I have tried different values of series resistor on the loading motor but this just makes it struggle to operate the heavy mechanism and the oscillation still occurs at the end of travel.
I'm putting this to one side again for now as it is taking up too much time for no return. I'll try the new motors when they arrive and report on the result. If it still doesn't work it will be time for an eBay auction and to wave goodbye to it.
At this moment I really won't be sorry to see it go!
I'm putting this to one side again for now as it is taking up too much time for no return. I'll try the new motors when they arrive and report on the result. If it still doesn't work it will be time for an eBay auction and to wave goodbye to it.
At this moment I really won't be sorry to see it go!
It isn’t unusual to briefly reverse a motor to slow it down rapidly. But the duration of the reverse pulse must be brief or else the type of “oscillating” fault observed from this CD player tray drive system will occur.
It’s of interest that there are 3 logic level control signals between the MCU and the tray motor driver IC.
I expect the oscillating logic signal which goes into oscillation controls motor direction, forward or reverse.
But what are the other two control signals?
Do they ever change?
At least one of them must signal the motor to start or stop.
And what about that other motor which raises/lowers the clamper arm?
Does it have another 3 control lines from the MCU?
Another question: What does this MCU use for a clock source?
It’s likely the MCU clock rate determines the timings for detecting the end-of-travel switches and also the duration of the motor “reverse” pulse.
I will look at the service manual and try to figure this out. I always like a good puzzle. And this is definitely a challenging puzzle!
-EB
It’s of interest that there are 3 logic level control signals between the MCU and the tray motor driver IC.
I expect the oscillating logic signal which goes into oscillation controls motor direction, forward or reverse.
But what are the other two control signals?
Do they ever change?
At least one of them must signal the motor to start or stop.
And what about that other motor which raises/lowers the clamper arm?
Does it have another 3 control lines from the MCU?
Another question: What does this MCU use for a clock source?
It’s likely the MCU clock rate determines the timings for detecting the end-of-travel switches and also the duration of the motor “reverse” pulse.
I will look at the service manual and try to figure this out. I always like a good puzzle. And this is definitely a challenging puzzle!
-EB
I watched the video.
It appears the tray opens without oscillation.
Does the oscillation occur only when it is moving into the closed position?
Likewise it looks like the clamper oscillates only when moving into the “clamped” position. It appears to release the clamp without oscillating.
Is it accurate to say that the oscillation frequency seems about the same for both the tray and the clamper?
Could there be some sort of hidden mechanical interlock between the tray and clamper mechanisms which is perhaps broken or misaligned? And also largely invisible?
-EB
It appears the tray opens without oscillation.
Does the oscillation occur only when it is moving into the closed position?
Likewise it looks like the clamper oscillates only when moving into the “clamped” position. It appears to release the clamp without oscillating.
Is it accurate to say that the oscillation frequency seems about the same for both the tray and the clamper?
Could there be some sort of hidden mechanical interlock between the tray and clamper mechanisms which is perhaps broken or misaligned? And also largely invisible?
-EB
A series Zener would drop the Zener voltage in one direction and conduct fully in the other. Its a case of getting it the right way around.
There are 3 logic levels from the micro to the motor conroller that operate both loading and clamper motors. It's input 3 that doesn't look right on the scope but I will need to confirm the logic states on each with the scope for each stage of operation. If one of these levels is wrong, I thinking maybe I can disconnect it and produce the correct level from the other two using simple logic gates.
The oscillation does occur when the tray opens fully/unclamps as well but not alway, which is peculiar. The oscillation frequency is the same for the loading and clamping phases.
I can't see any other mechanical system that would act as an interlock but will take the mech apart for a closer inspection.
The main clock source for the micro comes from a crystal than is divided down from another chip on the board. I think it's running at around 8Mhz but I'll check this later too.
I like the zener diode idea but as with series resistors, I think the motor drive is already quite weak and any drop in voltage just slugs it to the point that it stuggles to move the tray in and out.
