Philips CD104 tweaks

Some players do briefly 'kick' the platter at power on but if it spins properly but only with a disc present then it sounds like it has focused on the disc.

We had to have an LPM (laser power meter) to meet Sony requirements but I never really liked them if I'm honest.
 
Hi Karl,
Some CD players will spin at approximately the correct speed (a touch fast) before focus search begins. Some "kick" as you have seen.

First test, does focus search occur? Second, do we have laser light? If not, is there laser current and how much is it? At that point we condemn or pass the laser diode.

The LPM was a necessary evil, they were way more expensive than they should have been. But I learned to use it to save me troubleshooting time. There is a good reason why they were a requirement from almost every distributor of CD players. So like them or not, they were a valuable tool that could save you some time. Of course, an accurate 'scope was also required along with the specific test CDs. I had a few sets and they were too expensive as well. No, test CDs for CD player setup cannot be copied or duplicated.

So, normally the sequence is:
  1. tray end switch in closed position?
  2. laser head position switch closed, then head positioned?
  3. focus search with laser on
  4. focus lock? - spin disc (spindle motor)
  5. EFM data lock to PLL?
  6. lock disc speed to PLL ---> crystal clock
  7. track to data portion lock and play
Philips radial mechanisms don't have a head position switch, so you just make sure it swings to the inside of the CD.
 
  • Like
Reactions: Mooly
First test, does focus search occur? Second, do we have laser light?
🙂 that's the one.

So, normally the sequence is:
That's the order it all comes together. Philips were always the odd ones out in many things.

A decent scope is the must have tool for any CD work, you simply can not manage without one as it tells you so much on how healthy the signal from the pickup is. Even looking at the RF just while it tries to read the TOC tells much.
 
  • Like
Reactions: anatech
Yup!
The eye or RF pattern is like a blood test. It tells you so much about how things are running.

Edit:
Service work isn't easy. It requires training and equipment. Just because someone gets something running has no bearing if it is running correctly or not. Without the proper equipment and materials, you have nmo way of knowing. "It sounds good and I can hear distortion" isn't even close to accurate.

So you fix it free. You are probably ripping yourself off since the equipment is likely not performing as it should. So you get reduced performance and usually higher distortion. CD alignment directly impacts distortion. The signal on a CD is not digital. It comes off as an RF waveform in analogue form. This is decoded into digital, and this is where digital errors occur most commonly. Garbage in, garbage out. Bit perfect? Nope.
 
  • Like
Reactions: Mooly
I installed a Kwak clock and replaced some electrolytic caps with Oscons.


IMG_7843.jpeg
 

Attachments

  • IMG_7842.jpeg
    IMG_7842.jpeg
    979.8 KB · Views: 44
  • IMG_7845.jpeg
    IMG_7845.jpeg
    1.1 MB · Views: 41
  • IMG_7844.jpeg
    IMG_7844.jpeg
    939.9 KB · Views: 41
  • IMG_7841.jpeg
    IMG_7841.jpeg
    842 KB · Views: 40
Hi,
I have a CD104 here with a few problems. From least to worst:
  • when ejecting the tray, the motor keeps running for a few seconds more even after the tray is fully out.
  • the player can only play back discs for a short while. Once it "warms up", it becomes unable to read the discs. If I let it sit for a while, then on the next try it will read the TOC fine and play back discs for as much as 5-10 minutes. But once it is warmed up, it becomes progressively more blind, eventually reaching the point where it just doesn't play anything or even show the TOC.
Also, seeking (searching) seems to be pretty slow too.
- there is no audio output on the RCA jacks. The cables have proper continuity so something else must be broken before the output phase.

The unit seems to have no visible faults, I can't spot any dry solder joints or corrosion anywhere. The only other problem I had is the CDM1 laser board having an extra PCB which is held in place by legs only, and of course it fell out while testing... I soldered the extra PCB together with the legs holding it to prevent this. This made no difference in functionality.

I've done a near full recap on the system since, the only caps I didn't replace were a single one on the front panel (largely irrelevant, the inputs all function correctly) and all the ones on the CDM1 laser board (the board that has the two ribbon connectors on it) due to running out of the correct type caps. But every cap on the power, servo and decoder boards have been changed to brand new ones. Unfortunately, this made zero difference.

I've tried checking the TDA1540s and they both seem to get the same voltage inputs/outputs. The reed relays seem to function in as much that they show no continuity when the unit is off, but they show continuity when the unit is powered.

Could anyone help me with where I should begin looking for problems? I don't own a scope to check any waveforms, only a decent multimeter.
 
I'm afraid a good scope is an absolute essential for any real diagnostics, it tells you so much.

