Sony SCD-C555ES SACD changer Measurements
I measured the Sony SCD-C555ES SACD player's performance using my usual CD test signals stored on a CD-R, and with DSD-encoded signals from Sony's "tentative" test SACD (TGZD 90005). All the tests were performed using the two-channel left/right outputs, with occasional spot checks from the multichannel jacks.
The maximum operating levels at 1kHz were 2.088V (CD) and 2.068V (SACD), and the output polarity was correct for CD data. (I don't have an SACD with a suitably asymmetrical waveform.) The source impedance was 110 ohms in the midrange and treble, rising to 133 ohms at 20Hz. Error correction on CD replay was one of the best I have encountered, the SCD-C555ES coping with gaps in the data up to 2.4mm long without dropping out.
Channel separation (not shown) was better than 100dB over much of the audioband, decreasing to 95dB at 20kHz due to the usual capacitive coupling between channels. The frequency response with CD playback (fig.1) was identical with both pre-emphasized and normal data, and showed a 0.5dB rise at 20kHz, which might be just audible as a little extra HF "air." Playing back test tones from the Sony test SACD showed (fig.2) that the rise continued above the audioband, reaching +1.3dB at 40kHz, above which the player's output dropped like a stone, due to the analog low-pass filter used to prevent the DSD encoding's ultrasonic noise content from upsetting downstream components.
Fig.1 Sony SCD-C555ES, CD frequency response at -12dBFS, without emphasis (top) and with emphasis (bottom). (Right channel dashed, 0.5dB/vertical div.)
Fig.2 Sony SCD-C555ES, SACD frequency response at -3dBFS (right channel dashed, 0.5dB/vertical div.).
The beginning of the rise in HF noise with SACD can be seen above 10kHz in fig.3, which shows spectral analyses of the Sony's analog output while it played back a dithered 1kHz tone at -90dBFS from both CD and SACD. While the SACD signal has a noise floor in the midrange and treble around 5dB lower than that from CD, the DSD encoding used on the former results in more top-octave noise. With the measurement bandwidth extended to 200kHz (fig.4), the huge rise in ultrasonic noise due to DSD is readily evident. Without the analog low-pass filtering already mentioned, which can also be seen in this graph, this noise would continue rising in level with increasing frequency.
Fig.3 Sony SCD-C555ES, 1/3-octave spectrum of dithered 1kHz tone at -90dBFS, with noise and spuriae (from top to bottom below 10kHz): 16-bit CD data, DSD SACD data (right channel dashed).
Fig.4 Sony SCD-C555ES, 1/3-octave spectrum of "digital black," with noise and spuriae (from top to bottom above 10kHz): 16-bit CD data, DSD SACD data (right channel dashed).