The oscillation does occur when the tray opens fully/unclamps as well but not alway, which is peculiar. The oscillation frequency is the same for the loading and clamping phases.
I can't see any other mechanical system that would act as an interlock but will take the mech apart for a closer inspection.
The main clock source for the micro comes from a crystal than is divided down from another chip on the board. I think it's running at around 8Mhz but I'll check this later too.
I like the zener diode idea but as with series resistors, I think the motor drive is already quite weak and any drop in voltage just slugs it to the point that it stuggles to move the tray in and out.
Last edited:
So what about trying a cap across the motor (a big one) and a series resistor to the motor. If the motor only kicks for a very short period the cap just might slug the response. That would work equally for both directions. The cap needs to be bipolar but miniature 1000uF or even higher are very small in low voltages. Just try a single cap as an experiment (with resistor to the motor) and see. It will be OK as a one off with reverse bias.
I still think a good test is to see if the motor does this reverse kick without the tray in operation. Just operate the limit switch manually and see. Take the belt of the tray.
I still think a good test is to see if the motor does this reverse kick without the tray in operation. Just operate the limit switch manually and see. Take the belt of the tray.
I did remove one of the loading belts and move the tray by hand to watch the motor operate. That's how I found it was going into reverse when the closed switch operates.
The reverse period is too long to be just a braking phase as it's around 5 seconds. This is what seems wrong to me about the whole sequence. A short reversing of the motor, say half a second or so might be more acceptable but not enough to reopen the tray closed switch contacts.
The reverse period is too long to be just a braking phase as it's around 5 seconds. This is what seems wrong to me about the whole sequence. A short reversing of the motor, say half a second or so might be more acceptable but not enough to reopen the tray closed switch contacts.
Although it’s reasonable to assume the four leaf-spring position sensor switches are configured to be “end-of-travel” limit switches, I’ve occasionally run into a situation where multiple switches needed to open or close simultaneously.
My current understanding is this CD player has a total of 4 switches associated with the tray and clamper:
#1 closes when tray is all the way out
#2 closes when tray is all the way in
#3 closes when clamper is fully open
#4 closes when clamper is fully closed
It would be helpful to verify these items:
Do switches #1 and #2 both remain open while the tray is moving?
Do switches #3 and #4 both remain open while the clamper is moving?
Does the “oscillatory” condition stop after 5 seconds, leaving the tray in the correct position (closed)?
Or does the tray return to the open position?
When you hold the tray closed with your finger does that cause the oscillation to stop?
Does the clamper get into oscillation as frequently as the tray?
If you manage to get it into a state where the tray is staying closed and the clamper is clamping the disc, then will the player function normally? Will it play a disc?
With a disc in the machine, does everything work OK as long as nobody pushes the open/close button?
This is such a fascinating puzzle for me that I would get one of these Akai CD-93 (or CD-73) machines mysel if they weren’t so rare and so $$$.
-EB
My current understanding is this CD player has a total of 4 switches associated with the tray and clamper:
#1 closes when tray is all the way out
#2 closes when tray is all the way in
#3 closes when clamper is fully open
#4 closes when clamper is fully closed
It would be helpful to verify these items:
Do switches #1 and #2 both remain open while the tray is moving?
Do switches #3 and #4 both remain open while the clamper is moving?
Does the “oscillatory” condition stop after 5 seconds, leaving the tray in the correct position (closed)?
Or does the tray return to the open position?
When you hold the tray closed with your finger does that cause the oscillation to stop?
Does the clamper get into oscillation as frequently as the tray?
If you manage to get it into a state where the tray is staying closed and the clamper is clamping the disc, then will the player function normally? Will it play a disc?
With a disc in the machine, does everything work OK as long as nobody pushes the open/close button?
This is such a fascinating puzzle for me that I would get one of these Akai CD-93 (or CD-73) machines mysel if they weren’t so rare and so $$$.
