I was asked to look at a Sugden CD21SE player,
Sugden Audio: A21SE
The owner knew what had happened, they accidently closed the tray with the disc upside down.
Now that shouldn't be any problem for any player, the player should either sit there and say "no disc" or throw it back out... but not so here... the display shows a row of dashes and it just sits there not responding to any command. Turn the power off (fully with plug out and wait 20 minutes) and retry... no good... drawer stays closed.
So top off and gently pushed the drawer out (common Philips mech like the short loader version of L1210) and out comes the disc... all is well.
Confirmed a couple of times that an upside down disc causes the lock up every time. The player finds "something" to focus on presumably the reflective label side, then quickly spins the disc the wrong way which is typical of a servo not locked to anything... and that's it... game over.
I think it has to be a design (software issue) and this was one of the very very first CD21SE's to be available.
Just such an odd issue I thought it deserved a mention. Maybe later players have modded software, there is a big eprom/micro with a paper Sugden sticker attached.
Sugden Audio: A21SE
The owner knew what had happened, they accidently closed the tray with the disc upside down.
Now that shouldn't be any problem for any player, the player should either sit there and say "no disc" or throw it back out... but not so here... the display shows a row of dashes and it just sits there not responding to any command. Turn the power off (fully with plug out and wait 20 minutes) and retry... no good... drawer stays closed.
So top off and gently pushed the drawer out (common Philips mech like the short loader version of L1210) and out comes the disc... all is well.
Confirmed a couple of times that an upside down disc causes the lock up every time. The player finds "something" to focus on presumably the reflective label side, then quickly spins the disc the wrong way which is typical of a servo not locked to anything... and that's it... game over.
I think it has to be a design (software issue) and this was one of the very very first CD21SE's to be available.
Just such an odd issue I thought it deserved a mention. Maybe later players have modded software, there is a big eprom/micro with a paper Sugden sticker attached.
A little update on this. I emailed Sugden wondering if there was updated software and received a very helpful and informative reply within the hour.
There is a mod, just a resistor to change on the servo board.
What a refreshing change these days for a company to even acknowledge a technical question let alone be so helpful. Well done...
There is a mod, just a resistor to change on the servo board.
What a refreshing change these days for a company to even acknowledge a technical question let alone be so helpful. Well done...
Hi Mooly,
I'm putting this in your thread rather than starting a new one because it's entirely in keeping with your title and highlights another possible design fault, although not really of Sugden's making.
This is a Sugden CD-21 Series 1 & 2 servo board. It looks like a 'buy out' and probably comes complete with the Philips based loader. The chipset is TDA1300T, SAA7327H and P87C58X2BBD with a pair of AM2088H for interface with motors and focus coils.
Below you will see some screen captures from the TDA1300 datasheet (my iPad will not allow me to attach a PDF) highlighting, under the heading 'laser control circuit section' the function of pin 17, the monitor diode input.
On page 1 in the General Description, they say....
"After a single initial adjustment the circuit keeps control over the laser diode current resulting in a constant light output power independent of ageing"
....which to me sounds like a really great idea.
Now have a look at the closeup of the PCB, where I have removed the TDA1300. Pin 17 has been tied to the VDD supply, completely negating, if I understand this correctly, a truly useful function.
My own experience with this loader has been deeply unhappy. Apart from laser and spindle motor issues, and the odd quirk Mooly has pointed out, these boards tend to be extremely temperamental, sometimes working for days on end, then refusing to read discs at all.
I was fortunate to have 3 CD-21s in at the same time, only one of which had a fully functional servo board. It enabled me to prove beyond any doubt that the boards themselves are half the problem with these machines.
I removed the TDA1300 because it was the most likely suspect. In 'board failure mode' the machine will initiate 3 or 4 fractional turns of the spindle motor, then give up. If no disc is present, it will go through the motions of lighting and focussing the laser. If I manually position the laser sled halfway through its travel, the processor will detect this and park the sled as it should.
So it seems to me it's not getting any feedback from the laser diode amplifiers / detectors and that does rather point to the TDA1300. I also tried switching the AM2088H devices, after all, if the spindle-motor drive is defective, that would also result in 'no disc'. The one 2088 drives the coils, the other the motors. As the focus action is good, swapping them over should kill the focus action and restore spindle drive if the SM chip is bad. Made no difference at all. Same weak fractional turns and good focus action.
Now take this board off and fit a known good one. Voila! - works perfectly. With a good board, the fractional turns are gone. The disc is spun-up and read in one go..... although if memory serves, if the machine has been powered off at the rear switch, it might need more than one attempt. But it at least manages a full rotation, as opposed to 1/16th.
Oh, and I did clean the FFC connector. Folded paper with a cleaning agent works well here. And I retensioned the contacts with a pin too. The flat cable has a nice snug fit, even before the clamp is slipped over.
Also tried heating the various chips with my hot-air station. Made no difference.
Anyone have anything to add?
I'm putting this in your thread rather than starting a new one because it's entirely in keeping with your title and highlights another possible design fault, although not really of Sugden's making.
This is a Sugden CD-21 Series 1 & 2 servo board. It looks like a 'buy out' and probably comes complete with the Philips based loader. The chipset is TDA1300T, SAA7327H and P87C58X2BBD with a pair of AM2088H for interface with motors and focus coils.
