CD Player in distress: Philips 3 disc carousel changer,

[Vince]

Hi people,

Just a few days after fixing my very first CD player :

https://www.eevblog.com/forum/repair/vintage-cd-player-repair-philips-cdc-486-6-disc-changer/msg3366804/#msg3366804

... here is another one. Also Philips, also from my dad.


CD player is not a stand-alone unit, it's part of an "all in one" / combination stereo system, branded Philips, that daddy bough in 2000.... Yes, as you can see from the pics, it's written crap and ugly all over it. I doubt Philips designed this horrible thing... they probably just hired some factory in China and tasked them with designing the thing, insisting that the total cost must be under 20 bucks... probably.

This thing is so crap that the old swing arm in comparison, is already starting to feel to me like a tremendously fine piece of engineering and build quality.

Neither me or my dad care the slightest bit for this stereo system as a whole. The only thing my dad likes, very much, in this thing, is the 3 disc CD changer, since that allows him to listen to several hours of uninterrupted music. A functionality he values a lot.
So although I don't care for this ugly crap thing, however it would be cool to be able to make my dad happy again... that's what drives me in this particular repair 

I doubt anybody would care helping me repair this things, and I understand that... but well, if only one single person posts one single sentence that leads me to the solution, then that's all I need ! 


OK so let's get started.

 I started working on it yesterday. First big fear I had, since it's an all in one design, is how the hell am I going to ACCESS the CD player, and work on it in situ !  Luckily that was easily solved as you can see : it's actually not badly designed. The turntable at the top, is removed in seconds with just 4 screws. Then the CD player is next in line. A dozen screws later and the whole CD player unit can be pulled out of the system. It's connected to the chassis only by 2 ribbon cables, gifted with connectors. Cables are quite long, enough to conveniently lay the CD player sub-assembly flat on to the bench, next to the main chassis. This way it is accessible and can still be operated using the front panel buttons on the main unit.

OK so now it's like we had a "normal" stand-alone CD player to fix. So what do we have at hand...

The design/construction of it does not make it easy to work on sadly.  There is a single PCB as far as electronics go. It's screwed over the  head assembly, covering it completely ! You get to see only the solder side... though the silkscreen does nicely show / identify the various sections of the PCB, which is convenient :

- " Loader Control "
- " HF Amplifier "
- " Servo Drivers "
- " Decoder "
- " DAC Audio "

Silk screen gives some more useful information like, what pins drive teh various servos, power supply points, left and right audio channels, etc.  It's all very much appreciated because of course there is no schematic or service manual for this thing. Well, to be honest I have nt even bothered searching ! I can't believe this kind of crap products could have schematics available. Not even sure how I would go about identifying the thing... what model number ?!...

All that is good. What's NOT good is that the whole player is actually one big giant piece !  The carrousel, tray, head assembly, PCB.. .it's all one big indivisble unit : when you open or close the tray, the entire assembly comes out ! So, if I open the tray, the head, as well as the  PCB, move outward along with the tray, of which they are a part of. So it's not easy to probe the PCB if it's moving...

Also, if I want to access the component side of the PCB, I have to flip it, but then of course that means I can't open the tray or it the board will get ripped...
Also, if you wnat to flip the board, you have to desolder the terminals of the two big motors that drive the tray and rotate the carousel.. because the the motors are solders directly onto the PCB !  No cables, no nothing.

Also, it's near impossible to get to see the laser lens and photo diodes.. without destroying the tray with a chainsaw, as it's part of it, and covering the diodes 90%.


OK so that's for access and serviceability.

Symptoms now : old man complained that the player would not detect discs any more : it would cycle through the 3 discs of the changer, acting as if it found no discs in there. So basically, the very same symptoms the swing arm had !

Given what I learned from the swing arm repair, I did :

- Cleaned the laser lens thoroughly, using IPA and Q tips, because the tray was covered with lots of dust ! I mean a lot, what you would get if using the player in a garage, not in a living room... it's probably what he did. Component side of the PCB too was very dusty, so cleaned that. Solder side had weird milk / white stains all over it, don't know where that might come from. Flooded it with IPA again... that did it, clean now.

- Tried to see if the laser was firing... can't see anything but it's hard to get too as I said, that I can only see it from a very shallow angle, so I can't vouch for anything    Need to try harder.

- Checked the spindle motor : unlike the swing arm, the spindle here does turn. However I noticed that it seems to turn properly only when it's unloaded, ie when there is no disc to rotate. When it is loaded with a disc, I can see the spindle get moving, but it stops immediately, having moved only a tiny bit, a fraction of a turn, I don't know, a quart of a revolution, something like that, then it switches to the next disc. Seems weird to me. So maybe the motor is tired ? Or the driver is weak somehow. weak power rail, what have you. Could be so many things.

That's it. Spent only an hour on it last evening, so didn't get very far of course.

As always, any advise / contribution much appreciated... I understand it's not exactly fine piece of audio equipment so hard to motivate people on that one but... it's not about saving a piece of junk, it's about making the old man happy with a functional CD changer unit, nothing more.

A few piccies below.


So while everybody laughs at me spending time on this master piece... ahem... I will resume work on it.

Next to do I think :

- Try harder to see if I can see the laser firing.
- Check power supplies, based on the silk screen information I can find here and there on the PCB.
- Try to access the photo diode and clean them too : a clean laser is not help if the receiving side is obfuscated with dust, I guess...
- Check the spindle motor on the component tester and / or check currents in the windings on the scope, to see if all windings are healthy.

