Notes on the Troubleshooting and Repair of Compact Disc Players and CDROM Drives


[Document Version: 2.86] [Last Updated: 05/25/1998]

Chapter 1) About the Author & Copyright

Notes on the Troubleshooting and Repair of Compact Disc Players and CDROM Drives

Author: Samuel M. Goldwasser
Corrections/suggestions: | Email

Copyright (c) 1994, 1995, 1996, 1997, 1998
All Rights Reserved

Reproduction of this document in whole or in part is permitted if both of the following conditions are satisfied:

  1. This notice is included in its entirety at the beginning.
  2. There is no charge except to cover the costs of copying.

Chapter 2) Introduction

  2.1) Compact discs and the digital audio revolution

The transformation of CD players and CDROMs from laboratory curiosities
to the economical household appliances that have revolutionized the musical
recording industry and have made possible multimedia computing depend on
the availability of two technologies: low power low cost solid state
laser diodes and mass produced large scale integrated circuits.  Without
these, a CD player using 1960's technology would be the size of dishwasher!

Most of us take all of this for granted rarely giving any thought to
the amazing interplay of precision optics and complex electronics - at
least until something goes wrong.  The purpose of this document is to
provide enough background on CD technology and troubleshooting guidance
so that anyone who is reasonably handy whether a homeowner, experimenter,
hobbiest, tinkerer, or engineer, can identify and repair many problems
with CD players and possibly laserdisc players, CDROM drives, and optical
storage drives as well.

Even if you have trouble changing a light bulb and do not know which end of
a soldering iron is the one to avoid, reading through this document will
enable you to be more knowledgeable about your CD player.  Then, if you
decide to have it professionally repaired, you will have a better chance of
recognizing incompetence or down right dishonesty when dealing with the
service technician.  For example, a bad laser is not the most likely cause
of a player that fails to play discs - it is actually fairly far down on the
list of typical faults.  A dirty lens is most likely.  There - you learned
something already!

  2.2) Scope of this document

This document was developed specifically for the troubleshooting and repair
of the CD players in component stereo systems, compact stereos, boomboxes, car
units and portables, as well as CDROM drives (including the Sony Playstation).

The primary differences between these types will relate to how the disc is
loaded - portables usually are top loaders without a loading drawer or tray:

However, as a result of the level of miniaturization required for portables
and to a lesser extent, CDROM drives, everything is tiny and most or all of
the electrical components are surface mounted on both sides of an often
inaccessible printed circuit board with the entire unit assembled using
screws with a mind of their own and a desire to be lost.

For other types:

* Laserdisc players and optical disk storage units have much in common
  with CD players with respect to the mechanical components and front-end
  electronics.  Therefore, the information contained in this document can
  represent a starting point for their troubleshooting as well.  However,
  they may include additional servo systems (optical pickup tilt, for
  example), as well as additional and/or different signal processing

* DVD (Digital Versatile - or Video - Disc) players (which are just now
  becoming widely available), will suffer from many of the same problems as
  CDs and Laser Discs.  Thus, a familiarity with the operating and repair of
  current technology will give you a head start on the amazing wonders (and
  similarly amazing problems) to come.  There is a great deal of information
  on DVD technology in the DVD FAQ.  Electronics Now, December, 1997, has a
  nice article by Steven J. Bigelow covering everything from the DVD format to
  installing and using a DVDROM drive in your PC.

Note that throughout this document, the term 'CD player' is used most often.
However, it should be understood that in most cases, the information applies
to CDROM drives, game machines using CDs like the Sony Playstation, laserdisc
players, MiniDisk players/recorders, DVD players, and other types of optical
disk systems.  Also see the document specifically devoted to these other
technologies: "Notes on the Troubleshooting and Repair of Optical Disc Players
and Optical Data Storage Drives".  Also, where I remember, the term 'disc' is
used to denote a read-only medium (e.g. a regular audio CD or LD) while 'disk'
is used for one that is recordable (e.g., CD-R or MiniDisk).

Note: Links to all the diagrams and photographs referenced from this document
can be found in Sam's CD FAQ Files.

