Last week, when I powered on my Philips CD Player (CD931), I heard some mechanical grinding noise and the tray moved only a few centimeters outward and then stopped.

The CDM9 tray loading operates like this. The tray has a linear row of teeth driven by a 10-teeth plastic pinion on a plastic pulley spindle which in turn is belt driven by a small DC servo motor.

All the teeth on the plastic pinion have been sheared. I searched the web and apparently numerous people have had the same incident happened to them. The small plastic spindle and pulley costs me nearly US$20 from a seller on eBay but I have no choice and ordered one.

Today, the new spindle arrived and I fitted it to the tray loading mechanism. I explicitly make sure that the end limit switches operates correctly before powering ON. The tray closing stop limit switch operates a few mms before the tray reaches mechanical stop.

With the tray fully extended outwards, upon powering ON, it retracts back into the player but the servo motor did not stop and made a struggling noise when the tray can move in no further. I cut power quickly and re-check limit switches but limit switch operation seems to be OK. I then placed a voltmeter across the motor and switches ON. The DC servo motor did not stop after end limit is actuated, but power to the motor is removed by the control circuit *after* the motor is struggling when the end stop is reached.

So, the designer is NOT using the limit to stop the motion but relies on the high motor stalling current to trip the travel.(eblc1388: Not true, see post below)

I understand why it was designed like that as it is quite common in places like factories or process plants to use torque as a means for stopping actuators. This make sure the actuators are really closed tight.

My complaints is why use flimsy plastic pinion gear that could not withstand repeated torquing operation? The said mechanical force is always on a specific spindle tooth because there is no slip between the pinion and the row of longitudinal teeth on the moving tray.

I don't want my new spindle to shear after just a while.

Any comments? Is there any recommendations or modifications for it?

 

What I had done to a couple of CD drives that had that problem, was that I remounted the "linear Row of teeth" (which is referred to as a rack in a rack and pinion gear setup) on a rubber membrane, I had to slowly grind down the backside of the rack to make room for some .063" "BUNA N" rubber strips. I then glued the rubber strip to the rack and then in turn glued the whole assembly to the drive tray. The rubber strip gives the rack a type of "shock" absorber to absorb some of the torque applied by the motor when stalling on its end points. The rubber is stiff enough to not cause any problems with the opening or closing, but pliable enough to move some when too much torque is applied.

My .02

 

The rubber is stiff enough to not cause any problems with the opening or closing, but pliable enough to move some when too much torque is applied.

Interesting BMorse, that would not be an easy modification for the faint hearted.

After posting the above, someone emailed me with a circuit of the tray motor driving circuit. Apparently I was wrong to assume the motor stop on torque. It is nothing like that.

The motor is start/stop via timed routine by the MCU inside the player. There is only two wires coming to the main board from the limit switch(es). The limit switch(es) closes when the tray is at both ends of the travel. If after a certain time during the close/open operation the limit switch signal is not happening, the MCU will immediately reverse the operating direction. If the limit switch does operate, the MCU will then decide when to switch off the motor supply.

I found this out by putting three diodes in series of the servo motor trying to reduce its speed. The motor did move more slowly but it was not to the liking of the MCU and 70% into the closing stroke the MCU said its enough and reverse the tray direction.

Therefore I come up with this modification and hopefully it will do what I wanted. I haven't fitted it into the player yet but I don't see any problem so far. Apparently there is no problem during the opening stroke as the motor stop immediately upon the limit switch actuation. No change is therefore necessary to the opening operation.

Basically the circuit just places three diodes in series with the motor whenever the limit is activated, reducing the motor torque. If there is no signal from the limit switch, the N-CH MOSFET is driven hard ON and motor will be running with full speed as before the modification.

 

Interesting BMorse, that would not be an easy modification for the faint hearted.

You are right about that, it will take some real good steady hands and a lot of patience.....

Let us know how the modification you did works out....

 

The modification has been made with satisfactory result.

Before the modification, I was using two series diode in the closing path of the motor so the motor operates more slowly. (see photo) This did help to solve the closing problem to a great extend but it takes about 50% longer for the motor to close the tray.

The final modification is what I have proposed in the previous post. It only changes the motor speed when it has closed to its end of travel. Therefore the speed of closing is unaffected for 98% of the stroke and the tray appears to operate as usual. In fact, the only noticeable effect to an user is the noise from the stalling motor is gone and there is no more "jerking" seen on the driving belt when the motor stalls. This is due to slower speed and current of the motor and also it takes a bit longer to reach the dead stop while the motor is stopped by the MCU through a fixed timer. So in effect the motor stalling time becomes shorter then before modification.

The next photo shows the additional circuit built by just soldering the component together without using a circuit board. The last photo shows the complete modification fitted to the player.