-EB
Try this test with belt removed:
1) Start condition: Tray full open
2) Press “close” button.
3) Close the tray with your hand moving it at (roughly) normal speed.
4) Observe what the motor does after the tray closes the “tray closed” switch.
If this test still causes the motor to rotate in reverse for a full 5 seconds that is really strange.
Regarding faulty MCU chips, during my lengthy electronics career I observed that most faulty MCU chips had “total failures” where they didn’t function at all. Rarely an MCU would have a failure of one single I/O pin. usually stuck at either logic high, logic low, or open circuit.
In this case perhaps there is one bad pin on this MCU?
Does this MCU operate on a 5V VCC rail? Please confirm.
If it runs on 5V:
Digital output pins should measure 0V or 5V.
Input pins should detect the transition between 0 and 1 at about 2V
I haven’t examined the service manual yet. I will try to do this soon.
-EB
I looked at the datasheet for the BA6247 motor driver IC. This is one wacky device!
View attachment BA6247FP-Y,BA6238A.pdf
It contains "1 1/2" H bridges. Not 2.
Because of this it can only drive one motor at a time.
The "Vr" pin seems extremely important because the analog voltage at this pin determines the rotational speed of the motors.
One of my theories is that the motors should rotate more slowly.
Finding a way to reduce Vr could accomplish that.
-EB
View attachment BA6247FP-Y,BA6238A.pdf
It contains "1 1/2" H bridges. Not 2.
Because of this it can only drive one motor at a time.
The "Vr" pin seems extremely important because the analog voltage at this pin determines the rotational speed of the motors.
One of my theories is that the motors should rotate more slowly.
Finding a way to reduce Vr could accomplish that.
-EB
I want to learn more about this "transistor and zener diode" connected between the MCU and the BA6247. The root cause of both the tray and clamper oscillation may be tied to this: Vr (pin 8).
According to the BA6247 datasheet the Vr pin is "pulled high" when unconnected (open circuit). This applies the full 12V to the motors to run them at full speed.
To reduce the output voltage and slow down the motors, Vr must be pulled down to a somewhat lower voltage. If Vr is grounded then the motors won't get any voltage.
Perhaps the "transistor and zener diode" are supposed to reduce the motor voltage at "end of travel?"
What signals and voltages are present at the MCU pin which connects to this "transistor and zener circuit"?
Use an oscilloscope with DC coupling to view it.
Then press open/close button to start the tray into motion.
Does the logic state of this MCU pin ever change?
-EB
I'm still reading...
5 seconds sounds like some default timeout period, its not running for 5 seconds when the tray does its dance i.e when the switch closes and then opens again as the tray reverses.
It would be instructive to know if all those (on other forums) saying it is the uP have actually replaced it and nothing else and whether it was an instant fix.
Well...
How many versions of the Akai CD-93 model exist?
Evidently one version has an LED/dual photosensor module in addition to 3 (not 4) mechanical leaf-spring switches for detection of tray position.
The schematic in the service manual I just downloaded from hifiengine shows this LED/photosensor module and it also shows 3 (not 4) leaf switches:
The lower section of this schematic requires careful interpretation:
First of all the box showing +15 at the Vr pin (8) of the motor driver is totally incorrect. There must be no connection to a power supply rail here. Perhaps they were suggesting this voltage would be close to the VCC rail voltage of the driver IC when the motors are resting (not running)?
Because the Vr pin has an internal pullup it will rise to a value close to the VCC of the driver IC when transistor TR5 is switched off.
But when the MCU commands transistor TR5 to turn on, then Vr should be pulled down to the zener voltage of the (unmarked) zener diode. Unfortunately the zener diode isn't labeled on the schematic Sigh... Old schematics were often less than perfect. And they are still like this today.
