Technical
How Do CDs Work
Like gramophone
records, the information on optical discs are recorded on a spiral
track. However, with a CD the laser starts reading the disc from
the inside ring (table of contents) and ends up on the outside.
When play back starts, a laser beam shines on the ridges and lands
on the data membrane layer. If you look at the image on the right
you can see the data layer moving in grey.
During playback,
the number of revolutions of the disc decreases from 500 to 200
rpm (revolutions per minute) to maintain a constant scanning
speed. The disc data is converted into electrical pulses (the bit
stream) by reflections of the laser beam from a photoelectric
cell.

Cutaway
view of the laser pickup. Depending on whether the laser beam
hits a ridge or a land, the laser beam is reflected and received
by the
photoelectric cell. The disc data is converted into electrical
pulses (the bit
stream) by reflections of the laser beam off of a photoelectric
cell.
When the laser beam
strikes "land", the beam is reflected onto a
photoelectric cell. When it strikes a "ridge", the
photocell will receive only a weak reflection. Thus the
photoelectrical cell receives series of light pulses corresponding
to the ridges and lands in the disc. These light pulses are the
foundation of binary 'digital' data. A simple substitution for the
weak signal "0" and the in-focus signal "1"
results in a pure digital playback without alteration, every time,
without failure or degradation.
In music playback,
a D/A-Converter (digital to analogue converter; DAC) converts the
series of pulses (binary coding) from a decimal place to a
waveform which can be then processed for amplification. The longer
the decimal place the better the sound. Current standard CD audio
is 44,100 pulses per second and 16 bit (decimal places) in digital
word length. Thus a 24 bit system sounds all that much better, in
fact DVD audio is set to allow 24 bit AND pulse at 97,000 times
per second! Go DVD go!!!
The Compact
Disc player mechanism. The laser pickup reads the disc from below.
Thanks to this
optical scanning system, there is no friction between the laser
beam and the disc. As a result, the discs do not wear, no matter
how often they are played. However, they must be treated
carefully, as scratches, grease stains and dust might intercept or
diffract the light, causing whole series of pulses to be skipped
or distorted. This problem can be solved, as during the recording
the Cross Interleaved Reed Solomon Code (CIRC) is added, which is
an error correction system that automatically inserts any lost or
damaged information by making a number of mathematical
calculations. Without this error correction system optical disc
players would not have existed, as even the slightest vibration of
the floor would cause sound and image distortions.
Scanning the disc (part II)
When the laser beam
hits land, all of its light is reflected and the cell gives off
current. When the laser beam shines on a ridge, half of the light
hits the upper surface and the other half hits the lower down
service. The difference in height between the two places is
exactly a quarter of a wavelength of the laser beam light, so the
original beam is totally eliminated by the interference between
the beam reflected from the surface of the disc and the beam
reflected from the ridge. The photocell does not produce current.
It should be noted
that the ends of the ridges seen by the laser are "ones"
and all lands and ridges are "zeros"; thus turning on
and off the reflection is one, steady state is a string of zeroes.
As it is not possible to have two ones next to each other, Eight
to Fourteen Modulation (EFM) is used to convert 8-bit data bytes
to 14 bit units that always have a minimum of 2 and a maximum of
10 zeros between ones. This makes the pits/ridges and lands
separating them 3 to 11 bits long, no less, no more. This
conversion is done in hardware using a ROM lookup table. To
connect these 14 bit units 3 merge bits are used to make sure that
there are no "ones" too close to each other. In audio,
the third merge bit is used to make sure that the cumulative
lengths of the lands and ridges stay equal in the long run,
otherwise a low frequency component is created that the processing
amplifiers can not handle. Thus 8 data bits are actually 17
channel bits on the disc, but called 16 bit for naming
conventions.
There
are 20,000 tracks like this one on one compact disc.
The scanning must
be very accurate because the track of ridges is 30 times narrower
than a single human hair. You can see the "ridge" in the
illustration above -it is the DARK ROUND CIRCLE. When the laser
light is over top of it, the light 'splits' in two, causing a weak
signal. There are 20,000 tracks on one audio compact disc. The
lens which focuses the laser beam on the disc has a depth of field
of about 1 micro;m (micrometer = one-millionth of a meter).
