Mark Levinson No.37 CD transport & No.36S D/A converter Measurements

Sidebar 3: Measurements

Checking the error-correction performance of the Mark Levinson No.37 transport with the Pierre Verany test disc indicated no problems through track 34, where it began to mistrack. This is a very good though not exceptional result. Track 27 is the required minimum CD standard.

The rest of the measurements presented are for the No.36S. Unless otherwise noted, all of the results are for the unbalanced output mode. (Selected measurements were repeated from the balanced output and were not significantly different from the results shown here.)

The Levinson No.36S has an output of 1.85V (3.7V, balanced), left and right, when decoding a 1kHz, 0dBFS (full-scale) sinewave. The output impedance was 6.6 ohms, left & right (13.2 ohms left, 13.6 ohms right, balanced). The Levinson was noninverting from the unbalanced outputs, with a positive-going impulse reproduced as positive at its outputs. The balanced output is configured with pin 2 positive. DC offset was 0.3mV in the left channel, 0.6mV in the right.

The frequency response of the Levinson No.36S was very flat, with less than a 0.25dB drop at 20kHz (fig.1). The de-emphasis error—for that very small percentage of CDs which use pre-emphasis—is insignificant (also fig.1, displaced for clarity). And the crosstalk shown in fig.2 is so low in both channels as to be insignificant as well.

Fig.1 Mark Levinson No.36S, frequency response (top) and de-emphasis response (bottom) (right channel dashed, 0.5dB/vertical div.).

Fig.2 Mark Levinson No.36S, crosstalk (R-L channel dashed, 10dB/vertical div.).

Fig.3 shows the processor's output spectrum decoding a -90dB, 1kHz dithered sinewave signal. The blip at 2kHz was repeatable. The only other anomalies are 120Hz and 240Hz power-supply noise components—at extremely low levels. Using the same type of spectral analysis, fig.4 shows the result of the Levinson's decoding of "digital silence" (all data words zero) in the unbalanced mode, out to 200kHz. The same-level power-supply noise is present—and little else. Fig.5 shows the fade-to-noise with dither, indicating outstanding linearity down to below -110dBFS.

Fig.3 Mark Levinson No.36S, spectrum of dithered 1kHz tone at -90.31dBFS, with noise and spuriae (16-bit data, 1/3-octave analysis, right channel dashed).

Fig.4 Mark Levinson No.36S, spectrum of digital silence (16-bit data, 1/3-octave analysis, right channel dashed).

Fig.5 Mark Levinson No.36S, departure from linearity (right channel dashed, 2dB/vertical div.).

Mark Levinson
P.O. Box 781
Middletown, CT 06457
(860) 346-0896