CD drive

Overview

Supports speeds of 1x and 2x and can switch between them.

The tracking-servo (balance and gain) is self-adjusting on all consoles, while the focus-servo needs potentiometer-adjustments (bias and gain) on earlier consoles. From PU-18 and up, both the focus and tracking servos are self-adjusting at disc-spinup, which creates an audible "beep" coming from the optical pickup's coils during auto adjustment.

Electronics

The CD subsystem of a first generation PU-7-based PlayStation consists of the following hardware components:

• BA6398FP: BTL-driver (for electrically driving the motors)
• CXA1782BR: Servo controller + RF-amp
• CXD2516Q: CD DSP
• CXD1199BQ: CD decoder (with 32 KB of external RAM)
• Sub-CPU/Mecha-Con

Over the years, these components have been consolidated into fewer, more complex ICs.

Models (mechanism)

• KSM-440AAM
  • Compatible with PU-7, PU-8 and PU-16 boards
  • Can in theory be mounted on PM-41 boards after replacing the top plastic shield, but the cable doesn't have the reinforcement/pull-tab for the non-ZIF connector of PM-41
  • Usually has issues with premature mechanical wear due to the pickup gliding on a plastic peg rather than metal/nylon. Some late production original service parts made around/after mid 1996 have the same cast metal pickup as later units do
• KSM-440ABM
  • Not actually officially used in the PlayStation, but in the "CD-ROM Discman" PRD-150
  • Construction very similar to KSM-440AAM, and as such, it has issues with premature mechanical wear due to the laser block gliding on a plastic peg rather than metal/nylon
• KSM-440ACM
  • Compatible with PU-7, PU-8 and PU-16 boards
  • Adds a plastic piece to the base which will rest above a piece of foam on the internal RF shield of the console
  • Can in theory be mounted on PM-41 boards after replacing the top plastic shield, but the cable doesn't have the reinforcement/pull-tab for the non-ZIF connector of PM-41
  • Early revisions of KSM-440ACM have issues with premature mechanical wear due to the pickup gliding on a plastic peg rather than metal/nylon; later ones have the pickup base made of cast metal and a gliding element made of nylon, as do all the later drives. The change happened around mid 1996
• KSM-440ADM
  • Compatible with PU-18 and PU-20 boards
  • Has a plastic shield at the bottom to prevent the long cable running under the mechanism from getting jammed in it
• KSM-440AEM
  • Compatible with PU-22 and PU-23 boards
• KSM-440BAM
  • Compatible with PM-41 boards (PlayStation One)
  • Can be mounted on PU-7/PU-8/PU-16 boards after replacing the top plastic shield (these boards have a ZIF-connector, but installation - while being a bit more finicky - works fine nonetheless)

While all drives are electrically compatible with each other, mechanical compatibility is not always given, mostly due to differing flex cable lengths on the pickup. In case of the compatibility of the PS One drive (KSM-440BAM) with early PS1 drives (KSM-440AAM/ACM and maybe ABM) and vice versa, the top plastic shield on the drive needs to be replaced to make them compatible. Additionally, installation of early drives into a PS One can be finicky, because they lack the reinforcement plastic tab at the end of the cable, since early consoles had a ZIF connector which didn't require the reinforcement tab. There is also some slight difference with a plastic peg on the drive assembly that sits on the rubber cushion of the top motherboard RF-shield.

Motherboard/CD-Drive mechanical compatibility

compatible	KSM-440AAM	KSM-440ABM	KSM-440ACM	KSM-440ADM	KSM-440AEM	KSM-440BAM
PU-7	Yes	unknown	Yes	No	No	Partial
PU-8	Yes	unknown	Yes	No	No	Partial
PU-16	Yes	unknown	Yes	No	No	Partial
PU-18	No	No	No	Yes	No	No
PU-20	No	No	No	Yes	No	No
PU-22	No	No	No	No	Yes	No
PU-23	No	No	No	No	Yes	No
PM-41	Partial	unknown	Partial	No	No	Yes

Comparison between KSM-440AAM (which does have the wear/FMV stuttering issues) and KSM-440ACM (which mitigates these issues)

Copy protection + Region locking

In optical drives (CD, DVD, BluRay etc), the laser automatically follows the groove by moving the lens sideways (tracking). This is necessary, since tracks on optical discs are neither perfectly centered nor perfectly circular.

The PlayStation exploits this inherent feature of the CD format by having a specially crafted wobbly groove at the beginning of each original disc. The console now monitors the tracking movements the laser has to do to follow this groove and extracts a signal out of this. After this, a string can be extracted from that signal. This string is either SCEI (for Japan and Asia discs), SCEA (for North America discs) or SCEE (for all PAL region discs). The console then compares this to a string it expects, which differs depending on the region of the console. Hence, this method combines both, copy protection and region locking.

