" in the January 2008 issue), as well as, for some tests, our Audio Precision System One and the Miller Audio Research Jitter Analyzer.

The Cary's maximum output level from its balanced jacks was 5.9V with both CD and SACD data; it was 2.93V from the unbalanced jacks. Both sets of outputs preserved absolute polarity; ie, were non-inverting. (The XLRs are wired with pin 2 hot.) The output impedance was very low, at 198 ohms at all frequencies, balanced, and 99 ohms unbalanced. Tested for error correction with the Pierre Verany Test CD, the CD 306 was the best player I have encountered, there being no audible glitches until track 37, which has 3mm gaps in the data spiral. However, while there were no audible problems with tracks 34–36, which have slightly shorter gaps, errors were being flagged in the S/PDIF output words.

The Cary's frequency response was flat within the audioband with both CD and SACD data (fig.1, top pair of traces). Any response error playing back preemphasized CD data was negligible (fig.1, bottom pair of traces). Playing back SACDs, the output began to roll off above 30kHz, reaching –3dB at 45kHz and –12dB at 68kHz. Channel separation (not shown) was superb at >130dB below 1kHz, and still 115dB at 20kHz.

Fig.1 Cary CD 306 SACD Pro, frequency response at –3dBFS into 100k ohms, SACD data (top), response with deemphasis, CD data (bottom). (Right channel dashed; 2dB/vertical div.)

The rather complicated-looking graph in fig.2 shows the spectrum of the CD 306's balanced output under a variety of conditions, taken by sweeping the center frequency of a 1/3-octave bandpass filter from 20kHz down to 20Hz. (I do this test for historical continuity with the library of digital tests I have published going back to 1988.) The top pair of traces shows the spectrum of a dithered 1kHz tone at –90dBFS. The traces peak at exactly –90dBFS, implying minimal linearity error, and are dominated by the recorded dither noise, implying that the Cary's dynamic range is greater than that of the CD medium. The next two pairs of traces overlap below 3kHz; they are the spectra of a dithered tone at 1kHz, first with SACD data, then with external 24-bit data (footnote 1). Both show a drop in the noise floor of around 15dB, suggesting that the Cary player's ultimate resolution is between 18 and 19 bits. The bottom two pairs of traces in fig.2 show the spectra of a dithered tone at –120dBFS, again with both SACD data and with external 24-bit data. The tone is clearly resolved at the correct level, and no harmonic or power-supply spuriae are visible.

Fig.2 Cary CD 306 SACD Pro, 1/3-octave spectrum with noise and spuriae of dithered 1kHz tone at –90dBFS with 16-bit CD data (top), SACD and external 24-bit data (middle at 2kHz and top at 20kHz, respectively), and of dithered 1kHz tone at –120dBFS with SACD and external 24-bit data (bottom at 1kHz). (Right channel dashed.)

The rise in the noise floor in fig.2 with SACD data is most likely due to ultrasonic noise, due to the format's aggressive noiseshaping, increasingly leaking into the 1/3-octave bandpass filter's passband. An FFT-derived spectrum of the same signal (fig.3) shows uniformly low noise, with again no trace of spuriae. Plotting the CD 306's linearity error with CD data (fig.4) really shows only the effect of the recorded dither noise. The Cary's reproduction of an undithered 16-bit sinewave at exactly –90.31dBFS is essentially perfect (fig.5), with the three DC voltage levels described by these data in clear evidence, with excellent waveform symmetry. Increasing the bit depth to 24 or playing back DSD data gave the same result: an excellent sinewave with low noise (fig.6).

Fig.3 Cary CD 306 SACD Pro, FFT-derived spectrum with noise and spuriae of dithered 1kHz tone at –90dBFS with SACD data (left channel blue, right red).

Fig.4 Cary CD 306 SACD Pro, linearity error, 16-bit CD data.

Fig.5 Cary CD 306 SACD Pro, waveform of undithered 1kHz sinewave at –90.31dBFS, CD data (left channel blue, right red).

Fig.6 Cary CD 306 SACD Pro, waveform of dithered 1kHz sinewave at –90dBFS, SACD data (left channel blue, right red).

The Cary offered very low levels of harmonic distortion. Fig.7, for example, is an FFT-derived spectrum of the CD 306's balanced output while playing back a full-scale 1kHz tone from the Sony Test SACD into a high impedance. The only harmonic visible is the second, lying at a roots-of-the-universe –120dB (0.0001%) in the left channel and –116dB (0.00015%) in the right. The second harmonic rose by 10dB with external 24-bit data and is joined by a trace of third harmonic (fig.8), but this is still negligible in absolute terms. And intermodulation distortion was also vanishingly low, even into the demanding 600 ohm load (fig.9).

Fig.7 Cary CD 306 SACD Pro, balanced output, spectrum of 100Hz sinewave at 0dBFS into 200k ohms, SACD data (left channel blue, right red; linear frequency scale).

Fig.8 Cary CD 306 SACD Pro, balanced output, spectrum of 100Hz sinewave at 0dBFS into 200k ohms, external 24-bit data (left channel blue, right red; linear frequency scale).

Fig.9 Cary CD 306 SACD Pro, balanced output, HF intermodulation spectrum, 19+20kHz at 0dBFS peak into 600 ohms, CD data (left channel blue, right red; linear frequency scale).

The Cary demonstrated excellent rejection of word-clock jitter, though the measured level dropped a little as the oversampling ratio was increased. With the CD 306's oversampling set to 384kHz, for example, the Miller Analyzer indicated just 246 picoseconds peak–peak of jitter—and, as can be seen in fig.10, this was almost entirely data-related, and close to the residual level in the test signal. However, some modulation of the noise floor to either side of the fundamental tone can be seen in this graph; in this respect, the Cary performs very similarly to the Krell Evolution 505, which Fred Kaplan reviewed in the September 2008 issue of Stereophile.

Fig.10 Cary CD 306 SACD Pro, high-resolution jitter spectrum of analog output signal, 11.025kHz at –6dBFS, sampled at 44.1kHz with LSB toggled at 229Hz, CD data. Center frequency of trace, 11.025kHz; frequency range, ±3.5kHz (left channel blue, right red).

Finally, the CD 306's digital output operated correctly at 96 and 192kHz, when those rates were selected with the front-panel pushbutton.

Overall, Cary's CD 306 SACD Professional Version offers superb measured performance that is close to the state of the art.—John Atkinson



Footnote 1: The sample of the original CD 306 I examined truncated external 24-bit data to 16 bits. By contrast, the Professional Version performed in exemplary fashion with external data.—John Atkinson

X