Below you will see some screen captures from the TDA1300 datasheet (my iPad will not allow me to attach a PDF) highlighting, under the heading 'laser control circuit section' the function of pin 17, the monitor diode input.
On page 1 in the General Description, they say....
"After a single initial adjustment the circuit keeps control over the laser diode current resulting in a constant light output power independent of ageing"
....which to me sounds like a really great idea.
Now have a look at the closeup of the PCB, where I have removed the TDA1300. Pin 17 has been tied to the VDD supply, completely negating, if I understand this correctly, a truly useful function.
My own experience with this loader has been deeply unhappy. Apart from laser and spindle motor issues, and the odd quirk Mooly has pointed out, these boards tend to be extremely temperamental, sometimes working for days on end, then refusing to read discs at all.
I was fortunate to have 3 CD-21s in at the same time, only one of which had a fully functional servo board. It enabled me to prove beyond any doubt that the boards themselves are half the problem with these machines.
I removed the TDA1300 because it was the most likely suspect. In 'board failure mode' the machine will initiate 3 or 4 fractional turns of the spindle motor, then give up. If no disc is present, it will go through the motions of lighting and focussing the laser. If I manually position the laser sled halfway through its travel, the processor will detect this and park the sled as it should.
So it seems to me it's not getting any feedback from the laser diode amplifiers / detectors and that does rather point to the TDA1300. I also tried switching the AM2088H devices, after all, if the spindle-motor drive is defective, that would also result in 'no disc'. The one 2088 drives the coils, the other the motors. As the focus action is good, swapping them over should kill the focus action and restore spindle drive if the SM chip is bad. Made no difference at all. Same weak fractional turns and good focus action.
Now take this board off and fit a known good one. Voila! - works perfectly. With a good board, the fractional turns are gone. The disc is spun-up and read in one go..... although if memory serves, if the machine has been powered off at the rear switch, it might need more than one attempt. But it at least manages a full rotation, as opposed to 1/16th.
Oh, and I did clean the FFC connector. Folded paper with a cleaning agent works well here. And I retensioned the contacts with a pin too. The flat cable has a nice snug fit, even before the clamp is slipped over.
Also tried heating the various chips with my hot-air station. Made no difference.
Anyone have anything to add?
Attachments
And a word of warning if you plan to rework one of these boards.
See those copper coloured patches where the IC was? Yes, they glue them on too!
This is totally unnecessary. I have made double sided boards with smd parts on both sides and we never had to glue them down yet. And yes, they were reflowed in an oven, not hand soldered.
Apart from using 20 times more glue than (un)necessary, there is an excellent chance of ripping the tracks off the board if you are not anticipating the glue - which incredibly, is totally impervious to heat and does not soften at all! I would recommend you try and soften it up first with some paint thinner or other strong solvent. Not ideal..... You could also try running a thin strong steel wire under the chip to see if it is free. Not sure if there is enough space though...
See those copper coloured patches where the IC was? Yes, they glue them on too!
This is totally unnecessary. I have made double sided boards with smd parts on both sides and we never had to glue them down yet. And yes, they were reflowed in an oven, not hand soldered.
Apart from using 20 times more glue than (un)necessary, there is an excellent chance of ripping the tracks off the board if you are not anticipating the glue - which incredibly, is totally impervious to heat and does not soften at all! I would recommend you try and soften it up first with some paint thinner or other strong solvent. Not ideal..... You could also try running a thin strong steel wire under the chip to see if it is free. Not sure if there is enough space though...
Last edited:
This is the official mod.
(I'll look at your other posts later)
Excellent! Thanks Mooly
Right....have applied the mod to one machine, so far so good. It was not reading discs before I changed the resistor (was intermittent), I specifically checked again - but now it does.
However....I will curb my boyish enthusiasm for the moment. As previously noted, these sets can work perfectly for days, then not at all.
But this is surely a step in the right direction. If it's still working this time next week I'll be a happy chappy
I will report back on this one.
Now for the next one
However....I will curb my boyish enthusiasm for the moment. As previously noted, these sets can work perfectly for days, then not at all.
But this is surely a step in the right direction. If it's still working this time next week I'll be a happy chappy
I will report back on this one.
Now for the next one
Ok, this is the board I removed the TDA1300 from. Here it is put back on again. Sheesh...30 min to get the damn thing off because of the glue and 45 sec to put it back again. That poor little chip was abused quite badly but it seems to have survived. So the mod has been done and it too plays
But will it play tomorrow? Or next week?
Will report back on this one too
But will it play tomorrow? Or next week?
Will report back on this one too
Attachments
good to know its working OKThe LPC (laser power control) feature is standard on all CD/DVD/SACD players and relies on a photodiode mounted on the same die as the laser diode itself. So the photodiode receives a true and direct representation of the emitted light.
The circuit itself is just a simple feedback arrangement around an opamp with the circuit altering the laser diode current to maintain the light level at some preset value... as set by the trimmer pot. Its absolutely necessary due to the characteristic of the diode which is extremely temperature sensitive. A constant current drive would give a light output that varied enormously over the operating range of the player.
It might just be worth replacing the transistor that drives the laser diode... never had one fail on these but they can fail as a generalisation. Not often, I've probably had two or three over the years across various players.
When I first worked on this set I noticed some of the electrolytics had some 'road rash'. It took a while to connect the dots but eventually I realised this board had been on a belt sander! And when I checked, all the others also. Can this be good for a board containing such sensitive (and relatively high density) ICs?
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.