Intermittent problem Philips CD150

So, I got a CD150 on my bench, it looks nice and clean, doesn't look very used at all, so I think it's worth restoring and modding. Laser tests out fine (eye pattern is 1,5Vpp and laser is adjusted to 55mV) and player came to me working intermittently, that is, losing track very often, especially on CD-R. Measured voltages, all normal.

I decided to clean first, obviously, laser lens and lubricate hall motor bearing. No change. Then, installed new caps on servo board first. Especially the infamous axial blue Philips cap seemed to me a possibility. But, although there was an improvement (reading CD-R better), it still lost track sometimes (especially when trying to change to a track further away). When it loses track (or is it focus? Is there a way to monitor some signal to see if focus is lost, for example? FE Lag signal? Shall I scope that?), it makes a noise (swing arm beating against limit) and stops spinning, showing Err on display. One single time, it also showed Err 18. But I can't get an error code list anywhere for this player.

So, decided to recap most important caps on decoder board and display board, too (because I remember that there are some blue axial caps in and it contains the micrcontroller, so better go on the safe side).
It now seemed a little bit better, but still would easily stop when I would try to advance to a further away track (forward or backward).

Decided to try service program - would easily go to service position 3 and, strange enough, swing arm can be moved freely all the way without ever losing track (focus?) once. I mean, I tried really a lot, and it would never happen, only, obviously if I exceeded the end of the disc (btw, what happens is very similar to the intermittent problem while track changing). But the noise the swing arm makes when I command it back and forth in this service position, it's different from the noise it makes when it loses track while changing track manually in normal operation. Let me explain: in service position 3, it moves in 64 steps according to SM (you can hear that). When normal operation and I tell it to advance, say, 5 tracks, it makes like a sweeping, almost singing sound until it arrives and then seems to "lock in" and then starts playing. In the even of losing track, the same happens, except that close to the end of the movement, the singing gets deeper in sound and then it loses its position and bangs against the limit, disc stops and shows Err.
Strangely enough sometimes it works well and even manages to travel from track 1 to 20 and later back to 1, for example, without failure, several times in a row.

I don't have any CDM-2 based player here at the moment (the RAFOC is CDM-4), so I can't compare: is it normal for them to "sing" while the swing arm is moved further away during changing track manually? CDM-4 based players usually don't do this, for what I know, or just a little. Could this be some kind of parasitic oscillation of the radial servo? I mean, the easiest way to make the intermittent problem happen is when doing search forward or backward, first, it moves in steps like in service position (no singing sound), then it advances more, starts to sing here (and normally not loses track), but as soon as it advances very quickly, there it often loses track after some seconds. There, the sound is stronger and deeper.

To eliminate mechanical failures, I disassembled the whole RAFOC, cleaned and lubricated, nothing looked wrong, also cleaned the inner laser lens of the pickup.
What is strange with this behavior, is that it hasn't failed reading TOC once.
Or starting to play when press Play, never fails.

Tomorrow I will try to let it play alone a few discs from start to end to see if it fails on its own or if this only happens when we change the track.

Any ideas?

P.S: I might have found something by mere chance: in SM for another philips player based on CDM-4, it says as explanation for one of the errors: No TL pulse received within 0,5 sec during track jumping. Check the RE Lag circuit (Error during Search or Next/Previous).
This one sounds exactly like what is happening with this player! When I find some time, I will try to measure the TL signal. I'll also check the RE lag circuitry, which is explained in the SM, I think I'm onto something here.
 
Last edited:
Yes, I changed them. The 33uF I changed twice, to different models and higher values, as is used in later cdm-4, just to be sure (as I don't have Philips type), because of what I read on the Internet (personally I'm yet to find a Philips servo board where it really matters, all the ones I serviced until now were happy with whatever new cap I would use) that some types of caps would give bad results, but it was the same result. If you think it could be related, I can try different cap once again. But I don't think, because typical behavior when this cap is bad is that it has trouble reading TOC, difficulty reading and focussing in general. Here it seems to read just fine, even CD-R and scratched, also TOC not failing, but then on track change from time to time - it simply quits its job 😉 and you have to start playing the CD from the beginning lol. Today I'm going to let it play alone to see if track change also fails, I'll report back then.
 
An update: when I was trying to let it play, I discovered that there was no music. In the beginning, I remember it had once failed to play, I assumed bad connection at RCA, so I resoldered, since then I hadn't tested music out. I measured and found that -18V rail was missing - no power coming to voltage reg. I checked and found a cold solder joint, at least it's what I think it was, to be on the safe side I installed a wire link from rectifier diode to reg. Problem fixed: it now plays music again. BUT: intermittent track (focus?) loss is still happening - sometimes more, sometimes less, but for what I can tell, ONLY when you operate track next or previous or search buttons. It's playing happily alone right now. Until now no skips, even scratched cd and cd-r, whatever, even difficult discs which make some players with problems fail immediately it reads well (though haven't heard for long time yet). This is very odd. Am going to follow the "lead" I might have found in SM regarding the RE Lag circuit.
 