It is quite normal
for the (compact) disc to move back and forth 1mm during playback.
A flexible regulator keeps the lens at a distance of +/- 2 micro;m
from the rotating disc. For the same reason, a perfect tracking
system is required. The complex task of following the track is
controlled by an electronic servo system. The servo system ensures
the track is followed accurately by measuring the signal output.
If the output decreases, the system recognizes this as being
"off track" and returns the tracking system to its
optimum state.
Many CD players use
three-beam scanning for correct tracking. The three beams
come from one laser. A polarized prism projects three spots of
light on the track. It shines the middle one exactly on the track,
and the two other "control" beams generate a signal to
correct the laser beam immediately, should it deflect from the
middle track.
The Disc
The CD is a plastic
disc 1.2mm thick and 12cm in diameter, with a silver-colored
surface that reflects laser light. The maximum playing time for
music recorded on compact disc is 74 minutes. The CD has several
layers. First, to protect the 8 trillion microscopically small
pits against dirt and damage, the CD has a plastic protective
layer. On the top of this layer the label is printed. Then there
is the reflecting aluminum coating which contains the ridges.
Finally, the disc has a transparent carrier through which the
actual reading of the disc takes place. This plastic forms a part
of the optical system. Mechanically, the CD is less vulnerable
than the analogue record, but that does not mean that it must not
be treated with care.
The
CD has several layers. Notice how the ridges contain binary
information.
The protective
layer on the label side is very thin: only 0.002mm. Careless
treatment or granular dust can cause small scratches or hair
cracks, enabling the air to penetrate the evaporated aluminum
coating. This coating then starts oxidizing immediately at that
spot. If the CD is played extensively, it may be advisable to
protect the label side with a special protective foil,
which is commonly available in shops.
A CD must never be
bent, so care should be taken when removing it from the jewel
case. Even slight bending causes stress fractures. The aluminum
then becomes deformed, causing some ridges to be blocked. As a
consequence, error correction always has to be applied in that
area, affecting the final sound.
The reflecting side
of the CD is the side that is read. People tend to set the CD down
with the reflecting side up. But the more vulnerable side is not
the reflecting side but the label side. On the label side, the
reflecting layer with its ridges has been evaporated. The
sensitive layer on the reflecting side has been protected better
than the one on the label side. It is therefore better to store
CDs with the reflecting side down. It is best to store the CD back
in the jewel case, where it is safely held by its inside edge.
Never
write on the label side, even with a felt-tipped pen. The ink may
penetrate the thin protective coating and affect the aluminum
layer.

If a smear,
however small, remains on the CD, much information is lost.
CDs are easily scratched, and should never be cleaned with just
any cloth. CDs should be cleaned radially: not along the grooves,
but at right angles to the direction of the grooves. If a smear,
however small, should remain on the CD, running along the
direction of the grooves, much information would be lost. It is
advisable to use special CD cleaner that operates with a rotating
brush at right angles to the direction of the grooves.
Many people think
that the digital CD is produced completely digitally, but this is
not always the case. Many CDs have an analogue master tape as
their source tapes still kept in the library of the record
company, used in the past to make records. The quality of a CD
made from analogue tape can be surprisingly high. A CD recorded,
processed and dubbed digitally does not always sound better than a
CD produced with one or two analogue processing stages.
To indicate what
stages have been treated in what ways, a useful three-letter code
is used on recordings. The letters represent: the recording, the
editing/mixing process, and dubbing, respectively. They are
printed on the CD and/or on the insert label in a rectangular box.
There are three possibilities: DDD (completely digital CD); ADD
(analogue recording, digital processing and dubbing); and AAD
(analogue recording and processing, digital dubbing). Many CDs
carry the ADD or AAD indication. This does not mean that they are
inferior to the DDD CDs!
Much information
in this section courtesy Philips -Inventors of the CD
|