  2.3) Types of problems found in CD players

Many common problems with CD players can be corrected without the need for
the service manual or the use of sophisticated test equipment (though a
reliable multimeter will be needed for any electrical tests and an oscilloacope
of at least 5 MHz bandwidth is highly desireable for servo alignment and more
advanced troubleshooting).  The types of problems found in a CD player can be
classified into several categories:

1. Mechanical - dirt, lubrication, wear, deteriorated rubber parts, dirty/bad
   limit switches, physical damage.  A dirty lens (coated with dust, tobacco
   smoke residue, or condensed cooking grease) - easily remedied - is probably
   the number one cause of many common problems: discs not being recognized,
   seek failure, audible noise, and erratic tracking, sticking, or skipping.

   Even many professionals may mistake (either accidentally or on purpose)
   these symptoms being due to much more serious (and expensive) faults.
   Don't be fooled!

   Cleaning of the lens and any other accessible optical components (usually
   only the turning mirror, if that) and a mechanical inspection should be the
   first things done for any of these problems (and as periodic preventive
   maintenance especially if the equipment is used in a less than ideal
   environment).  See the section: "General inspection, cleaning, and lubrication".

2. Electrical Adjustments - coarse tracking, fine tracking, focus, laser power.
   However, some CD players no longer have some of these adjustments.  The
   servo systems are totally digital - they either work or they don't.

3. Power problems (mostly portables) - weak batteries, inadequate, defective,
   or improper AC wall adapter.

4, Bad connections - broken solder on the pins of components that are stressed
   like limit or interlock switches, or audio or power jacks, internal
   connectors that need to be cleaned and reseated, broken traces on flexible
   cables, or circuit board damage due to a fall.

5. Electrical Component Failure.  These are rare except for power surge (storm
   and lightning strike) related damage which if you are lucky will only blow
   out components in the power supply.  (Or, plugging a 3 V portable into the
   12 V of your automobile.  You can probably forget about this even being a
   CD player again.)

6. Incompatible geographic location :-).  This doesn't really apply to CD
   players but may be a factor with equipment like Sony PlayStations and
   very likely with DVD players.  In their infinite wisdom (or greed),
   manufacturers are including 'country codes' on the discs so that a game
   or movie sold in one place cannot be used in another.  So, if you bought
   a disc on the other side of the world and it doesn't work at home, thank
   the lawyers.....

You can often repair a CD player which is faulty due to (1) or (2) except
for laser power which I would not attempt except as a last resort without
a service manual and/or proper instrumentation if needed - improper adjustment
can ruin the laser.  If discs are recognized at all or even if the unit only
focuses correctly, then laser power is probably ok.  While the laser diodes
can and do fail, don't assume that every CD player problem is laser related.
In fact, only a small percentage (probably under 10%) are due to a failure
of the laser diode or its supporting circuitry.  Mechanical problems such as
dirt and lubrication are most common followed by the need for electrical
(servo) adjustments.

The solutions to category (3) and (4) problems are obvious - but it may take
a conscious effort to remember to check these out before assuming that the
fault is due to something much more serious.

Category (5) failures in the power supply of component (AC line powered) CD
players can also be repaired fairly easily.

Most other electrical failures will be difficult to locate without the service
manual, test equipment, and a detailed understanding and familiarity with
audio CD technology.  However, you might get lucky.  I have successfully
repaired problems like a seek failure (replaced a driver chip because it was
running excessively hot) and a door sensor failure (traced circuitry to locate
a bad logic chip).  Since so much of the intelligence of a CD player is in the
firmware - the program code inside the microcontroller, even the schematic may
be of only marginal value since I can pretty much guarantee that the firmware
will not be documented.  The service manuals rarely explain *how* the equipment
is supposed to work - and then perhaps only in poorly translated Japanese!

You can pretty much forget about repairing electrical problems in portable
equipment other than perhaps bad connections (usually around the audio or
power jacks, internal connectors, interlock switch (since it is stressed), or
elsewhere due to the unit being dropped).  Nearly everything in a portable
(and most CDROM drives for that matter though this is not quite as bad) is
itty-bitty surface mount components.  There is generally only minimal useful
information printed on the circuit board.  Tracing the wiring is a nightmare.
Even the test points and adjustments may be unmarked!

  2.4) Repair or replace?

While CD players with new convenience features are constantly introduced, the
basic function of playing a CD has not changed significantly in 15 years.
None of the much hyped 'advancements' such as digital filters, oversampling,
one bit D/As, and such are likely to make any difference whatsoever in the
listening pleasure of most mortals.  The people who care, do so only because
they are more concerned with the technology than the musical experience.  Most
of these so called advances were done at least in part to reduce costs - not
necessarily to improve performance.