The voltage of this zener diode must be identified. This can be done by temporarily connecting the collector of TR5 to ground with a cliplead. Doing this should reduce the voltage at the Vr pin. Power must be applied to the CD player to measure this. The CD player should be in a resting state where none of the motors are running.
Also, what is the VCC voltage at the driver chip power input pins (7 and 9)?
Oddly the schematic fails to show any connections to these pins.
Next a visual inspection of the transport mechanism must be conducted:
And exactly where are the 3 leaf-spring switches located?
I would assume that "down sw" and "up sw" apply to the clamper mechanism.
And that "open" detects tray position.
Finally let's take a closer look at the MCU:
Pins 61-64 detect the leaf-spring switches.
When a switch is closed then the corresponding MCU input pin should be at ground potential.
The MCU has internal pullup resistors which should pull all 4 pins up to the MCU VCC (probably 5V) when the switches are open.
Occasionally an internal pullup resistor fails. No problem: Add an external pullup (usually 4K7 ohms) to fix it.
Note: There is an unconnected pin 63 labeled "close."
I assume this would get used for a model that has a mechanical "tray closed" switch rather than the LED/photodetector sensor. If there is a mechanical leaf-spring tray-closed switch then it probably should be connected to pin 63.
There is a different set of MCU input pins for the LED/photosensor module. These are pins 33 and 34.
When light is falling on the photosensors this should cause a logic high level on these 2 pins.
When the light is interrupted the pins should be at logic level 0 (near ground voltage).
No external pulldown resistors are shown. However it's unusual for an MCU to have an internal "pulldown" on an input pin.
If this player has the LED/photsensor, then both pins 33 and 34 must be checked to make sure they swing all the way from logic low to logic high when the photosensors operate.
If there is no LED/photosensor present in this machine then its likely pins 33 and 34 should both be connected either to VCC (5V) or to ground. I'm not comfortable with "unconnected" input pins on an MCU.
Well, all of this should be enough to keep us busy for a while and out of trouble...
-EB
How many versions of the Akai CD-93 model exist?
Evidently one version has an LED/dual photosensor module in addition to 3 (not 4) mechanical leaf-spring switches for detection of tray position.
The schematic in the service manual I just downloaded from hifiengine shows this LED/photosensor module and it also shows 3 (not 4) leaf switches:
The lower section of this schematic requires careful interpretation:
First of all the box showing +15 at the Vr pin (8) of the motor driver is totally incorrect. There must be no connection to a power supply rail here. Perhaps they were suggesting this voltage would be close to the VCC rail voltage of the driver IC when the motors are resting (not running)?
Because the Vr pin has an internal pullup it will rise to a value close to the VCC of the driver IC when transistor TR5 is switched off.
But when the MCU commands transistor TR5 to turn on, then Vr should be pulled down to the zener voltage of the (unmarked) zener diode. Unfortunately the zener diode isn't labeled on the schematic Sigh... Old schematics were often less than perfect. And they are still like this today.
The voltage of this zener diode must be identified. This can be done by temporarily connecting the collector of TR5 to ground with a cliplead. Doing this should reduce the voltage at the Vr pin. Power must be applied to the CD player to measure this. The CD player should be in a resting state where none of the motors are running.
What is the voltage at Vr pin with TR5 collector grounded?
What is the voltage at Vr pin with TR5 collector open?
What is the voltage at Vr pin with TR5 collector open?
Also, what is the VCC voltage at the driver chip power input pins (7 and 9)?
Oddly the schematic fails to show any connections to these pins.
Next a visual inspection of the transport mechanism must be conducted:
Does this LED/photsensor device actually exist?
It is labeled "PHOTOSENSOR PCB."
Therefore a tiny PC board should be present somewhere inside the transport mechanism.
It should be located in such a way that it can detect both "tray open" and "tray closed."
It is labeled "PHOTOSENSOR PCB."
Therefore a tiny PC board should be present somewhere inside the transport mechanism.
It should be located in such a way that it can detect both "tray open" and "tray closed."