The fun begins...

[Vince]

...

[tautech]

Next to do I think :

- Try harder to see if I can see the laser firing.
- Check power supplies, based on the silk screen information I can find here and there on the PCB.
- Try to access the photo diode and clean them too : a clean laser is not help if the receiving side is obfuscated with dust, I guess...
- Check the spindle motor on the component tester and / or check currents in the windings on the scope, to see if all windings are healthy.

And clean that grubby PCB ! 

Based on the dust loading on it dust could be the main problem.

Stackers/multiple disc player IME always check which slots are populated with a CD immediately on closing the door.
Smart units will show the CD names on the display when you toggle through them so it really needs be tried with commercial CD's and not some PC written one.

Good luck and following along for the ride. 

[Vince]

Making progress.

Tried harder to see the laser. Finally found a tight spot where I can see it, at a nice angle too. Turned the lights off in the lab. I can very clearly see the red laser being turned on just after the spindle is started, then I can see the focus servo moving up and down, then the laser being turned off, then it moves on the the next slot in the carousel and does this cycle again.
So that's a good start...  this cycle is normal from what I understand. Drive is doing what's supposed to, and the laser works fine.

Then I checked power rails. Found only two : +10V and +5V.  I guess the +10V is like in the swing arm, powering the laser LED and all the various motors. It is there, and no ripple, all good. Just like the swing arm, the 10V rail read very high : 11,6V here.
The 5V now : Iat first I was baffled, it was plain missing, reading 0,00V !  How can that be ? I mean obviously it powers the digital stuff that drives the smarts of the player... so it had to be there... made no sense. So I probed it while at the same time pushing buttons on the front panel to make the player do something : oh, 5V pops up ! somehow, it is present ONLY when the player is doing something : opening the tray, operating the carousel... then as soon as it's done, power rail gets to 0V.... don't understand.... but I guess it must be a feature, so I will call it good or else I fear I might be chasing yet another red herring, my specialty I admit...

OK, going back to the lab...

[Vince]

And clean that grubby PCB ! 

ARGH... it's already been cleaned... as I said I flooded it with IPA and brushed it with my static safe "tooth brush" style brush but... the only pic I had of the PCB, component side, was taken BEFORE the cleaning, sorry! 

Pic of the solder side is current though, taken after cleaning the white stains.

Stackers/multiple disc player IME always check which slots are populated with a CD immediately on closing the door.

Yep it does that indeed. That's how I trigger it to see it dance and have a chance of observing the laser being turned on and off.

Smart units will show the CD names on the display when you toggle through them so it really needs be tried with commercial CD's and not some PC written one.

My unit is definitely not smart !    Way too cheap and crap to embed any smartness...
I do have proper commercial CD's now... dug out my collection of CD's from back in the day. So no more CDR for me, don't need that anymore.

[Vince]

OK, first thing I did to help in working on the thing, was to make it easier to pull the board. As I said it's flipped upside down, and is covering the head assembly. So everytime I want to look at the component side or the head, I have to desoldered 4 huge blobs of solder that hold the PCB on the terminals of the two big motors that drive the tray and the carousel. So I desoldered the motors and connected them to the board with long wires, so I can flip the board without having to disconnect the motors. I used regular wires, but they are too think to allow for the PCB to sit flat on the motors' butts, which makes the PCB interfere/ clash with the metal work as the tray  is operated and the PCB gets moved with it.
Idea was good but first implementation was poor. Si I didn't give up and instead of regular PVC insulated wires, I used some enameled wire salvaged from whatever choke I could find in my parts bins. It's so thin that it solves the problem completely. Success. I can now operate the drawer / drive yet still pull the board in and out in an eye blink.

Then, just for goo measure, I reflowed the joints. I mean...100% of them... really 100%. Took an hour to do, cost me an arm and a leg worth of flulx and flux remover chemical, and also probably cost me 10 years worth of life expectancy since I don't have a fumle extractor. Yes, getting one is my long list of "to do/get" things...
It didn't change anything, but at least now I can rule out bad joints, which is one more step towards the solution...

Then started studying the board. The silk screen helps for sure.
No doubt it's a Philips... only Philips chips on the board, and I also saw the same tendency I found on the swing arm, for fusible resistors protecting all the chips. Half a dozen of these resistors. All 3.3 ohms. All tested good.

Half the board space is devoted to the motor drivers. No less than 6 motors in the there. 3 chips in all. Each can drive 2 motors. They are  TDA7073. They run a single positive supply yet can drive the motors in both directions using a differential design (the terminals of the motors are floating, not referenced to ground). They are also designed to run on a single supply, 5V typically though they can take more than that.  So it's more integrated / modern than the swing arm, which had home made drivers using op-amps and discrete bipolar trannies to drive the motor, as well as a symmetrical power supply.

Then we have the DAC, a TDA1311A.

Then we have the main chip on the board, SAA7378GP, a "CD7" whatever that means, which is basically a single chip solution to handle the grunt work. Datasheet says it's basically the combination of a decoder (based on CD6 SAA7345GP ) and a Digital Servo ( based on a DSIC2 TDA1301T ), but with little improvements to both of these chips, while they were at it.