Therefore, unless you really do need a 250 disc CD changer with a
remote control that has more buttons than a B777 cockpit and 2000 track
programmability, a 10 year old CD player will sound just as good and
repair may not be a bad idea.  Many older CD players are built more solidly
than those of today.  Even some new high-end CD players may be built around
a mostly plastic optical deck and flimsy chassis.

If you need to send or take the CD player or CDROM drive to a service center,
the repair could easily exceed the cost of a new unit.  Service centers
may charge up to $50 or more for providing an initial estimate of repair
costs but this will usually be credited toward the total cost of the repair
(of course, they may just jack this up to compensate for their bench time).
Parts costs are often grossly inflated as well - possibly due to a deliberate
effort on the part of manufacturers to discourage repair of older equipment.
However, these expensive parts do not really fail nearly as often as is
commonly believed - the laser is not the most likely component to be bad!
Despite this, you may find that even an 'authorized' repair center will want
to replace the expensive optical pickup even when this is not needed.  I do
not know how much of this is due to dishonesty and how much to incompetence.

If you can do the repairs yourself, the equation changes dramatically as
your parts costs will be 1/2 to 1/4 of what a professional will charge
and of course your time is free.  The educational aspects may also be
appealing.  You will learn a lot in the process.  Thus, it may make sense
to repair that bedraggled old boombox after all.

Chapter 3) CD Digital Audio Technology

  3.1) General Introduction to CD Technology

Information on a compact disc is encoded in minute 'pits' just under the
label side of the CD.  The CD itself is stamped in much the same way as
an old style LP but under much more stringent conditions - similar to the
conditions maintained in the clean room of a semiconductor wafer fab.  The CD
pressing is then aluminum coated in a vacuum chamber and the label side is
spin-coated with a protective plastic resin and printed with the label.

CD-Rs - recordable CDs use a slightly different construction.  CD-R blanks
are prestamped with a spiral guide groove and then coated with an organic dye
layer followed by a gold film, resin, and label.  The dye layer appears
greenish and deforms upon exposure to the focused writing laser beam to form
pits and lands.

The newest variation - DVDs or Digital Versatile Disks (or Digital Video
Disks depending on who you listen to) - implement a number of incremental
but very significant improvements in technology which in total add up to a
spectacular increase in information density - almost 10:1 for the same
size disc.  These include higher frequency laser (670 or shorter visible
wavelength), closer track spacing, better encoding, and a double sided disc.
According to early reports on the final specifications, DVDs will be able
to store 8 times the audio of current CDs at a higher sampling rate and
bit resolution, 2 hours of MPEG encoded high quality movies, and
all kinds of other information.  Raw data capacity is somewhere between
5 and 10 GBytes.  See the section: "Comparison of CD and DVD Specifications"
for additional information.

  3.2) CD information storage and playback

The actual information to be recorded on a CD undergoes a rather remarkable
transformation as it goes from raw audio (or digital data) to microscopic pits
on the disc's surface. For commercial or professional audio recording, the
process starts with pre-filtering to remove frequencies above about 20 kHz
followed by analog-to-digital conversion, usually at a sampling rate of 48 K
samples/second for each stereo channel. The resulting data stream is then
recorded on multi-track digital magnetic tape. All mixing and pre-mastering
operations are done at the same sampling rate.  The final step is conversion
through re-sampling (sample-rate conversion including some sophisticated
interpolation) to the 44.1 K samples/second rate actually used on the CD (88.2
K total for both channels).  (In some cases, all steps may be performed at the
44.1 K rate.)

That is followed by extremely sophisticated coding of the resulting 16-bit
two's-complement samples (alternating between L and R channels) for the
purpose of error detection and correction. Finally, the data is converted to a
form suitable for the recording medium by Eight-to-Fourteen modulation (EFM)
and then written on a master disk using a precision laser cutting lathe.  A
series of electroplating, stripping, and reproduction steps then produce
multiple 'stampers', which are used to actually press the discs you put in
your player.