And exactly where are the 3 leaf-spring switches located?
I would assume that "down sw" and "up sw" apply to the clamper mechanism.
And that "open" detects tray position.
Finally let's take a closer look at the MCU:
Pins 61-64 detect the leaf-spring switches.
When a switch is closed then the corresponding MCU input pin should be at ground potential.
The MCU has internal pullup resistors which should pull all 4 pins up to the MCU VCC (probably 5V) when the switches are open.
Occasionally an internal pullup resistor fails. No problem: Add an external pullup (usually 4K7 ohms) to fix it.
Note: There is an unconnected pin 63 labeled "close."
I assume this would get used for a model that has a mechanical "tray closed" switch rather than the LED/photodetector sensor. If there is a mechanical leaf-spring tray-closed switch then it probably should be connected to pin 63.
Is anything actually connected to MCU pin 63?
There is a different set of MCU input pins for the LED/photosensor module. These are pins 33 and 34.
When light is falling on the photosensors this should cause a logic high level on these 2 pins.
When the light is interrupted the pins should be at logic level 0 (near ground voltage).
No external pulldown resistors are shown. However it's unusual for an MCU to have an internal "pulldown" on an input pin.
If this player has the LED/photsensor, then both pins 33 and 34 must be checked to make sure they swing all the way from logic low to logic high when the photosensors operate.
If there is no LED/photosensor present in this machine then its likely pins 33 and 34 should both be connected either to VCC (5V) or to ground. I'm not comfortable with "unconnected" input pins on an MCU.
Well, all of this should be enough to keep us busy for a while and out of trouble...
-EB
While trying to determine whether tray motion is detected by mechanical switches or by an LED/photodetector I compared the Akai CD-93 service manual to the CD-73 service manual. These two models are similar in some ways but quite different in others.
The mechanical construction of the CD-73 model is similar to other mid-range CD players. Its lower chassis is a typical "stamped and bent" sheet steel item. Its CD transport frame is made from a smaller sheet-steel stamping. These items resemble many late 1980's era Japanese CD players.
The CD-93, on the other hand, has a very robust and thick die-cast base section. It's CD transport components are mounted directly to several posts which extend up from the die-cast base plate. There is no "sub-chassis" for the CD transport.
The CD-73 has a very small standard DC spindle motor (probably Mabuchi RF-310 or RF-320).
The CD-93 contains a much larger "pancake type" brushless spindle motor.
Returning to the question of "where is the photo-sensor PC board?"
There is a photo on page 7 of the CD-93 service manual which locates it near the rear of the left-hand side of the tray. About 1-2 inches forward from the belts and pulleys.
One curiosity about the tray drive mechanism is this:
I also discovered a far more complete schematic which shows all of the pullup resistors and several other electronic parts which weren't depicted in my previous post of schematic details. There are two different groups of schematics in these service manuals. A "detailed" schematic and also a simplified "functional" schematic (Akai calls it a block diagram"). I'll post details from the "detailed" schematic soon.
-EB
The mechanical construction of the CD-73 model is similar to other mid-range CD players. Its lower chassis is a typical "stamped and bent" sheet steel item. Its CD transport frame is made from a smaller sheet-steel stamping. These items resemble many late 1980's era Japanese CD players.
The CD-93, on the other hand, has a very robust and thick die-cast base section. It's CD transport components are mounted directly to several posts which extend up from the die-cast base plate. There is no "sub-chassis" for the CD transport.
The CD-73 has a very small standard DC spindle motor (probably Mabuchi RF-310 or RF-320).
The CD-93 contains a much larger "pancake type" brushless spindle motor.
Returning to the question of "where is the photo-sensor PC board?"
There is a photo on page 7 of the CD-93 service manual which locates it near the rear of the left-hand side of the tray. About 1-2 inches forward from the belts and pulleys.