The tray problem could be an issue with any micro switches used to detect when the tray is fully out (and also in).

For the playability issues you need to look at the RF (with a scope) and also use a scope to look at the signal coming from DAC's. The board that 'fell out' sounds kind of odd... maybe the player has a hidden history and has been worked on and modified in the past.

You need a scope though 🙂 a good one of at least 20MHz bandwidth (for an older analogue scope).
 
  • Thank You
Reactions: anatech
Yes. 0.5 uS/div horizontal for a proper display.

Resolder the "gribblets". Remove old solder, I apply a touch of liquid flux, then resolder using a minimum of new solder, both sides. Also component leads tha connect top and bottom traces.

No 'scope? Don't touch.
 
  • Like
Reactions: Mooly
The board that 'fell out' sounds kind of odd... maybe the player has a hidden history and has been worked on and modified in the past.
Every component seems original and it still had the blue caps. I don't think it was ever serviced, though a few of the cable harnesses look different from the others.
The small board fell out because I poked it, because I've never seen a sub board connected like that. It was only held in place by the little connectors clinging into it like claws. From the few pics I've seen, this board standard on the CDM1.

I'll try to check the individual ICs voltage inputs to see if there's something wrong in the output stage. It would be good to at least get the audio output working. Then I'll try recapping the laser board. And, of course, fix the gribblets.
 
Replaced the remaining caps on the CDM1 board. Now the player can search/fast forward very quickly and even search backwards as well. It also reads the TOC immediately. So this improved the CD reading behavior considerably.

Unfortunately the unit still "overheats" after a few minutes, and stops seeing discs altogether after that.
 
Small update.
I've measured all the voltage regulators on the power board. Their output voltages are all fine, within a few % of the values listed in the service manual (for ex. -18V measured as -17.7V). The -31.5V measured as -34V but that's an input voltage and the output was correct.

Next I left the player running and then touched each IC for heat. Most got barely warm, but the SAA7010 and SAA7020 were hot enough to burn my finger. I tested some heatsinks and found that when I put a heatsink on the SAA7010, the player is be able to play for much, much longer. Additionally the SAA7020 doesn't heat up so much either as long as the SAA7010 has a sink on it, and the player can switch tracks faster as well.
I have not glued the heatsink on so far, only let it held by gravity while the player is upside down, so the heat transfer may not be the best. Once the heatsink heats up, the player once again stalls. I was rotating two heatsinks to keep the chip "cool" and it has been going on for nearly an hour this way.

So my hunch was right, a chip is overheating. But, why?

SAA7010 pin 15 and pin 28 correctly measure as +12V and +5V respectively. Pin 22 correctly measures as 4.3218 MHz (or thereabouts, it fluctuates a bit, but I'm reading it with a multimeter).
other measurements:
pin 6 - 5V - service manual says 4.7V
pin 7 & 8 - 1.38V - service manual mentions 1.8V for these both.
pin 12 - 3.73V- service manual mentions 4.6V
pin 13 - 3.64V - service manual mentions 4.6V
pin 17 - 3.5V - service manual mentions 2.4V
pin 18 - 4.45V - service manual mentions 5.1V

Additionally there's a component marked as 2501 on the manual (with a value of 22n, not sure what it is). If I do as much as touch the upper lead of it in any way - finger, multimeter, anything - the player nearly immediately stops playing. Tilting it while touching the outer cylinder won't do this, so probably not a bad joint, but even the most gentle contact with the lead will make the player drop out of playing.
 
Hi zyrobs,
The chip that is hot is probably bad. If supply voltages are high, that will cook things.

Use an oscilloscope on the power lines to make absolutely certain they are clean without noise or dropouts. Some are low, probably due to higher than normal current draw.

Use thermal epoxy to attach a heat sink. The self adhesive ones don't stay over time. So not normal epoxy or super glue, thermal epoxy.

Some circuit areas are very sensitive to induced noise. You may have a cracked solder connection, or it just may be a sensitive area of the circuit. You don't have to actually touch an RF circuit to make it fail.
 
  • Like
Reactions: Mooly
I don't have a scope so that's out.

Just by keeping the heatsink on with gravity (running the player upside down), it already gets hot to the touch. So even with thermal epoxy improving the contact, I don't think a heatsink will be a long term solution. Additionally there's still the other issue of the unit not giving any sound.

I looked around for a replacement SAA7010 but only found chinese junk, chips on sale from the US with crazy shipping costs, or entire full boards for 100€.
Well, the unit also lacks all sound output whatsoever so replacing the decoder board might be prudent... but then again, the 7010 being cooked would also explain the lack of audio output, since it is the chip handling the CD Audio demodulation, and the rest of the board could very well be fine...