Of course, it is possible to create your own CDs with a modestly priced CD-R
recorder (which does not allow erasing or re-recording).  Now, re-writable CD
technology with fully reusable discs enables recording and editing to be done
more like that on a cassette tape

Like a phonograph record, the information is recorded in a continuous spiral.
However, with a CD, this track (groove or row of pits - not to be confused
with the selections on a music CD) starts near the center of the CD and
spirals (counterclockwise when viewed from the label side) toward the outer
edge.  The readout is through the 1.2 mm polycarbonate disc substrate to
he aluminized information layer just beneath the label.  The total length
of the spiral track for a 74 minute disc is over 5,000 meters - which is more
than 3 miles in something like 20,000 revolutions of the disc!

The digital encoding for error detection and correction is called the
Cross Interleave Reed Soloman Code or CIRC.  To describe this as
simply as possible, the CIRC code consists of two parts: interleaving
of data so that a dropout or damage will be spread over enough physical area
(hopefully) to be reconstructed and a CRC (Cyclic Redundancy Check) like error
correcting code.  Taken together, these two techniques are capable of some
remarkable error correction.  The assumption here is that most errors will
occur in bursts as a result of dust specs, scratches, imperfections such as 
pinholes in the aluminum coating, etc.  For example, the codes are powerful
enough to totally recover a burst error of greater than 4,000 consecutive
bits - about 2.5 mm on the disc.  With full error correction implemented (this
is not always the case with every CD player), it is possible to put a piece
of 2 mm tape radially on the disc or drill a 2 mm hole in the disc and have no
audio degradation.  Some test CDs have just this type of defect introduced

Two approaches are taken with uncorrectable errors: interpolation and
muting.  If good samples surround bad ones, then linear or higher
order interpolation may be used to reconstruct them.  If too much data has
been lost, the audio is smoothly muted for a fraction of a second.  Depending
on where these errors occur in relation to the musical context, even these
drastic measures may be undetectable to the human ear.

Note that the error correction for CDROM formats is even more involved
than for CD audio as any bit error is unacceptable.  This is one of many
reasons why it is generally impossible to convert an audio CD player into a
CDROM drive.  However, since nearly all CDROM drives are capable of playing
music CDs, much can be determined about the nature of a problem by first
testing a CDROM drive with a music CD.

  3.3) CD (disc) construction

The information layer as mentioned above utilizes 'pits' as the storage
mechanism.  (Everything that is not a pit is called a 'land'.)  Pits are
depressions less than .2 um in depth (1/4 wavelength of the 780 nm laser light
taking into consideration the actual wavelength inside the polycarbonate
plastic based on its index of refraction).  Thus, the reflected beam is 180
degrees out of phase with incident beam.  Where there is a pit, the reflected
beam from the pit and adjacent land will tend to cancel.  This results in high
contrast between pits and lands and good signal to noise ratio.  Pits are
about .5 um wide and they come in increments of .278 um as the basic length of
a bit (encoded, see below) on the information layer of the disc.

Each byte of the processed information is converted into a 14 bit run length
limited code taken from a codebook (lookup table) such that there are no fewer
than 2 or more than 10 consecutive 0s between 1s.  By then making the 1s
transitions from pit to land or land to pit, the minimum length of any feature
on the disc is no less than 3*p and no more than 11*p where p is .278 um.
This is called Eight-to-Fourteen Modulation - EFM.  Thus the length of a pit
ranges from .833 to 3.054 um.

Each 14 bit code word has 3 additional sync and low frequency suppression bits
added for a total of 17 bits representing each 8 bit byte.  Since a single bit
is .278 um, a byte is then represented in a linear space of 4.72 um.  EFM in
conjunction with the sync bits assure that the average signal has no DC
component and that there are enough edges to reliably reconstruct the clock
for data readout.  These words are combined into 588 bit frames.  Each frame
contains 24 bytes of audio data (6 samples of L+R at 16 bits) and 8 bits of
information used to encode (across multiple frames) information like the time,
track, index, etc:

                Sync                    (24 + 3).
                Control and display     (14 + 3).
                Data                    (12 * 2 * (14 + 3)).
                Error correction        ( 4 * 2 * (14 + 3)).
                                        588 total bits/frame

A block, which is made up of 98 consecutive frames, is the smallest unit which
may be addressed on an audio CD and corresponds to a time of 1/75 of a second.
Two bits in the information byte are currently defined.  These are called P
and Q.  P serves a kind of global sync function indicating (among other
things) start and end of selections and time in between selections.  Q bits
accumulated into one word made of a portion of the 98 possible bits in a block
encodes the time, track and index number, as well as many other possible
functions depending where on the disc it is located, what kind of disc this
is, and so forth.