I guess I don't have to care about the DAC part of the board, since it's downstream of the main chip. It just cannot have any say on the operation of the CD player itself. All it could possibly do is give crap sound or no sound at all, but it would not keep the discs from being detected.

As for the 3 motor drivers.... well clearly most of them work fine. The two big ones for the tray and the carousel sure work just fine. The one driving the focus servo appears to be working fine as I could see it moving up and down. The tracking motor I guess we don't care for now ? Or maybe we do, I don't know if it's being used when trying to read/detect the CD ? Just in case, I will investigate a bit in this area.
Then there is a new motor that wasn't there in the swing arm. It's referred to in the datasheet as the "sled" motor. I guess that's the one that moves the head on a horizontal plane to do the fine tracking ?
Then of course there is the spindle motor. I still need to check the motor itself, but at a guess I would say it's good. I mean, been thinking about it behaves. The player when it spins the disc to try to detect it, always rotates it by the exact same amount, and it always starts promptly. A quarter turn. If the problem was a defective motor, defective winding say, I would expect it's behaviour to be somewhat random. It would spin a different amount each time, and it would not always start as promptly / energetically as it does. Its behaviour seems too "consistent" to me to be indicative of defective motor. Still, will quickly test it of course.

- Cleaning : meant to clean the photodiodes after cleaning the laser lens. But looking at the head, I can't see how I would go about doing that... looks like the photodiodes are embedded, sealed inside the head itself, so basically they are not exposed to the outside world, so nothing to clean ?! Only thing to clean is the laser lens ? Well that's done. So now I can rule out dirty photodiodes I guess... one less problem, good.




So what's next ?

Motors work, lens is clean, laser is firing up, focus motor is working... what's left that could keep the main chip from reading data/ detecting the disc ?  That would be the photo diodes, and whetever circuitry around it.

Since the CD player is basically a single chip design, reading its 45 page datasheet will basically tell me a lot about the architecture and workings of the board itself. Will try to match what I see on the board, and the info I see in the datasheet... should teach me lots.
At a quick glance, I see that the chip allows for various configurations/arrangements of photodiodes, called "astigmatic focus", "single Foucault " and " double Foucault ". They also talk about " satellite " diodes.... whatever they are ? I assume the photodiode array is used to read back the data itself, and the satellite diodes maybe are like what the wing arm schematic called " monitor " diode ? A diode that doesn't read data, but just monitors the output power of the laser so that the closed loop driving the laser diode can regulate its current ? I don't know.

Anyhow. Plan is concentrate on the main chip datasheet to try to figure as much details as possible on the design of the board, then test it's various parts. Concentrating first on the photodiodes : do they work ? Does any data / bit stream come out of them and get into the chip ?

[Vince]

Been reading the datasheet of the main / decoder chip.  I have my answer as to what those " Satellite " photodiodes are. They are used in pair always, and are there to implement the tracking. Nothing to do with the swing arm "monitor" diodes... I don't see any such thing in this 3 disc changer. Laser doesn't seem to be monitored at all, it's open loop. I guess you just have to cross fingers...

Spindle motor : powered it with the lab power supply. Works just fine. It starts from as low as 1,5Volt or so.
Tested the tracking motor too. It too, works fine.
It's the sledge motor that I don't know how to test as I can't see it moving, too small and buried in the head assembly.

Anyway, thanks to Google, I now have serious hopes of fixing this thing. I searched for the model number of this stereo system, Philips AS665C and to my surprise there are actually many hits ?!  Extremely useful / interesting ones at that, not just random posts on forums.

So I found :

1)  A website that sells spare parts for consumer devices. They somehow have spares for this thing. They sell the laser / optical assembly, for only 30 Euros, and the entire mechanism,with tracking etc, for barely more, 40 Euros. Wow... I am impressed.

https://www.brelect.fr/index.php?page=res_liste_produits.php&motcle=PHILIPS%20AS665C&cat=0&fam=0&type=CHAINE+HIFI

2) A Philips document, for internal use, for their service techs. A 65 page compilation of all the service bulletins pertaining to this very model ! 
So could search all this info and see what they might have to say about the CD player...

3) Kept the best for dessert... found a freaking SERVICE MANUAL !!!!!!   29 pages, of which 4 pages on the CD part specifically. A foreword in Brazilian language, that would be Portuguese, not exactly my area of expertise. Then 3 pages of goodness : A full pin-out of the 3 connectors on the board, then a block diagram of the player, then... a schematic !  YES !!!!!  I am saved !!!!    A nice schematic at that : integrated circuits are represented with their internal bloc diagram, and pin names. There is even a handful of waveforms.

I can't believe there is so much good stuff available for such a junk piece of consumer electronics. I am shocked ! 
Well, thank you Philips then, I appreciate very much !

[Runco990]

Hi!   

I used to work as a tech for a major retailer for a few years.  This type of unit was the absolute worst trash of the era, but it did work.  We HATED them!

The most common faults in these are the ribbon cables, motors and the optical pickup.  Just because it lights up does NOT mean it's good.  We changed a TON of them.

The ribbons go bad especially the one for the tray, causing a not read.  The spindle motors often would not spin up fast enough to catch and complete the disc read cycle and the cpu would give up.  Sometimes a little oil fixed this, but for warranty repairs we changed parts for new.  Yes, this series of crap often broke during warranty.  Philips or not.

