Dual CD 3590RC player mandness [SOLVED]

[shakalnokturn]

More for the records than seeking help, if I did skip something obvious let me know...

This Dual CD player I'd had a first look at a while ago and left aside, can't remember previous conclusions.

Dug it out 3 days ago and have since been spending my evenings with it - running circles around my sanity.

With belts replaced, switch contacts and lens cleaned, gave it a disk that it wouldn't read just displaying "0".
The pickup wouldn't return to index position on loading. Bringing the pickup against the switch, loading disk then moving it by hand just off the switch it would read index and play any CD normally until stopped.

Track seek and skip also worked in both directions.

At this point I set out for a schematic but had no luck so I started looking for other CD players using the same YM3815 DSP/control IC with a service manual available. There were some from JVC, Onkyo, Fischer... The JVC ones were the easiest to read.
(After searching quite a bit I think it is very similar to the Lenco HCD-390 or Yoko F-420R players. Hardly more common, just as much service information...)


With approximate schematics (general layout of the sled servo is similar but component values change) and some probing I had a better view of what was supposed to happen.

The YM3815 only directly controls pins FEMOF(24), FEM+(25), FEM-(26) when load, seek or skip operations are required, during normal play the tracking servo takes care of shifting the feed (sled) motor before the tracking coil on the pickup can't keep up.

During skip forward operations FEM+ will be pulsed high while FEM- is held low, for backward track skips FEM- is pulsed high while FEM+ is held low. FEMOF remains low so the tracking servo can lock on the newly selected position.
This works well once the player has been helped to read the index which would mean that the feed motor control is working as intended.

On loading a disk FEMOF is held high which disables the coarse tracking servo, FEM+ is held low, FEM- held high which should move the pickup to centre position until it hits the switch but it didn't...

There was some voltage on the feed motor (-1.4V after the drive transistor's Vbe drop) but just not enough to get it moving.
Running short of Ideas I tested the motor on a lab PSU, and did find that although it was working there was a mechanical issue with it, something loose like a chip of magnet inside would jam it at times.

I replaced the motor with the exact same model, still no return to centre...

FEMOF transistor was replaced with no effect, to be noted that on all JVC's schematics the transistor is NPN with Collector to ground while in my player it's NPN with Emitter to ground, I think JVC have things wrong on the schematic.

In a last attempt before finally giving-up and sending it to the parts heap I modified the motor control amplifier's gain (replacing 120k feedback resistor by 220k), and after readjusting the tracking trimmers have a working player.

Now why the f*** would it not work? I just can't get my head around it, the op-amp output voltages were consistent with the theoretical values...

[fzabkar]

Dried grease?

Did you measure the voltages at pins 2 & 3 and then calculate the expected output?

What are the high and low levels of the FEM signals? Are they 5V logic?

[shakalnokturn]

No, the cogs and sliding guides were really smooth, only the original motor was a little erratic.

Yes, calculated value was very close to measured value...

Yes, FEM +/- are 5V logic levels with a good swing, within 0.15V of rails.