Information on a CD is recorded at a Constant Linear Velocity - CLV.  This is
both good and bad.  For CD audio - 1X speed - this CLV is about 1.2 meters per
second.  (It really isn't quite constant due to non constant coding packing
density and data buffering but varies between about 1.2 and 1.4 meters per
second).  CLV permits packing the maximum possible information on a disc since
it is recorded at the highest density regardless of location.  However, for
high speed access, particularly for CDROM drives, it means there is a need to
rapidly change the speed of rotation of the disc when seeking between inner
and outer tracks.  Of course, there is no inherent reason why for CDROMs, the
speed could not be kept constant meaning that data transfer rate would be
higher for the outer tracks than the inner ones.  Modern CDROM drives with
specs that sound too good to be true (and are) may run at constant angular
speed achieving their claimed transfer rate only for data near the outer edge
of the disc.

Note that unlike a turntable, the instantaneous speed of the spindle is not
what determines the pitch of the audio signal.  There is extensive buffering
in RAM inside the player used both as a FIFO to smooth out data read off of
the disc to ease the burden on the spindle servo as well as to provide
temporary storage for intermediate results during decoding and error
correction.  Pitch (in the music sense) is determined by the data readout
clock - a crystal oscillator usually which controls the D/A and LSI chipset
timing.  The only way to adjust pitch is to vary this clock.  Some high-end
players include a pitch adjustment.  Since the precision of the playback of
the any CD player is determined by a high quality quartz oscillator, wow and
flutter - key measures of the quality of phonograph turntables - are so small
as to be undetectable.  Ultimately, the sampling frequency of 44.1 K samples
per second determines the audio output.  For this, the average bit rate from
the disc is 4.321 M bits per second.

Tracks are spaced 1.6 micrometers apart - a track pitch of 1.6 um.  Thus a 12
cm disc has over 20,000 tracks for its 74 minutes of music.  Of course, unlike
a hard disk and like a phonograph record, it is really one spiral track over 3
miles long!  However, as noted above, the starting point is near the center of
the disc.  The width of the pits on a track is actually about .5 um.  The
focused laser beam is less than 2 um at the pits.  Compare this to an LP: A
long long playing LP might have a bit over 72 minutes of music on two sides or
36 minutes per side.  (Most do not achieve anywhere near this much music since
the groove spacing needs to vary depending on how much bass content the music
has and wide grooves occupy more space.)  At 33-1/3 rpm, this is just over
1,200 grooves in about 4 inches compared to 20,000 tracks on a CD in a space
of just over 1.25 inches!  The readout styles for an LP has a tip radius of
perhaps 2 to 3 mils (50 to 75 um).

  3.4) And you thought driving on a narrow winding country road was tough!

To put the required CD player servo system performance into perspective, here
is an analogy:

At a constant linear velocity of about 1.2 meters per second, the required
tracking precision is astounding: Proper tracking of a CD is equivalent to
driving down a 10 foot wide highway (assuming an acceptable tracking error of
less than +/- .35 um) more than 3,200 miles for one second of play or over
14,400,000 miles for the entire disc without accidentally crossing lanes!
Actually, it is worse than this: focus must be maintained all this time to
better than 1 um as well (say, +/- .5 um).  So, it is more like piloting a
aircraft down a 10 foot wide flight path at an altitude of about 12 miles (4
mm typical focal length objective lens) with an altitude error of less than
+/- 7 feet!  All this while the target track below you is moving both
horizontally (CD and spindle runout of .35 mm) 1 mile and vertically (disc
warp and spindle wobble of up to 1 mm) 3 miles per revolution!  In addition,
you are trying to ignore various types of garbage (smudges, fingerprints,
fibers, dust, etc.) below you which on this scale have mountain sized
dimensions.  Sorry for the mixed units.  My apologies to the rest of the world
where the proper units are used for everything).

The required precision is unbelievable but true using mass produced technology
that dates to the late 1970s.  And, consider that a properly functioning CD
player is remarkably immune to small bumps and vibration - more so than an old
style turntable. All based on the reflection of a fraction of a mW of
invisible laser light!

Of course, this is just another day in the entertainment center for the CD
player's servo systems.  Better hope that our technological skills are never
lost - a phonograph record can be played using the thorn from a rosebush using
a potter's wheel for a turntable.  Just a bit more technology is needed to
read and interpret the contents of a CD!