Maybe this insight will help.  You could have another problem, but these were the most common.  I have a shoe box full of pickups from those days.  The boards actually were very reliable.  It was remarkable what they could build for a nickle and sell for $99.

Hope this helps a little.

[Vince]

Thanks for your input ! 

The bad news is that from what you say.. basically it could be anything and I have no way of diagnosing reliably the problem, unless I had spare parts to swap.

Spending, say, 30 Euros for a pick-up is tolerable... only if I can be 100% sure it will fix the problem. But if it needs to be 30 Euros here, then 20 Euros there, and another 50 Euros there, ans still no fixed... then it's no-no. Not going to throw 100 or 150 Euros of random parts to finally get it, maybe, fixed...  the whole system is not worth that much... especially since the VFD display is so worn out you can't see squat.

So... I can only hope I will get lucky and find something so wrong with it that there will be no doubt as to what the fix is.

So, to sum it up... it's not looking good   
At least I have the schematic now, so I stand a fair chance, it's not 100% hopeless.

Crossing fingers...

[laneboysrc]

The player you have is of course a very different design as the swing-arm, as you assumed correctly.

The swing arm is a single-beam design, the drive in the 3-disc changer is a 3-beam design. The difference is that the single beam "weaved" slightly left and right of the pits to stay on track, while the 3-beam design has a larger reading spot and two auxiliary light spots slightly to the left and right of reading beam that are uses to stay on track. Very different technologies.

While development, in order to debug drive related issues, we always had the drive separated form the changer mechanics. This way it was easier to see what is going on and measure things. The mechanics is just a few wires for motor and switches anyway.

If the disc is not detected, then things fail early on. Usually the sequence is roughly as follows:
- MCU moves the lens to the inner-most home position
- MCU moves the lens slightly outwards to ensure it is on data part of the disk
- Laser gets turned on
- MCU initiates the focus ramp. If a disc is present and the drive can get focus, the chipset locks the lens into focus and notifies the MCU
- MCU turns on the disc motor (note: some designs switch the disc motor on before starting to focus, that depended on the drive manufacturer and what worked best)
- After a spin-up timeout, the chipset usually reports to the MCU that the PLL is locked onto the data stream
- MCU starts reading meta-data to find out where it is on the disc, and if necessary jumps back into the "TOC" area to read the table of contents

So what exactly does your drive do?
You see the laser, but does the lens go up/down? (without disc this is easy to see)
Does the disc motor turn on?
Do you hear some kind of ticking sound and see the lens assembly (we called that "sledge") moving in or out?
Hint: if the drive has issues with reading data and staying on the track (e.g. weak laser, poor disc, dirt on the lens, mechanical issues with the lens actuators) then often you can hear some faint noise emitting as the laser skids across the track and the chipset is trying to counter-act.
When things are quiet and you hear regular ticks, then the laser is staying on track nicely and the MCU may try to jump back and forth to read a particular area on the disc.

cheers, Werner

[shakalnokturn]

Those CDM12 / VAM1201 do fail quite often.

If you have doubts on the photodiode side check the amplitude of the "eye pattern" of course for that you need to know the correct amplitude.

Sometimes blowing down on the side of the lens with dry air you can remove some dust from the prism and gain the missing light return.

If the pickup is on the end because of dust or just end of life sometimes you can get a little more working time out of them by upping the laser power.

[Vince]

The player you have is of course a very different design as the swing-arm, as you assumed correctly.
The swing arm is a single-beam design, the drive in the 3-disc changer is a 3-beam design. The difference is that the single beam "weaved" slightly left and right of the pits to stay on track, 

Yes  indeed, I seem to remember seeing the head of the swing arm constantly "oscillating" slightly while playing the disc.

While development, in order to debug drive related issues, we always had the drive separated form the changer mechanics. This way it was easier to see what is going on and measure things. The mechanics is just a few wires for motor and switches anyway.

Yes, the changer mechanismis in teh way tiem and makes my life extremely miserable, indeed ! 
I would LOVE as you say, to separate it from the player itself ! Trouble is... I just don't see how I can do that! 
I mean, the changer and the player appear to be sooooo very much intertwined. For instance, the various switches that report to the MCU about the mechanical state of the changer... they are NOT part of the changer, no. All these switches are soldered directly on to the PCB / electronics !    So if I get rid of the changer, the MCU won't see those switches doing there thing anymore, so MCU will freak out and not allow for a normal operation of the player.... I am stuck with the changer mechanism it seems ! 

If the disc is not detected, then things fail early on. Usually the sequence is roughly as follows:
- MCU moves the lens to the inner-most home position
- MCU moves the lens slightly outwards to ensure it is on data part of the disk
- Laser gets turned on
- MCU initiates the focus ramp. If a disc is present and the drive can get focus, the chipset locks the lens into focus and notifies the MCU
- MCU turns on the disc motor (note: some designs switch the disc motor on before starting to focus, that depended on the drive manufacturer and what worked best)
- After a spin-up timeout, the chipset usually reports to the MCU that the PLL is locked onto the data stream
- MCU starts reading meta-data to find out where it is on the disc, and if necessary jumps back into the "TOC" area to read the table of contents

Thanks.

So what exactly does your drive do?

My player does :

1) Turns laser ON, I can see it.
2) Then moves the Focus motor / lens up and down, once.
3) turn on the spindle motor : disc is rotated by about a quarter of a revolution, no more no less. Just a quarter. This lasts only a fraction of a second of course.
4) then it fails to detect the disc/read data I assume, because the changer then switches to the next disc, and the whole cycle repeats.

