Digital-to-analog conversion in playing a Compact Disc

Digital-to-analog conversion in playing a Compact Disc

Summary.

The 16 bit words from the error-correcting circuit are converted into an analog signal by a 16 bit conversion system. This system consists of a digital transversal filter, in which the signal is oversampled 4 times (sampling rate 176.4 kHz) and then filtered in such a way that signals at frequencies above 20 kHz are attenuated by 50 dB after digital-to-analog conversion. The filter is followed by a noise shaper, which rounds off to 14 bits with negative feedback of the rounding-off error of the preceding sample. Next there is a 14 bit digital-to-analog converter, which is followed by a low-pass third-order Bessel filter. The signal-to-noise ratio of the complete system is about 97 dB. Even though the lowpass filter has a sharp cut-off the system is phase linear. The entire system, except for a few operational amplifiers, is contained in three integrated circuits; one for the digital filter (for both of the stereo channels) and two for the two digital-to-analog converters.

Link -> Philips tech. Rev. 40,174-179,1982, No. 6