One curiosity about the tray drive mechanism is this:
The CD-73 model has 2 rubber belts.
In contrast the CD-93 model has 3 rubber belts.
In contrast the CD-93 model has 3 rubber belts.
I also discovered a far more complete schematic which shows all of the pullup resistors and several other electronic parts which weren't depicted in my previous post of schematic details. There are two different groups of schematics in these service manuals. A "detailed" schematic and also a simplified "functional" schematic (Akai calls it a block diagram"). I'll post details from the "detailed" schematic soon.
-EB
Apologies for my slow response to the last few posts and thank you for your time and input helping to diagnose this problem. I got sidetracked repairing a Pioneer AVR that needed a new DSP chip and then its FL display decided to stop working just to try my patience.
I will try to answer some of the questions posed:
Do switches #1 and #2 both remain open while the tray is moving?
Yes
Do switches #3 and #4 both remain open while the clamper is moving?
Yes
Does the “oscillatory” condition stop after 5 seconds, leaving the tray in the correct position (closed)?
Or does the tray return to the open position?
The tray opens again after the oscillation stops. Sometimes it will move to the clamping stage but mostly it just opens again.
When you hold the tray closed with your finger does that cause the oscillation to stop?
Yes but you can feel the tray pushing back.
Does the clamper get into oscillation as frequently as the tray?
Yes the clamper does the same as frequently
If you manage to get it into a state where the tray is staying closed and the clamper is clamping the disc, then will the player function normally? Will it play a disc?
Once the tray is clamped down, the disc will spin and the player functions normally.
With a disc in the machine, does everything work OK as long as nobody pushes the open/close button?
Yes
Try this test with belt removed:
1) Start condition: Tray full open
2) Press “close” button.
3) Close the tray with your hand moving it at (roughly) normal speed.
4) Observe what the motor does after the tray closes the “tray closed” switch.
If this test still causes the motor to rotate in reverse for a full 5 seconds that is really strange.
This is how I found the motor was reversing and yes its about 5 seconds before it changes direction again.
Does this MCU operate on a 5V VCC rail?
Yes
What signals and voltages are present at the MCU pin which connects to this "transistor and zener circuit"?
I've not looked at this closesly but will add it to the 'to do' list
How many versions of the Akai CD-93 model exist?
I don't know the answer to this - mine is the CD73 and therefore doesn't have the photo sensor, just the 4 x leaf switches.
The schematic for the CD93 has some differences and the loading mech is different too as it has 3 belts and an extra pulley. I have no idea why it is different from the CD73... perhaps they had trouble making it work?!
I will try to answer some of the questions posed:
Do switches #1 and #2 both remain open while the tray is moving?
Yes
Do switches #3 and #4 both remain open while the clamper is moving?
Yes
Does the “oscillatory” condition stop after 5 seconds, leaving the tray in the correct position (closed)?
Or does the tray return to the open position?
The tray opens again after the oscillation stops. Sometimes it will move to the clamping stage but mostly it just opens again.
When you hold the tray closed with your finger does that cause the oscillation to stop?
Yes but you can feel the tray pushing back.
Does the clamper get into oscillation as frequently as the tray?
Yes the clamper does the same as frequently
If you manage to get it into a state where the tray is staying closed and the clamper is clamping the disc, then will the player function normally? Will it play a disc?
Once the tray is clamped down, the disc will spin and the player functions normally.
With a disc in the machine, does everything work OK as long as nobody pushes the open/close button?
Yes
Try this test with belt removed:
1) Start condition: Tray full open
2) Press “close” button.
3) Close the tray with your hand moving it at (roughly) normal speed.
4) Observe what the motor does after the tray closes the “tray closed” switch.
If this test still causes the motor to rotate in reverse for a full 5 seconds that is really strange.
This is how I found the motor was reversing and yes its about 5 seconds before it changes direction again.
Does this MCU operate on a 5V VCC rail?