The exact way this signal is being extracted by the drive electronics and how it's compared differs between motherboard revisions. PU-7, PU-8 and PU-16 have the circuit built from discrete op-amps and passive components. PU-18 and PU-20 have this circuit condensed into a semi-custom analog IC based on NEC's µPC5032 from the CHS-A family, and PU-22 and all later boards do all of this inside the DSP/CD-ROM decoder/SPU combo IC.

The extraction circuit requires electric adjustment using a potentiometer on PU-8, PU-16, PU-18 and PU-20 to properly function. Later boards don't require adjustment. The potentiometer setting is called "Push-Pull" in official documentation and - on some board revisions - also labeled as such on the board. On PU-8 and PU-16, the potentiometer is RV702. On PU-18 and PU-20, it's RV703 (which is the only potentiometer on these boards). Proper adjustment requires an oscilloscope, a special test disc (SCD-2700) and a way of turning off the tracking servo. Rough adjustment can be done using an oscilloscope and a standard audio CD (originals only, no burnt discs) only and is more than enough to get the drive into a reliable state.

Additionally, some PlayStation BIOS versions(depending on both, the region and the revision) also implement some further checks, which unlike the wobble signal check are not done by the drive hardware but completely in software. These checks involve checking the content on the PlayStation logo screen, which is actually loaded from disc and not part of the BIOS. Either the 3D PS-logo, the licensing text, or both are checked.

Lastly, some games implement their own copy protection measurements, e.g. LibCrypt. As additional protection, some games check for the presence of a modchip and won't work properly if they detect one, no matter if the game disc is original or not.

Issues

Mechanical

• KSM-440AAM, KSM-440ABM and early KSM-440ACM have their optical pickup blocks in a plastic case rather than the cast metal found in all later units. This causes the arms that slide on the drive block's chassis to wear down pretty quickly by friction, thus skewing the laser out of adjustment. This is an infamous issue on early PS1 consoles. It is not caused by heat; instead it is just premature mechanical wear. Placing the console on it's side as has been famously done by many owners of early PS1 consoles back in the day only causes the laser to not slide on that rail at all, as it will mostly "hang freely" in the air around that rail.
  • Fix: Replacing either the whole drive unit with a late KSM-440ACM or even a KSM-440BAM (from a PS One, see notes above in the models section) or just the optical pickup with one of the aforementioned drive unit models.
• Dust covering the top of the lens
  • Fix: Dust on the external lens surface can easily be cleaned off using just a cotton swab, optionally with some CD/DVD lens cleaning fluid (as can often be found as part of DVD drive cleaning sets)
    • Hint: Don't use isopropyl alcohol, as it will dissolve and smear around even the smallest amounts of fats and oils that landed on there with the dust, leaving them as smeary residue on the lens. If you used isopropyl alcohol, you need to additionally wipe the lens a bit with a dry cotton swab or - even better - proper lens cleaning fluid.
• Dust getting inside the optical pickup
  • Fix: to do

Laser unit

The PlayStation drive's pickup employs a "Laser Coupler", which is a single unit consisting of a Laser Diode (LD), Photo Diode (PD, for monitoring the laser output power) and Photo Detector IC (PDIC, the actual laser signal detector).

Adjustments

Power Supply

Some revisions of the power supply have a potentiometer to fine-adjust the output voltage. Normally, changing this should not be necessary, unless repair work has been done to the power supply. Adjustment is done using a multimeter and setting the 3.5V output to exactly 3.5V, measured on the power supply itself with the console turned on. This will also affect the 8V output, which however does have a far larger tolerance, since it will get converted to a stable 5V on the motherboard through a linear voltage regulator which supports a range of input voltages. The other use of the 8V output is for the BTL-driver and the wobble-extraction circuit, so on boards up to including PU-20, especially on PU-7 to PU-16, recalibrating the drive electronics could be necessary if reading issues arise after changing the power supply output voltage.

Potentiometer on Laser/Pickup

By far the most common advice for tweaking the console's disc reading performance is to turn the potentiometer on the disc drive's laser flex cable. This is usually not a good idea.

It happens under the assumption that this would directly just increase the laser power. However, what it actually does is adjusting the bias of the power monitoring photo diode for the APC (automatic power control) circuit. This setting is done exactly once in the factory using a laser power meter in a special jig that the drive unit is placed in outside the console. It is not meant to be changed ever again when servicing a console. It does NOT directly control laser power, which is instead done by the APC circuit. The APC circuit will automatically compensate for aging/weak lasers, keeping the output power constant. Turning this potentiometer will actually increase output power beyond specification. The only reason this sometimes resolves disc reading issues is because it increases the drive's tolerance to other factors like dirt inside the lens assembly or on the laser IC, skewed laser sled (due to mechanical wear), unstable spindle speed due to dried-up/insufficient motor lubrication etc. It does typically NOT fix the actual issue. Hence, this setting should ONLY be changed as a last resort, if everything else fails, and in that case, replacement of the optical pickup WILL be required at a later point.