You see the laser, but does the lens go up/down? (without disc this is easy to see)

Yes, I can very clearly see the lens move up down quite a bit, 100% certain.

Does the disc motor turn on?

Yes, discs start rotating, but does only a quarter turn and then stops. So it does turn but only for a split second, literally.

Do you hear some kind of ticking sound and see the lens assembly (we called that "sledge") moving in or out?

Nope... the whole action ( turning on the laser, moving the lens up and down, then spinning the disc for a 1/4 turn) lasts only a second or two, so not much time to hear anything really...

[Vince]

Those CDM12 / VAM1201 do fail quite often.

Ah well.... point taken. I will not hesitate to come to this diagnosis then, if it all points to it. I guess a crap head is only fitting in a crap stereo system eh ? It's all consistent...
I am just glad that somehow the head is still available 20 years later, not only that, but also that it costs only 30 Euros, a rather moderate price by todays standard, especially considering it's a "high-tech" part, not a screw or a capacitor...

If you have doubts on the photodiode side

I don't really know what I am doing as you can see, but I thought that well, the photodiodes are part of the closed "loop" responsible for reading the disc, so it is as suspect as anything other component in that loop. Laser, lens, focus, HF amplifier, decoder chip... can be anything really...

Sometimes blowing down on the side of the lens with dry air you can remove some dust from the prism and gain the missing light return.

Hmm yes, I do need to get some cans of dry compressed air... I need to ! 

If the pickup is on the end because of dust or just end of life sometimes you can get a little more working time out of them by upping the laser power.

The schematic, doesn't show any adjustment for the laser output. Indeed on the board there is not trimmer anywhere to be seen.
However on the head itself, on the flex PCB, as you can see from prior pictures I posted, there are a few component, among which I can see a minuscule trimmer. I assume it's meant to adjust the laser output.... could try playing with that.

check the amplitude of the "eye pattern" of course for that you need to know the correct amplitude.

" eye pattern ".. hey that rings a bell !!!  Look at the schematic !  There is  waveform attached to the HF amplifier output, labelled " eyte pattern " !    It gives interesting info : the signal should be 900mV peak to peak, and riding over a +2,5V DC offset.

So basically watching that eye pattern / HT amplifier output, will tell me if there is any data being read at all, and if the amplitude is too low I will know that problem comes from there !  That would be giant step forward, let me measure that !!  Gimme some time... need to fight with the changer to do that... and the disc rotates for only a split second so it won't be easy to measure that eye pattern ! Glad storage scopes exist...

Stay tuned !!! 

[shakalnokturn]

Those pickup assemblies were used in a wide range of players ranging from your cheap 'n awkward carrousel to higher end models like the Marantz CD6000.

If you manage to catch the eye pattern during CD detection and it looks too low in amplitude try turning the trimmer on the pickup CW 2-3° at a time see if that improves things.
(Yes it is the laser power.)

[Vince]

Those pickup assemblies were used in a wide range of players ranging from your cheap 'n awkward carrousel to higher end models like the Marantz CD6000.

Oh OK, good to know ! I was being a bad mouth then, sorry for that ! 


OK, so here is the result of probing the output of the HF amplifier for that eye diagram....

The PCB person at Philips was kind enough to add a test point for the HF output, and label it clearly on the silk screen. Someone was thinking, refreshing...

I was excited at teh idea of posting a nice screen shot of the eye diagram for us to discuss about... however I won't need to, because what I saw on the screen doesn't exactly warrant a screen shot, to be honest !

What I saw can be described simply using words : I saw NOTHING !


HOUSTON, WE HAVE A PROBLEM !!!!   

OK let's be more specific...  I did see "something" : the 2.5V DC offset. Well, almost 2.5V. I don't know, maybe 2,3 or whatever... good enough for me.  However it was a flat line, a solid line.... ZERO signal riding on that offset ! Dead as a dodo ! 

Tried all I could think of to see that signal, but to no available.

I guess I should be happy ? No signal at all, must be easier to diagnose that a weak signal.

So unless I succumbed again to a red herring and I simply failed to capture that signal on my scope... let's say that I was semi-competent in operating the scope, and that there is indeed no signal at all there.

Thoughts :

1) I do get the 2.5V offset. From the schematic, this doesn't mean that there is anything coming out of the photodiode, since they seem to be AC coupled. So the DC offset is just the operating bias intrinsic to the amplifier, nothing more.

2) Even if the diodes or laser were tired, or even if one or two of the 3 diodes failed completely, I would hope that at least SOME signal would manage to make it through, and be visible at the output of the amplifier... SOME signal. But I get a solid, flat line... zero signal at all.


So, what to think of that ?!

I guess it means the HF amplifier is kaput ? So need to investigate the amp.... 

- First probe its input to confirm that there is no signal present.

- Then I could inject a sine wave with my sig gen at the input of the amp, and see if it comes out the other end...

- I don't know how much gain this amp is supposed to have ?? x10 ? x100  ??  Any clue / ball park  figure ?
It comprises no less than 5 bipolar transistors, so I would expect some serious gain out of the thing ?!
I won't claim to understand in fine detail how this amp works exactly, Looks like the first and last/fifth transistors are just buffers. The fourth has the same 1.8k resistor in both its collector and emitter, so I guess it provides no amplification and just sets the 2.5V DC offset (supply is 5V).
So that leaves only 2 transistors, second and third, to do the - actual amplifying.