Yes
What signals and voltages are present at the MCU pin which connects to this "transistor and zener circuit"?
I've not looked at this closesly but will add it to the 'to do' list
How many versions of the Akai CD-93 model exist?
I don't know the answer to this - mine is the CD73 and therefore doesn't have the photo sensor, just the 4 x leaf switches.
The schematic for the CD93 has some differences and the loading mech is different too as it has 3 belts and an extra pulley. I have no idea why it is different from the CD73... perhaps they had trouble making it work?!
Looking at the CD73 schematic, the zener (D19) on the Vr pin of the motor controller is HZ6A2, which is rated at 5.3V to 5.6V, so the motor supply voltage is being dropped to roughly half supply when the transistor is switched.
I need to get my scope on this point and see when and if it is being energised in the loading/clamping sequence.
I need to get my scope on this point and see when and if it is being energised in the loading/clamping sequence.
Last edited:
I watched your video several times. As far as I can tell the tray drive system in your machine looks like the CD-73. There are only two belts and I see something which looks like a leaf switch close to where the CD-93 service manual shows the LED/photodetector module would be.
The CD-73 schematic shows 4 leaf switches for tray & clamper detection. In contrast the CD-93 schematic shows 3 leaf switches and an LED/photodetector module.
I suggest downloading both service manuals: CD-73 & CD-93. Hifiengine has decent scans of both. That's where I got the manuals I'm looking at.
The same MCU is used in both models. Curiously it has one input pin labeled "93/73." This is pin 57. It should measure 5V for the CD-73 model because there is a pullup resistor module labeled 1B3 (47k x8). In contrast the CD-93 schematic has an optional jumper (JW23) which evidently must be installed to ground pin 57 for the CD-93 version which uses the LED/photodetector module instead of the leaf switch for "tray closed." The jumper grounds pin 57.
For the CD-73 MCU pins 34 & 35 are unconnected. For the CD-93 with LED/photodetector these 2 pins have 2K7 resistors pulling them to ground.
I'm always suspicious of unconnected MCU input pins. If pins 34 & 35 are totally floating in your CD player, then you might wish to try connecting them to ground. It's probably OK to connect them directly to ground. The 2K7 resistors are needed only when using the LED/photodetector system.
-EB
The CD-73 schematic shows 4 leaf switches for tray & clamper detection. In contrast the CD-93 schematic shows 3 leaf switches and an LED/photodetector module.
I suggest downloading both service manuals: CD-73 & CD-93. Hifiengine has decent scans of both. That's where I got the manuals I'm looking at.
The same MCU is used in both models. Curiously it has one input pin labeled "93/73." This is pin 57. It should measure 5V for the CD-73 model because there is a pullup resistor module labeled 1B3 (47k x8). In contrast the CD-93 schematic has an optional jumper (JW23) which evidently must be installed to ground pin 57 for the CD-93 version which uses the LED/photodetector module instead of the leaf switch for "tray closed." The jumper grounds pin 57.
For the CD-73 MCU pins 34 & 35 are unconnected. For the CD-93 with LED/photodetector these 2 pins have 2K7 resistors pulling them to ground.
I'm always suspicious of unconnected MCU input pins. If pins 34 & 35 are totally floating in your CD player, then you might wish to try connecting them to ground. It's probably OK to connect them directly to ground. The 2K7 resistors are needed only when using the LED/photodetector system.
-EB
My take on the schematics is that this zener is either 5V or 6V. Today's standard values are 5.1V, 5.6V, 6.2V. Of course there's probably nothing wrong with the original zener.
Take a close look at the detailed schematic on page 26 of the service manual. It's the same page for both models (CD-73 or CD-93).
There is also a 2K2 resistor (R59) between the MCU output pin "L.D gain" which is pin 43 on the MCU and the base of transistor TR5. With this series resistance the MCU output pin should rise above 3V if/when it tries to slow down the motors by outputting a logic high signal.
-EB