help philips cd 584 resets

[martinn]

forgot to attach the bottom side headphone section showing the opamp should be in the other way round

PlasticIsGood:
yes, the CDM4 does not like to play CDRs, a workaround is to burn them at slower speed and then it reads them. This pink floyd cd is an original and it was read without problems

[PlasticIsGood]

I find lead-free solder hard to work with, so I always use the old lead/tin stuff. It seems to mix OK with whatever's already there. A dry joint is one where the solder has failed to "wet" the metal. This can be because of contamination, or too low a temperature, but also maybe because it's too hot and the flux is burning before it does its work.

Where the solder meets the wire, there should be a fillet where it's flowed up the wire, not a depression where it looks like it's been repelled by the wire surface. Perhaps a cooler iron, for a little longer, in your case. Fresh flux should always be present at the join whilst soldering, so don't use the iron to transport the solder to the joint.

Soldering instructions seem never to say what temperature is best. I've found around 275C is OK for lead/tin, allowing quick work to avoid overheating the components. Other solders need to be hotter, but it's hard to tell by how much.

I think the 4560 is used because it copes well with a relatively low resistance at it's output. When looking for an alternative, check for minimum load resistance, or some indication of distortion v output current. Better, make yourself a headphone amp.


Some regulators can ring or oscillate if a too-low ESR cap is used close to the output. They need the damping provided by the cap's resistance. Your oscons are probably best placed around the decoder and filter chips, but you really need an oscilloscope to check if the change is for better or worse. Transferring noise from power to ground, especially without absorbing any of its energy, can be counter-productive.

I got a nice Philips dual-trace, 60MHz scope off ebay for not much money. Much more fun than a multimeter.

[poynton]

Originally Posted by PlasticIsGood

Chin up! I'd be happy with a player that refused to play Pink Floyd.

I once had a cd player that would only play 2 Eric Clapton cds and nothing else.

Obviously, some players have taste !!


.

[martinn]

Originally Posted by PlasticIsGood

I find lead-free solder hard to work with, so I always use the old lead/tin stuff. It seems to mix OK with whatever's already there. A dry joint is one where the solder has failed to "wet" the metal. This can be because of contamination, or too low a temperature, but also maybe because it's too hot and the flux is burning before it does its work.

Where the solder meets the wire, there should be a fillet where it's flowed up the wire, not a depression where it looks like it's been repelled by the wire surface. Perhaps a cooler iron, for a little longer, in your case. Fresh flux should always be present at the join whilst soldering, so don't use the iron to transport the solder to the joint.

Soldering instructions seem never to say what temperature is best. I've found around 275C is OK for lead/tin, allowing quick work to avoid overheating the components. Other solders need to be hotter, but it's hard to tell by how much.

I think the 4560 is used because it copes well with a relatively low resistance at it's output. When looking for an alternative, check for minimum load resistance, or some indication of distortion v output current. Better, make yourself a headphone amp.


Some regulators can ring or oscillate if a too-low ESR cap is used close to the output. They need the damping provided by the cap's resistance. Your oscons are probably best placed around the decoder and filter chips, but you really need an oscilloscope to check if the change is for better or worse. Transferring noise from power to ground, especially without absorbing any of its energy, can be counter-productive.

I got a nice Philips dual-trace, 60MHz scope off ebay for not much money. Much more fun than a multimeter.

Tried today with ~300 C for soldering and it is much better. I was using higher temperature, because I saw somewhere that for desoldering it was better and then I continued using it for soldering as well... maybe not the best idea

I replaced the 78L15 TO-92 100mA regulator with a 78S15 TO-220 2A and a resistor (yellow-violet-gold-gold) which had the gold multiplier stripe broken and measured 16 Ohm istead of 4,7 Ohm.

Not to make the trip to the electronics shop just for a regulator I also came back with some capacitors. Basically the power section has all new caps now. And the filter chips have the oscons back.

I'd say the sound is a tad cleaner, but it still won't play all my cd's... it resets (reproducibly) at the same spots of different tracks usually when some drums start to kick in

I'll give it another try tomorrow and stick tonight (again) to one cd that plays without problems

[Mooly]

The content on the CD can't have any possible connection to it cutting out. That sounds like a mechanical/disc issue.

[martinn]

Originally Posted by Mooly

The content on the CD can't have any possible connection to it cutting out. That sounds like a mechanical/disc issue.

