Hi Martin,

I remember that "backward spinning" from a blog I came across a while ago. There one member said this:

If it's a Philips based chip set then the D/A or the Digital Filter IC produces a clock which is fed back to the servo processing IC's which is where the Spindle Motor Drive originates. The signal used to be called (on older units where I have had this fault many times, apologies for not thinking of it !!) MC.
If this clock goes missing then the Spindle Motor doesn't spin or may spin backwards.

Link (see post #17): https://www.diyaudio.com/forums/digital-source/352614-philips-cdm12-disk-recognition-2.html

I'm not sure if this applies to your player as well but it's certainly something worth looking at.

PS: Pls note, this is just a pointer based on theory only. I have no hands-on experience with Ouverture!

Thanks, that is an interesting find - even though I don't really know how to measure this.

The strange thing is that it does not happen all the time. Only when selecting the first couple of tracks. So it does feel a bit unlikely that the clock signal could differ depending on laser position.

I am leaning towards trying to replace the round capacitors with regular ones. I have no experience or equipment for soldering surface mounted components.

I appreciate all input! :-)

Martin wrote:

The strange thing is that it does not happen all the time. Only when selecting the first couple of tracks. So it does feel a bit unlikely that the clock signal could differ depending on laser position.

Yes, I know - it's not the first thing that would spring to mind when you see your symptoms but an intermittend contact problem (be it a cold solder joint or whatever) can cause the strangest problems! I did download the service manuals for the Beosound Ouverture and did not find any "MC signal" in the schematics -- except for the "Motor Control" signal, which directly controls the CD spindle. Any problems there would sure cause strangest behaviour in the CD drive motor.

Martin wrote:

I am leaning towards trying to replace the round capacitors with regular ones. I have no experience or equipment for soldering surface mounted components.

Hmm, do you have any specific reason for changing the caps or is it just a "let's see what happens" approach?
Recapping is not really a magic cure, you know? CDM12 uses a different design than the CDM4, which was used in the early models of Ouverture. All the caps you see in your picture are just smoothing caps for the supply lines and I kind of doubt that they're failing. The known capacitor problem in the CDM2 and 4 servos were different. It was a capacitor in the laser current control circuit, but you don't have that in the CDM12 at all.

Soldering SMD's is not so hard as it may look. The caps are normal two-legged components and proper heat application with a good 50W temperature controlled soldering iron and desoldering braid should work fine.
Of course, a controlled hot air gun and a PCB preheater would be better but not absolutely necessary. If you plan on doing more SMT work in future, it's worth looking at those tools. You can get low cost, no name Chinese products off the internet for some 30-60 Euros.

I looked at your uTube video again and started studying the circuit diagram.
There's no technical and justifiable reason to turn the disc backwards. So it's safe to assume that something is seriously wrong there and that this needs to be addressed before moving forward.

There are only 2 parts that affect the spindle motor: IC7003, the decoder chip with the motor control circuitry and IC7002, the driver IC.
If I were in your shoes, I'd do following:
1) measure the drive motor voltage; pin 9 of IC7002 should be + and pin 12 should be minus. Run the drive as you did in the uTube and if you ever read a negative voltage across the motor terminal, you confirmed a fault situation and you can proceed to the next step 
2) measure the output at the decoder; pin 22 of IC7003 should be + and 23 minus; do the same tests as before
If you read the same negative voltage when the CD turns backwards you can go to the next step.
If the output is always positive then you have a problem in the connection from IC7003 and 7002, or the driver 7002 is damaged

3) There are several clock signals used on 7003. I would guess that it's not the actual clock that drives the processor but another timing clock that come from the Servo CPU 7001. If the 7003 processor clock were to fail, the program should force a safe shutdown, but if any other timing signal is missing or intermittent, the firmware might interpret it as a valid signal, which then could seriously screw with the control function of the chip.
Ergo, compare the clock input pins on 7003 with the respective output pins on 7001.
I'm confident you'll find something wrong there.

I'm very greatful for your advice! I will see if I have probes small enough to do the measurements you suggest!

Also, it seems to try to move the laser carriage up despite the "end position switch" engaging at the top (measured to be closing OK). It must be a logical error if it simply does not care about reaching the end position. I guess it is one of the ICs that interprets the open/close of the end position switch?