- There are 2 big electrolytic caps in this amp, that might therefore constitute suspects isn't it...  both are 47uF caps. There is one right across the 5V supply. I doubt this one is not bad otherwise we would not get our DC offset at the output, and it would keep the entire drive from working...  However the second cap is deep into the amp circuitry itself, so could affect it without affecting the rest of the player. As you can see it's located on the emitter part of the 2 and third transistors, the ones that I believe do the actual amplifying. So worth a look.

- Oh, looks like I might be mistaken... maybe the first transistor is not a buffer, but does amplify too !  I mean look : there is some feedback from the 2nd and 3rd transistors, that gets summed, and fed back to the first transistor..... OK, so maybe then we have a 3 stage amplifier... so I would expect some decent gain out of it, right ? x100 or more....



I could also do a sanity check with the DMM, to test the 5 photodiodes, make sure none are shorted or open-circuit.

I could also power the spindle motor using my lab power supply, this way I would have all the time in the word to look for that eye pattern !    Of course for that to happen I would also have to bypass the decoder chip so as to force the laser to turn on...


So lots of things to do !

[shakalnokturn]

The VAM1202 laser assembly can be found for less than 20€ on ASWO store, similar price for the full "transport" on eBay from Germany.
(No affiliation with either of the evil f¥€£ers...)

[Vince]

Wow, thanks for the link ! Sadly it looks like ASWO only sells to businesses, and I don't own a business.. however on EBay boy are you right, prices are SILLY LOW !!!   

Be it the pick-up or the full transport, in either case it's 6.5 Euros, free shipping, brand new part, from china ! 

From German sellers, same price but 15 Euros shipping (from France, for such a small and light part... really ?! ... ).. so basically they advertise the same price as Chinese sellers so as not to be dismissed by potential buyers... and inflate their shipping charge to compensate and make a benefit ! Still, that comes out at 23 Euros for a brand new unit from Germany, hence fast shipping and easy returns if needed.. German sellers tend to be more honest and less troublesome than French ones... just not the same mentality.

OK so to sum it up, worse case scenario... I replace the entire transport and for 23 Euros all in, I am good ?! I rather like that ! 

But let's not take the easy route... let's do some more troubleshooting, just in case it's an electrical issue ! And even if not an electrical issue... I will still check that HF amplifier anyway, just for fun and to get some hands on experience...might be useful for the next CD player, should there be any.

Anyway, I just checked all the diodes. Disconnected the flat flex from the board, and did a diode check with the DMM.

Here goes :

- Laser diode : 1,4V drop, in reverse I don't get an open circuit, instead I get a 1.6V drop. Not worried about that : there is active circuitry on the flat flex. 3 active components (SOT23 packages) and half a dozen passive components. So it's messing up the reading.
The laser is firing so I hardly expected anything seriously wrong when testing it...

D1 and D5 (the two "satellite" diodes) : 0,87V and 0,93V respectively.
D2 to D4 : 0,94V and 0,95V and 0,90V respectively.

In reverse, open-circuit as expected.

So the photodiodes are not dead at all... might still be tired I don't know, but definitely no catastrophic failure at the least.

[Vince]

.... I don't understand...probed again the output of the HF amp.... and now I DO see a signal !  Weird. See below...

Probe is set to x10 so the gain is 2 Volts per division.  Time base is set as per required in the schematic : 0.5us.

So :

- We still have a DC offset of about 2.5V.. well more like 3V if I had to eyeball it... still in the ballpark, good enough for now.

- Signal is not an eye pattern.. .it's a perfectly stable signal. What you see on the picture is what I see in "run" mode or even in analogue mode. I didn't have to "freeze" / single shot the signal to get that picture. Signal doesn't look "fuzzy" like an eye pattern would..

- Signal amplitude : schematic, I misread it... it says 800mV not 900mV (writing is not very sharp...), but most importantly, it does NOT say that a healthy signal is 800mVpp... no, I just realized it says : must be AT LEAST 800mV , quite different !!!
In this case, as you can see on the picture, amplitude is only 300mV...so no good !

- I get that signal shape as soon as the laser is turned  on / disc is loaded... disc NOT rotating. Then when the disc is rotated a quarter turn... the signal does NOT Change in anyway shape or form ! Still as stable and monotone as you see it on the picture.. no modulation/variation at all... which I would have expected once the disc was being rotated. So that's suspicious isn't it...


- .... and I ket the best, the most fun, for dessert : just for completeness, I probed the output with NO DISC present. Guess  what ? Yes, I STILL do get that sine wave shaped signal, 100% identical, zero change ! Disc or no disc, I get that signal !

- I then probed the INPUT of the amplifier (with a disc present) : I get a slight DC voltage, 100 or 200mV, but ZERO signal. Flat line.... yet still get that 300mVpp 600kHz "sine wave" at the output !!!


So ? No signal from the pickup ? Amplifier losing its mind and "oscillating" on its own at 600kHz ? Hmmm.... am I cursed or what ?....

