Sound essentially consists of analog signals, which processing is associated with the problems of attenuation, noise, and deterioration. These issues are addressed when the original sound passes through an analog-to-digital converter (ADC) after which the resulting data can be distributed on CDs or via networks as digital sound. Following, these data are processed using a digital-to-analog converter (DAC) in the end user's digital audio device and output as analog sound.
In the digitization of analog signals, sampling is carried out at a certain frequency. To reproduce sound with the highest fidelity possible, higher sampling frequencies and bit rates are required. Today's high-resolution audio sources are characterized by sampling frequencies and bit rates superior to those used for CDs, enabling digitalization for true high-fidelity audio reproduction.
Phase noise and jitter
Faithful reproduction of high-resolution audio sources requires precise digital signal processing and analog sound output with reduced deterioration of the sound source in the digital audio device. This conversion accuracy depends on the noise characteristics (i.e., frequency components outside the target frequency) of the clock frequency of the audio device used.
The clock frequency spectrum for a circuit with zero noise has the form of a straight line (Figure 1, left). However, real-life spectra are modulated by noise and are characterized by an extra frequency component nearby (Figure 2, right) known as phase noise. The phase noise of a clock frequency influences DAC and makes the time interval irregular. This phenomenon is called jitter (see the figure below).
Accurate low-noise clock sources are necessary
In digital audio devices, the phase noise of the master clock influences DAC due to jitter, thereby impeding high-fidelity audio reproduction. To enhance sound reproducibility, a crystal oscillator for a master clock with superior phase noise characteristics (i.e., low jitter) is necessary.
Phase noise is expressed as frequency component levels measured outside a crystal oscillator’s original frequency and is based on the component level of the original frequency. Offset frequency is the departure from the initial frequency and is normally measured in the range of 1 Hz - 1 MHz.
Frequency stability (the characteristic by which frequency does not change over an extended period) is generally seen as an important property of crystal oscillators. However, audio devices require short-term rather than long-term stability. Against such a background, XO’s which have a frequency stability of about ±30 - ±100 ppm, are commonly used for master clocks. High-end audio systems require an OCXO with excellent short-term stability close to the carrier for higher quality sound.