As an answer on your previous post, I am aware of some of the differences of the two CD mechanisms and figured that even though the newer one does not have the same type of infamous capacitor problem for the laser it seem to fail a lot more frequently than the older one. Several posts have suggested that capacitors still could be the problem but nothing conclusive really... I just figured it was the only thing that I might actually try to do myself. If there is a problem with an IC, it would be both difficult to troubleshoot and also to fix.

I will look into the suggested voltage measurements!

My first thought was that the laser carriage end position switch engaged to late, causing the laser motor to try to move it too far but I cannot find anything mechanical suggesting that this would be the case. It engages only at the very top millimeters which would not be reached when trying to play track 2 or 3 for example.

Martin wrote:

My first thought was that the laser carriage end position switch engaged to late, causing the laser motor to try to move it too far but I cannot find anything mechanical suggesting that this would be the case. It engages only at the very top millimeters which would not be reached when trying to play track 2 or 3 for example.

😕 Well, no. No matter where that switch is located within the physically accessible range of the sledge, it shouldn't make that noise it made in your video.
The home switch is just a reference for the laser/sledge to know where to turn around and start looking for the lead-in and the TOC.
I noticed that the sledge motor is driven also via IC7002...so, if something's wrong there you could expect erratic behaviour in spindle motor as well as sledge motor!

If you don't have the right fine probes to measure at the SMT ICs directly, just solder a very fine wire (wire wrap wires are good!) to an easily accessible spot near the IC and connect your scope or multimeter there.

---------------------------

Afterthought...: But the home switch is connected to IC7004, which talks to IC7001, which talks to IC7003 and controls the sledge motor. So, If something is wrong with the communication clock signal in that path, neither 7001 nor 7003 will make the right decisions and send the right signals!

That still leaves the 2 most probable causes of your problem: faulty IC7002 or bad/intermittent clock/communication between those controller chips.

For some reason, it tries to move the laser further than to the top. That is the rattling noise heard. So it seems to not care that the home switch is triggered.

I trust your deductive skills here. It means that one or more ICs are at fault. To replace any of these would not really be possible for me, so maybe I will see if sourcing a replacement board (PCB 8) is the most realistic way forward - or perhaps changing to a CDM4 kit, which might be easier to find.

At least one other thread (

Martin wrote:

At least one other thread (link) mentioned that the large caps could cause problems without specifying them. But considering that sometimes really strange behaviour can be seen, do you think it is worth spending time on trying to find and replace the caps anyway?

That thread doesn't sound overly convincing, but yes. Your picture shows that the board version is from 1995 and if that's close to the manufacturing date, it's over 20 years old and that is a fairly old age for an e-cap. If I were in your shoes, I'd get replacement caps and change them. I might even change the e-caps on the other side of the board provided they are easily accessible. It won't take much time or money to get replacements. They are readily available from online stores like Mouser, digikey, Reichelt, etc. Just make sure you get quality parts. I just looked at Mouser and you can get Nichicons for approx EUR1.- a piece.

Martin wrote:

I trust your deductive skills here. It means that one or more ICs are at fault. To replace any of these would not really be possible for me, so maybe I will see if sourcing a replacement board (PCB 8) is the most realistic way forward - or perhaps changing to a CDM4 kit, which might be easier to find.

I don't think that the chips themselves are faulty (except maybe the driver IC 7002) but something strange is going on there. And that post in which one guy claimed he had many CDM12 units where some clock signal was missing or intermittent just fits the bill. To troubleshoot this you'd need at least a 2-channel DSO (digital storage oscilloscope). I'm not sure if you have that. If not, I'd try to get my hands on an old CD player that uses the same servo board PCB 88. Many Philips and Marantz players used that drive.
You can probably forget about trying to get replacement chips since they are obsolete and hard to come by. And the CPUs are programmed; so a blank new one is actually useless.

I did come across one laser repair site that claimed "disk spinning backwards or skipping tracks is a hallmark sign for a faulty laser".
But you did say you got an "unused, genuine Philips mechanism", right? Since some sellers on Taobao say exactly the same, I'm wondering, did you buy that from a reputable seller? And I guess you only replaced the OPU (optical pickup) and not the whole CDM12 Industrial drive, is that right?