Well, took my oscilloscope to do the measurements SM suggests for CDM-2 servo board.
This is what I found:

RD Signal (Focus ok) stays high, even while error occurs. SO, not a focus loss.

Eye pattern is 1,2V pp even on a hard to read CD-R, looks good, but when error occurs (and sometimes during track change), it seems to lose some of its waves (lines), is that normal? Doesn’t lose amplitude.

HFD signal goes low when track change asked (guess that's normal), and stays low if the error occurs.

TL and RE-dig show the varying square wave that responds if I knock against player, for example, so that should be normal, either.

DDOS: here, something interesting: shortly before failure occurs, this signal goes high and stays there while error occurs and player stops! Best observed when I ask player to change from track 1 to 10, for example: then, signal goes low, as requested, and if track change is successful, only when player starts again to play, signal goes high. But if error occurs, the signal goes high shortly before.

From what I understand, this signal is important, as while it is low, HFD drop outs are ignored, which is what must occur during track change. Could this show some possible problem with the responsible circuitry? I mean, if it would stop "surpressing" drop outs earlier than intended (which means, while track change is still ocurring), the HF detector would report HF loss and would stop everything, no?
Or am I just imagining this?

Haven't had time to properly test the RE_lag circuitry: the only thing so far is that the required impulses during track change requests appear in fact at test point 43.
 
Argh, sorry. False alarm with the DDS signal. I somehow stupidly measured on the wrong pin of the IC, I was confused for a second regarding the side of the IC, was like I was looking from the wrong side, so I didn't measure DDS but one of the DATA signals. Obviously, that will loose valid information shortly before failure occurs. Now I measured DDS and it is fine, does exactly what is intended.

RE-lag also doesn't take me anywhere, for what it seems: I tried injecting "low" signal to transistor so to test function and while playing, this causes some "trouble", so I guess it must be working. Also, transistor measures out fine when power is off and all signals while operating seem normal. I can safely assume that the capacitor which is being shorted by the transistor to clear "memory" is ok, it's not an electrolytic.

I will now do the tests they describe in SM for RE and DAC speed control. But I think they will measure out fine.

Btw, the problem can really be produced very often if we use the Search buttons and keep them pressed until larger jumps are done and leave it there like three seconds or so, then (at least with this CD-R), it almost always fails. While normal track changes, one by one or just a couple each time, normally succeed, only fail sometimes.
 
Well, just resoldered almost all solder joints on servo board - no change. Also changed the "infamous" laser supply cap once again (to 33uF "garden variety") - no change either.
Now, I may be up to something. I'm pretty sure that the problem is related to radial drive or radial processor/circuit. When I do the test described in SM for radial error correction (inject +5 and -5V with 120k to pin 5 of L272, then it doesn't happen like described - the unit makes exactly the same failure like happens during search - it exits service position 3 with the same full back movement of the swing arm banging against limit and losing track/focus (back to zero). When I inject the voltage without being in service position, the swing arm moves, btw. I don't know if that's normal, have no reference 🙁
 
For today, that's it. Last conclusion I came to: I suppose the procedure in SM is kind of flawed, as the injection of either positive or negative 5 v makes the swing arm move to either one or the other extreme?

While playing, radial error correction must be working, so I'm not sure what to think about all this. I scoped the radial out. When looking at signal at radial drive opamp L272 out, it looks like a saw tooth wave which reacts if I bump against the player, I suppose that's like it has to look.
At radial processor output, the signal is smaller, can't really identify saw tooth, it has some spikes. I'd have a hard time taking photos of the waveforms as I'm like "poking around" with the probe while observing. Now comes the interesting part: if I make the "anomaly" happen (by searching, for example), in the exact moment of the "anomaly", the radial out signal goes completely crazy: even at radial processor out there's a huge chaotic waveform (of higher frequency than before and irregular spikes), like about 7-8 V from peak to peak. At L272 out even worse, obviously. As this only lasts about half a second, I can't take a pic at all of this.

What can I assume from this? Any opinions? I had though about exchanging L272, just in case, but I think it's useless, as I think it's working fine, the problem comes from before. Now, can it be the radial processor? Or can it also be some kind of mechanical defect, making the whole circuitry go nuts when used for longer / more far away searching?
I tried to read through description how the process works compared to normal playing, but it's too complicated for me. Any hints?

Shall I simply swap the RAFOC with a known good CDM-4 I have here (has the normal motor, so no chance of simply swapping the whole CDM, unfortunately), that's some work and it's delicate, but I've done it before, this would allow me to understand if it's in the circuitry or in the RAFOC (it can't be the laser, but it could be the swing arm, right?).

For now, I'll let it play a disc or another to see if ALONE, without any search/track change requests it doesn't lose track. Then I'll report back. Tomorrow or Friday, if I have time, I might do the RAFOC swap.
 