  3.5) CD optical pickup operating principles

A diagram showing the major functional components of the three-beam optical
pickup described below is available in both PDF and GIF format:

* Get CDT3BP: cdt3bp.pdf or cdt3bp.gif.

This design is typical of older optical pickups (though you may come across
some of these).  Newer types have far fewer individual parts combining and
eliminating certain components without sacrificing performance (which may even
be better).  Additional benefits result is lower cost, improved robustness,
and increased reliability.  However, operating principles are similar.

The purpose of the optical pickup in a CD player, CDROM drive, or optical disk
drive, is to recover digital data from the encoded pits at the information
layer of the optical medium.  (With recordable optical disks, it is also used
to write to the disk medium.)  For CD players, the resulting datastream is
converted into high fidelity sound.  For CDROMs or other optical storage
devices, it may be interpreted as program code, text, audio or video
multimedia, color photographs, or other types of digital data.

Most of the basic operating principles are similar for single-beam CD pickups
and for pickups used in other digital optical drives.

It is often stated that the laser beam in a CD player is like the stylus of a
phonograph turntable.  While this is a true statement, the actual magnitude of
this achievement is usually overlooked.  Consider that the phonograph stylus
is electromechanical.  Stylus positioning - analogous to tracking and focus in
an optical pickup - is based on the stylus riding in the record's grooves
controlled by the suspension of the pickup cartridge and tone arm.  The analog
audio is sensed most often by electromagnetic induction produced by the
stylus's minute movements wiggling a magnet within a pair of sense coils.

The optical pickup must perform all of these functions without any mechanical
assistance from the CD.  It is guided only be a fraction of a mW of laser
light and a few milligrams of silicon based electronic circuitry.

Furthermore, the precision involved is easily more than 2 orders of magnitude
finer compared to a phonograph.  Sophisticated servo systems maintain focus
and tracking to within a fraction of a micrometer of optimal.  (1 um is equal
to 1/25,400 of an inch).  Data is read out by detecting the difference in
depth of pits and lands of 1/4 wavelength of laser light (about .15 um in the

* The laser beam is generated by a solid state laser diode emitting at 780 nm
  (near IR).  Optical power from the laser diode is no more than a couple of mW
  and exits in a wedge shaped beam with a typical divergence of 10x30 degrees
  in the X and Y directions respectively.

* A diffraction grating splits the beam into a main beam and two (first order)
  side beams. (The higher order beams are not used).  Note that the diffraction
  grating is used to generate multiple beams, not for its more common function
  of splitting up light into its constituent colors.  The side beams are used
  for tracking and straddle the track which is being read.  The tracking servo
  maintains this centering by keeping the amplitude of the two return beams

* Next, the laser beam passes through a polarizing beam splitter (a type of
  prism or mirror which redirects the return beam to the photodiode array), a
  collimating lens, a quarter wave plate, a turning mirror, and the objective
  lens before finally reaching the disc.

* The collimating lens converts the diverging beam from the laser into a
  parallel beam.

* A turning mirror (optional depending on the specific optical path used) then
  reflects the laser light up to the objective lens and focus/tracking

* The objective lens is similar in many ways to a high quality microscope
  objective lens.  It is mounted on a platform which provides for movement
  in two directions.  The actuators operate similarly to the voice coils
  in loudspeakers.  Fixed permanent magnets provide the magnetic fields
  which the coils act upon.  The focus actuator moves the lens up and down.
  The tracking actuator moves the coil in and out with respect to the disc

* The collimated laser beams (including the 2 side beams) pass through the
  objective lens and are focused to diffraction limited spots on the
  information - pits - layer of the disc (after passing through the 1.2
  millimeters of clear polycarbonate plastic which forms the bulk of the

* The reflected beams retrace the original path up until they pass through the
  polarizing beam splitter at which point they are diverted to the photodiode
  array.  The polarizing beam splitter passes the (horizontally polarized)
  laser beams stright through.  However, two passes (out and back) through
  the quarter wave plate rotates the polarization of the return beam to be
  vertical instead and it is reflected by the polarizing beam splitter toward
  the photodiode array.

The return beams from the disc's information layer are used for servo control
of focus and tracking and for data recovery.

* A cylindrical lens slighlty alters the horizontal and vertical focal
  distances of the resulting spot on the photodiode array.  The spot will then
  be perfectly circular only when the lens is positioned correctly.  To close
  or to far and it will be elliptical (e.g., elongated on the 45 degree axis
  if too close but on the 135 degree axis if too far).

