Compact Discs
One of the most important developments in sound reproduction in recent years has been the Compact Disc (CD). The original audio signal is converted into digital signals (see page 136) at a sampling rate of 44.1 kHz and using a 16-bit system. A version of these digital pulses is impressed on to a master disc using a laser. The latter produces a spiral track of indentations (‘pits’) in the surface. The regions between the pits are called ‘lands’. Moulds of the master are used to make the replay copies in a way that is not unlike the process used with vinyl discs.
Unlike a conventional vinyl record the spiral track on a CD starts at the centre and works outwards. The density of ‘information’ is very great: up to about 75 minutes of stereo sound can be recorded on a CD. To achieve this the spacing between the tracks is only 1.6 μm (1.6 × 10−3 mm). This works out at over 6000 tracks per centimetre.
The linear speed of record (or replay) is 1.25 m/s. This is constant throughout the disc, unlike a vinyl disc where the tracking speed is greatest at the outside and slowest on the inside. This means that the speed of rotation of a CD must vary from around 500 rev/min at the centre to about 200 rev/min at the outside.
In replay a small laser is used with a complex optical system so that the difference between pits and lands is detected in the reflection of the laser beam from the disc. Accurate tracking is achieved by splitting the laser beam into three separate beams. One is used to read data, the other two being projected on to either side of the track. A greater signal from the reflection of one compared with the other means that the laser is going to that side of the track and a servo system moves the laser until the outputs from the two side beams are equal again.
Error correction
With such close spacing of tracks and tight packing of information a CD is basically very susceptible to errors. The tiniest blemish could cause the loss of vast numbers of digits. To overcome this, complex error correction systems are used. To begin with, the digital data (the samples) are scattered following a particular code. The same code is used to re-assemble the data correctly on replay. Thus a flaw on the disc is, in effect, distributed over a large number of samples and further error correction processes on each sample can eliminate the effects of the flaw. In theory a hole 1–2 mm across in a CD should have no audible effect. Some 25 to 30% of a CD’s tracks is devoted to error correction!
Frequency response: 20 Hz to 20 kHz±0.1 dB. Signai-to-noise ratio: 100dB.
Harmonic distortion: less than 0.005%.
Channel separation: better than 90 dB.
CD player
1. Simplified optical system.
2. Mistracking.
3. Correct tracking.