[Vince]

Just did one more quick experiment : I am a pervert little bastard, I know.... couldn't help but tried to.... try to... run the player with the flat flex of the pick-up DISconnected ! I pulled the bloody cable altogether and ran the player like that. this way, no more laser and photodiodes, zero signal whatsoever feeding the HF amp, guaranteed  !     Guess what ? YES, the output of the amplifier is STILL showing the exact same signal as before, 100% the same !   

Oh I am loving this CD player more and more every minute, so much fun, never a dull moment ! 

So looks like the amp has some problem eh ? Or my scope probe is dodgy and I am picking up some 600kHz signal from somewhere... but unlikely, 300mV is a lot, and the signal is very strong and stable. Has to come from the amplifier output...

So will carry on investigating this amp... might get lucky, might indeed be faulty and I may not even need a new pickup, who knows...

[shakalnokturn]

Maybe be a cracked ribbon cable then?

[tautech]

Maybe be a cracked ribbon cable then?

Or crook joints onto components.

From early in the thread:

[Vince]

OK, just spent an hour or two on the thing.

- Ribbon cable :  looks fine to me. No kink, and when I tested the diodes and servos, I checked each and every pin, and got perfectly stable and reliable readings each time I barely caressed the contacts oin the flex with the DMM probes. Just a light touch and I immediately a plausible and stable reading, each and every time. So if it's dodgy, it must be a very rare intermittent contact, not a solid open-circuit, whereas when I measured the HF output I got zero, reliably, all the time... and I spent quite some time looking at that output. So I would say the cable is, at the least, good enough to be ruled out. At least it can't explain why I get no signal whatsoever.
Also, I reflowed all the tiny SMD components on the flat flex. Was fun. Cleaning the flux residues without getting any inside the pickup assembly was fun too. But it went OK.  Also reflowed, for the second time now... the joints of the flat flex connector on the main board. Also put some contact cleaner in the thing. Result : no change, still no signal.

- So I move on to the amplifier itself. A good night of sleep passed, but the symptoms are still there, 100% repeatable. That's a good thing I suppose... I like solid failures, easier to chase...
I did some more experiments :

1) I grounded the input of the amp : the 300mV oscillation at the ouput, immediately goes away, and I am left with just the expected / wanted 2V DC offset.

2) Took it a step farther : now I am testing the HF amp all by itself.  I disconnected again the flat flex, so have direct control of the input.
As for power supply, it runs on the +5V, so does the decoded chip and laser. Probleùm is that the main board powers that rail ONLY when it asks the player to play a CD. Since the player fails to detect discs, it works only for one or two seconds at a time, grand maximum, then 5V supply is turned off, then it loads a new disc and I have again 2 seconds worth of 5V at my disposal. I got bored....
So I just powered off the main unit, and powered the player, only the 5V rail (don't need the 10V rail, it's just for the mortors), from my external lab power supply. Now the amp is powered permanently, so I can work on it with so much more comfort ! I can fiddle with it as much as I want.
Power supply indicates a 90mA current draw, which seems about right, for a decoder chip which is a complex beast, and an amplifier made of 5 bipolar transistors. So looks fine to me.

With the input grounded, the output is again stable, no oscillation, just the DC offset. So looks to be working "fine".. well as it was when powered by the main unit. So looks like I didn't screw anything up with my external supply... a good beginning.

Then I proceeded to inject some signal at the input of the amp. Not knowing what amplitude, I just took a rough guess : eye diagram says the signal amplitude should be at least 800mVpp. That's a minimum. So this means it's normally larger than that. Let's just round it up to 1 Volt for convenience. The amp has 5 transistors, say that 2 or 3 of them actually amplify the signal. Let's be conservative and pretend each stage has a gain of x10 , OK sorry, 20dB       that means the overall gain of the amplifier is 60dB. So that means the input signal is supposed to be 1mV ! With my x10 probe, that would be 100uV. Scope sensitivity is 2mV maximum, so the signal would be lost in the noise floor basically ! 
No wonder I couldn't see anything when I probed the input... even if there was something to be seen... I just could not see it !

So I guess probing the input of the amp, was not so smart. Non wonder the schematics direct you directly to the output of the amp, and don't make you waste time checking the input...

So, I started with a 10mVpp sine wave at the input.  Low enough to not cause damage by reverse biasing anything in the decoder chip, and worse case scenario, the output of the amp would clip and that's it. But I should still be able to see a distinct signal, and a large one.

So, with 10mVpp at the input, what did the amp give me at the output ? I ask ?  Yes, NOTHING AT ALL ! Still zero signal riding on the DC offset ! No-thing !  Now I am open minded and eager to learn, so if you can find a reasonable explanation, I am all ears of course... I swear I am. Failing that, I will declare the amp kaput and keep troubleshooting it further.

I don't see how a new pickup could be of any help if the amp can't amplify AT ALL. Hell, it's even worse than that I guess... it it weren't amplifying at all, at least that means the output would serve me my 10mVpp input signal untouched... but I guess nothing. So technically, I get minus infinity dB worth of "gain"... input signal just isn't even making it's way to the output, amplified or not.

So whats do you think ? Am I again chasing a red herring, or does the above hold water ?!.... 

In the meantime, will try to trace my 10mV input signal inside the amp, see where it disappears...

[Vince]

Forgot : as for the frequency of the input signal, again no clue. Don't know the pass band of the amp.

So I used 500kHz, based on :

- The eye diagram wave form asks to set the sweep speed to 500ns / DIV. At that speed, 500kHz gives you a couple periods of the signal.