Only in case everything else has already been attempted and turning the pickup's potentiometer is actually the last resort:
The pickup has the individual current for driving it's laser diode to achieve the specified light emission power encoded on it's label, in the last 3 digits of the bottom printed line. E.g. if this line reads "Cu474", it means 47.4 mA, and if it reads "KQ497", it means 49.7 mA. The actual current used by the console set via it's APC circuit can be determined by measuring the voltage drop over the emitter resistor at the driving transistor for the laser diode on the motherboard, and then compared to the value on the label. On PU-8 and later, this resistor actually consists of two 22 ohms resistors in parallel, resulting in a resistance of 11 ohms. On PU-7, it's a single 10 Ohms resistor. To obtain the current value, apply Ohm's law: divide the voltage drop by 11 Ohms (10 Ohms on PU-7). If it deviates by more than +5 mA or -11 mA (measured at 25°C; current increases/decreases by 0.4 mA by 1°C higher/lower environmental temperature, so you need to know your room temperature to compensate for this), the laser diode has deteriorated past it's life (or the APC circuit is defective; however, this is rare). Note that this method of determining laser diode deterioration does not work if the pickup's potentiometer has been tweaked at any point after manufacturing as this will have moved the APC monitoring circuit out of adjustment!

ONLY if above method actually revealed a deteriorated laser diode could tweaking the potentioneter get a bit more life out of it. Also monitor the RF signal level using an oscilloscope; there is no need to increase the laser current any further once the RF signal level is at 0.95 V_p-p (PU-22 and later are even specified for just 0.90 V_p-p, which is enough for their DSP to achieve a good reading).

Potentiometers on motherboards

PU-7

• RV701 (focus servo gain)
• RV704 (focus servo bias)
• RV705 (sled motor voltage)
  • This potentiometer sits in series with a 68k resistor on the BTL-driver's channel 3 input B for the sled motor control signal

PU-8, PU-16

• RV701 (focus servo gain)
• RV702 (push-pull)
  • This actually regulates a signal coming from pin 5 of the pickup (labeled as NC on all known original schematics, probably for obfuscation) going into the analog SCEx string extraction circuit
• RV704 (focus servo bias)
• Pads for RV705 (sled motor voltage) do exist but are unpopulated/externally bridged; no calibration of this value needed
  • Jumpers on the board are set such that the sled motor control signal goes to the BTL-driver's input A of channel 3 via a fixed 68k resistor. The difference to input B, which is used on PU-7, is an additional internal 10k resistance inside the BTL-driver, resulting in a fixed total resistance of 78k

PU-18, PU-20

• RV703 (push-pull)
  • This actually regulates a signal coming from pin 5 of the pickup (labeled as NC on all known original schematics, probably for obfuscation) going into the analog SCEx string extraction circuit
• Pads for RV701 (focus servo gain) and RV702 (focus servo bias) do exist but are unpopulated; no manual calibration of these values needed, see below
• These boards introduced auto-bias and auto-gain for the focus servo via DSP
  • The servo circuits will be adjusted by the DSP for an ideal frequency response of 0 dB (± 2.5 dB) at 1 Khz. This fact is responsible for the audible 1 KHz "beep" sound during drive initialization in consoles with automatic gain control.
  • For verifying the correct operation of the focus auto-gain function, the official service documentation suggests using an "adjustment jig" and a "servo analyzer" connected between CL711 (P-P) and CL708 (VC). This "servo analyzer" likely consists of a 1 khz sinusoidal signal generator and an oscilloscope in X-Y mode, where the goal would be to retrieve a lissaijous-figure which is symmetrical about both axes, X and Y. The "adjustment jig" may or may not consist of one or two 1 Khz band-pass filters.
  • For verifying the correct operation of the focus auto-bias function, the official service documentation suggests using a jitter meter and an oscilloscope connected between CL704 (RFO) and CL710 (VC). Using a standard (i.e. non-burnt) audio CD in playback, the jitter should be below 9.3ns and the RF-waveform on the oscilloscope should have an amplitude of 0.95 - 1.35 V_p-p. No additional adjustment jig is used for this.

PU-22 and up

• These boards introduced a different way of extracting the SCEx string from the wobble groove, hence no push-pull adjustment needed
• These boards don't have any potentiometers to adjust. No calibration needed. The drive electronics are fully self-adjusting at disc initialization.

Procedures

On PU-18/PU-20, disabling the tracking servo during adjustment - as required for push-pull adjustment - should be possible by opening/closing the jumper SOL702 on the motherboard, located close to the RF-amp and the pickup connector.