  The central part of the photodiode array is divided into 4 equal quadrants
  labeled A,B,C,D.  Focus is perfect when the signal = (A+C)-(B+D) = 0.

  The actual implementation may use an astigmatic objective lens rather than a
  separate cylindrical lens to reduce cost but the effect is the same.  Since
  the objective lens is molded plastic, it costs no more to mold an astigmat
  (though grinding the original molds may have been a treat!).  It is even
  possible that in some cases, the natural astigmatism of the laser diode
  itself plays a part in this process.

* The side beams created by the diffraction grating are positioned forward and
  back of the main beam straddling the track of pits being followed (not
  directly on either side as shown in the diagram - but that was easier to

  Segments on either side of the photodiode array designated E and F monitor
  the side beams.  Tracking is perfect when the E and F signals are equal.

* The data signal is the sum of A+B+C+D.

In essence, the optical pickup is an electronically steered and stabilized
microscope which is extracting information from tracks 1/20 the width of a
human red blood cell while flying along at a linear velocity of 1.2 meters
per second!

See the sections: "Parts of a CD player or CDROM drive" and "Startup Problems"
for more information on the components and operation of the optical pickup and
descriptions and photos of some typical laser diodes, optical pickups, and
optical decks.

  3.6) Optical pickup complexity

The opto-mechanical design of optical pickups varies widely.  Originally, they
were quite complex, bulky, heavy, and finicky with respect to optical
alignment.  However, in their continuing effort to improve the design, reduce
the size and mass, and cut costs, the manufacturers have produced modern
pickups with remarkably few distinct parts.  This should also result in better
performance since each optical surface adds reflections and degrades the the
beam quality.  Therefore, the required laser power should be reduced and the
signal quality should improve.

* Generally, the most complex types are also the oldest.  With these, there
  were individual optical elements for each stage in the beam path and
  completely separate laser diode and photodiode array packages.  In short,
  while details varied, the overall construction was very similar to the
  diagram and description given in the section: "CD optical pickup operating principles".  These also had several optical adjustments - which in some
  cases needed frequent attention.

  An example of this type is the Sony KSS110C Optical Pickup.  Most components
  perform individual functions and it is larger and heavier than more modern

* The most common types still have a separate laser diode and photodiode array
  but may have eliminated the cylindrical and collimating lenses and perhaps
  the polarizer and quarter wave plate.  There are few if any adjustments.

  The Sony KSS361A Optical Pickup is typical of these mainstream designs.
  With very minor variations (mostly in mounting), various models may be found
  in all types of CD players and CDROM drives manufactured by Sony, Aiwa, and

  Another similar design is used in the Sanyo K38N Optical Pickup which is
  somewhat newer and more compact.

  For a diagram and detailed description of these mainstream pickups, see the
  section: "Sony KSS series optical pickups".

* Some manufacturers have gone to a combined laser diode/photodiode (LD/PD)
  array package which looks like a large LD but with 8 to 10 pins.  Aside from
  the objective lens assembly, the only other part may be the turning mirror,
  and even this is really not needed.  Such a pickup can be very light in
  weight (which is good for fast-access CDROM drives) and extremely compact.

  Eliminating the components needed to separate the outgoing and return beams
  should result in substantial improvement in optical performance.  The only
  disadvantage would be that the beams are no longer perfectly perpendicular
  to the disc 'pits' surface and this may result in a very slight, probably
  negligible reduction in detected signal quality - more than made up for by
  the increased signal level.

  The CMKS-81X Optical Pickup and Optical Pickup from Philips PCA80SC CDROM
  are typical of these modern designs.

  The smallest ones such as the Optical Pickup from the Philips CR-206 CDROM
  are only about 1/2" x 5/8" x 3/4" overall - just about the size of the lens
  cover!  For this single-beam pickup, there are absolutely NO additional
  optical elements inside.  A three-beam pickup would have a diffraction
  grating in front of the laser diode.

  For a diagram and detailed description of this type of pickup, see the
  section: "Super simple optical pickups".

Go to [Next] segment

Go to [Table 'O Contents]

Written by Samuel M. Goldwasser. | [mailto]. The most recent version is available on the WWW server http://www.repairfaq.org/ [Copyright] [Disclaimer]