First of all: it's fixed! 😀

Sorry for all this monologue, seems I've been talking to myself 😛
But maybe it's going to be helpful to someone, some day.

Good news: it now plays almost everything I throw at it, even badly scratched and "difficult" CD-Rs (those which even some modern players refuse, as they might have been in the sun or burning process wasn't perfect or cheap media).

Bad news: it wasn't the electronics as I thought (although it did have it's problems, which I fixed, as you can read above), it was the: SWING ARM. I had a spare RAFOC with a good laser, good swing arm but bad focus mechanism (parts missing), so I swapped the whole RAFOC, unsoldered the two pins of the focus/lens assembly and swapped the focus/lens assemblies, soldered pins back to the flex cable and mounted all back in. Checked lubrication of bearings. (What has to be taken most care in swapping RAFOCs is not to loose the little balls, not to bend or twist in any way the bearing plate and to avoid electrostatic discharge with putting a paper clip on flex cable end, shorting contacts, and also taking care not to put any weight against the pickup head so to not damage lens and focus springs).

After reassembling, I had to adjust laser current, by controlling eye pattern I settled at 47 mV, gives best result with some "bad" CD-Rs (here the laser starts to "whine" on some tracks, which the other strangely enough it didn't, don't know what's going on here, but it's only two discs, only a couple of tracks and it doesn't even stop or skip, just makes that whining (eye pattern gets noisier then).
Eye pattern has the expected 1,2Vpp on a CD-R, perfect. But this hasn't changed, the other laser also was perfect.
But now, it doesn't stall anymore when track changing, and no matter how "hard" I push it (going from one end of the disc to the other, for example, or making fast forward through the whole disc), the problem is gone, I couldn't reproduce it, except if I hit the player hard from outside, but this is normal. So, it was indeed the swing arm. Unfortunately I don't know what it had. But: this one seems to "swing" back and forth more easily by hand when player off, like it had better balance, the other also moved freely but wouldn't have that swing. And yes, I lubricated the other as required and cleaned. Probably it was mechanically damaged in some way.

A couple of things I learned about the electronics (while I was trying to find the fault):

1) the 1,5uF bipolar cap on these (CDM-2) servo boards at radial drive output is critical for CDM-4 laser. It looked "damaged" and I put a 1uF in, as I didn't have 1,5, and it would work much worse. Had to add a 470nF MKT in parallel.

2) the (in)famous laser out filtering cap: I tried a couple of different types of caps and different values. I settled on a higher value (100uF) as it would work best (and is used in some newer CDM based players, 47uF is standard, this one had 33). But differences were really small. And no, regarding laser current, nothing changed (as opposedly what I read on the net happened to some people).

3) That momentary strange "rail-to-rail" high frequency noise on radial drive output and some other points in the circuit and consequent stalling (and banging of swing arm against end position) when radial lock is lost, this seems to be normal, contrary to what I thought (I just had never scoped it before), as it also happens when you "hit" the swing arm during operation making it lose track, it's the same reaction. This again confirms that the problem of the other swing arm was mechanical: when it was left to simply play a disc without anything, it would play the whole disc, even CD-R, without hiccups! but if you would put it to move by track change or, worse, searching, then it would "miss" track and give up.

3) As I was "suspecting" of the RE Lag circuit (and the "memory" capacitor close to the radial processor, as well as the transistor which would short it), because I couldn't make the "low impulses" visible on scope, I noted a ripple on top of the input signal at base of the transistor. Turns out this is normal: comes from the ripple which is left on top of the non-regulated +10 and -10V rails, as the base is biased with a 10K resistor to +10V, not +5V. This is another lesson for me: don't always take strange signals for wrong.

4) don't poke around with oscilloscope probe at L272 radial drive opamp. I even knew that (there's one topic on diyaudio where someone had this happening), so why was I that stupid? Certain is that it looked and sounded like fireworks and I had to unsolder a L272 from another defective servo board I fortunately had here, otherwise I'd have to order the IC. It blew a hole in the package of L272!! I guess I must have shortened -10V (pin 4) to output.... :O 🙁 this was a minor set back, as I had the IC to exchange, but here's the warning. If you want to scope radial drive out, take the signal close to the diode a bit above, NOT at pin 3 of the IC...

Another adventure ends here. At least for now. Not sure what to do with the player yet. Might mod it when I have some time, but it has limited potential. Might also sell it like this. Or keep it like this. One thing is for sure: it's an entry level player, but I really like them, they are vintage, small format, look vintage (although plastic, but the LED display is great and this player is physically in great shape, laser is as good as new, one day this might be worth quite some money) and even sound quite good (TDA1540).

And, out of this I gained a perfectly working laser diode, which obviously is still in the pickup of the problematic RAFOC. This enables me to dream about a laser diode swap adventure with a great vintage high end player from Onkyo which I have and which has dead laser. Although I'm not sure I'm ready for this, very few people have done it and even less have had success.