I totally agree, but I'd like to cancel out a few possibilities and understand better so please correct me where I am wrong. My reasoning was that in the digital section the receiver or filter either work or not, since there is zero's and ones to process i.e. are content independable. The cd's sampling rate is 44.1 kHz meaning there is always the same amount of info to parse. Therefore content should not matter at all.

What about the tda1541a? It converts the signal to current, so if there is a faulty decoupling or supplying capacitor and does not recharge fast enough the dac runs out of current. Could that cause the player to reset or would the signal just grow weaker - more silent? But I am not experiencing any fluctuations in volume, the player plays normally until it resets at certain tracks.

On the other hand, if I am looking for a mechanical issue. The cd's are fine, tested on 4 original cd's that all played fine before and some CDRs. Interestingly, Mike Oldfields songs of distant earth is brand new and plays to the end without any problems if the lid is open. However, if the lid is closed it causes the player to reset at the beginning of track 4, which has a loud beginning (some spaceship rockets sound). Seems like something is overheating. That's why I've started looking at the power section and voltage regulators, but there's not much left to look at there
Could it be the cd mechanism got damaged during the few dozen times I've assembled and disassembled the cd player or maybe some bent cable? I can't help but wonder why a closed lid would make a difference? And the reset is reproducible to the second.

Getting my hands on an oscilloscope would show me what happens to the signal on reset and would definitely be more fun than using a multimeter. Have to save up some money first

[Mooly]

The DAC has no bearing on the servo operation as such. Nothing is fed back from the DAC to the front end.

All these issues started after you began changing parts... is that correct ?

If supply rail had a problem then that could cause any manner of problems but its not likely to be related to signal content. The difference in current drawn by the DAC wouldn't modulate the supplies in practice.

Again a scope is THE tool to use to really see whats going on.

If its overheating and/or the fault is reproducible to the second then try pausing the player for a few minutes before the problem part of the track and see if it still cuts out at the same point.

[martinn]

Originally Posted by Mooly

The DAC has no bearing on the servo operation as such. Nothing is fed back from the DAC to the front end.

thanks for the explanation, makes things clearer now
this means the problem lies in the cd mechanism or servo part or the power section

Originally Posted by Mooly

All these issues started after you began changing parts... is that correct ?

yes it played without any problems

Originally Posted by Mooly

If supply rail had a problem then that could cause any manner of problems but its not likely to be related to signal content. The difference in current drawn by the DAC wouldn't modulate the supplies in practice.

Again a scope is THE tool to use to really see whats going on.

If its overheating and/or the fault is reproducible to the second then try pausing the player for a few minutes before the problem part of the track and see if it still cuts out at the same point.

maybe the transformer is overheating / failing?
after playing to the end of songs of distant earth with the cover off and repeating the cd it resets itself on track 4
I've let the cd player coold down for 10 min and the same error (note that the voltage regulators all cooled down in this time, but the transformer was still hot). After one hour of being switched off it doesn't reset at track 4 anymore.

Tried the other cd which resets exactly 16s on track 5. If I pause at 10s into track 5 for about 10 min and press play it again resets at 16s. (the transformer is still cool).

If I want to measure voltages on capacitors supplying the cd mechanism the mechanism has to be removed to gain access to those parts. I am a bit reluctant to power on the cd player without all cables attached, but it would be vastly faster if I did not have to assemble the whole player each time I test something. How do you guys measure voltages across the board?

[martinn]

started looking at the schematics a bit more and found a discrepancy. There's a diode rectifier pointing the other way around as in the schematic. Unfortunately I haven't made a picture of the unmodified board, but I've marked the directions before exchanging them (can't 100% exclude I've made an error
though).
update: the schematics on the next page shows the diode in the other direction like it is on the board now


I can't burn parts if I switch the orientation of the diode, right? There will be just half the sinus converted or?

The reset error started before the diodes were replaced! That's why I've replaced them

[Mooly]

If you get the diode the wrong way round it can cause damage because it would rectify the "wrong" half of the cycle and give a negative instead of positive voltage for example. If it were in a bridge then it would overheat. The PCB diagrams are too faint to really see whats going on but you should be able to work it out from the circuit diagram.

That glass ??? looking diode in the picture looks frazzled.

Too many issues seem to have arisen with all this. With the greatest respect I think you have to realise that the work you have done is probably the cause of all the problems.