Oh, just realised maybe I ough to pay more attentin to how I apply my signal to the input old the amp... ground of the sig gen is connected to the ground of the amp... so  obviously my 10mV signal is not going to turn on the base-emitter junction of the input transistor, eh eh... ooops.
There is some feedback from the other transistors, that I guess lift the base of the first transistor above the turn on point. I guess. Could measure that of course. So I guess I ought to AC couple my sign gen to the input, using some cap.

OK rookie onboard, guy needs to turn on his brain before turning on any piece of equipment...

[Vince]

OK, some more.

Corrected my rookie mistake : added a cap to AC couple the sig gen to the input of the amplifier. Oh magic, the amp woks now... I see my signal at the output. I almost cried.

Then I spent some time experimenting with the amp to characterize it, see if it held water.

BTW, the old swing arm and Sony amp I repaired recently are already in service. Sitting on the shelves among the instruments, keeping me company while I was working on the amp. I appreciate them very much !  Glad I could repair them...

So, here are the results :

- DC offset : 2,80V instead of 2,5V in the manual. Good enough for the purpose I think...

- Bandwidth : still flat at 1MHz. Can't go higher than 1MHz, sig gen limitation.... my old Philips is from the '70s so it was never going to set the world on fire. Still, I Googled and found out that the bit rate for CD players was about 1,4Mbps. So since it's still flat at 1MHz, it's probably going to be fine enough at 1.4MHz...

- Signal excursion : can go up to 4,0V maximum, and down to 1,8V. So 2,2V of excursion. Manual asks for 0.8V minimum, so at 2.2 we should be all good. If you allow for some distortion (happens at the bottom of the wave form first), then the absolute lowest it will go, is 1,4V. Maximum stays at 4V, no matter what. So 2,6V of excursion.

- Input sensitivity : maximum amplitude of the input before the output gets distorted, is about 9mVpp. A little below 10mVpp anyway. Sorry but at those amplitudes the noise of the scope (and crappy sig gen, and noise picked from all my wiring on the bench, combined...) is very high so the picture is quite blurry.. taking precise measurements is a fine art... or an illusion, depending on the skill level of the operator...

- Then measured the gain, well, the amplification factor. See picture.  I gave it my best shot. Switched to digital mode so I could add some averaging to clean up and stabilize the input signal as much as possible. Still not perfect.  That gives an amplification of about x225.
Then I realized after the fact that this was not 100% correct : although the input signal is AC coupled of course, apparently there is still 2 minor divisions worth of negative offset. Due to the scope front-end of course... to be expected at the highest sensitivity factor, 2mV... guess I could calibrate that out, if I could be bothered...
Anyway, so the actual amplification would be a little lower. Maybe 200 or so. Anyway, it's roughly 200, precise number is irrelevant.

200 sounds good to me ? x10 would have been ridiculously low, suspicious. At the other end of the spectrum, had I measured 10,000 I would have screamed operator error !  But anything between x100 and x1000 sounds reasonable / plausible to me. So at 200 / 250, sounds good.

So... I will declare the HF amp... WORKING ! Congratulations M. amp ! 


So I guess it's time for a new 23 euros pickup from Ebay Germany... but before I inform the old man about my findings, I will try to play with the trimmer to increase the laser output, see if that does anything. Don't even know which way to turn it ! Guess I can stuff an am meter in there and see what it's doing exactly.... I don't like doing things blindly, makes me nervous ! 

[Vince]

OK, back from fiddling with the laser.

Turns out I can't measure the laser current directly. The 5V sent by the decoder chip, does not power directly the LED. It only powers the laser circuitry located on the flat flex on the pickup. Said flat flex has a separate 5V supply. Had no convenient way of sticking an am meter on that supply, nor even on the ground return. So instead I just measured the overall current drawn by the player, at the output of my bench power supply. Normally about 90mA. Turned the trimmer slowly, a bit in one direction, no change in current draw. A bit in the other direction... still no change. Turned it a bit more, and bit more... still no change hmmm.... strange. Then I did a full turn, by mistake, as there were NO stops at the ends of the track !?  As the wiper suddenly passed from one end to the other in one go... current changed at last, jumping to 250mA ! 
So I rushed to immediately put the wiper about half way on the track, where it was at the beginning... but current would stay at 250mA, would not budge. So basically I went from "stuck at 90mA" to "stuck at 250mA".... in either case the thing was stuck !  So I think there was something fishy about that laser driver circuitry to begin with, and might have been the root cause of the problem... probably caused a  low laser output hence no read.
After a few seconds at 250mA, eventually the magic smoke and smell, escaped from one of the 3 transistors on the flex !
So now there is DEFINITELY something wrong on that circuitry !       

I guess I could reverse engineer the circuitry and replace that transistor.. assuming I can identify it ! Not a given sadly, with tiny SMD parts like that 
Maybe I can replace that transistor and who knows, maybe the laser will be driven properly and give a decent output and the drive will work ?

I don't know if I can be bothered doing that... I guess I should, I should... but the pickup being available and so cheap... if I replace that transistor, I will have to order it from Farnell and the minimum order of 15 Euros will be more or less the price of the pickup... yet no guarantee that it will fix my old pickup. So.... I guess the sensible thing to do, is to order a new pickup ?!................................

Looks like this repair is coming to a conclusion...