jonathanpotato

The OWH needs to be UNDER the metal plate. See pictures of the post : Metal slats needs to be straightened. Do the "virtual MD" trick (no MD inside) Put the OPU/OWH to the center (>> button) Unscrew a little (1 or 2mm) to release the OWH (red) Pass gently the OWH under the metal plate Re-screw the OWH on OPU Use scalpel (or fine tool) to twist big metal slats (yellow) to re-adjust the OWH height Use scalpel to twist the little metal slats to make the head flat against to the disc (orange). HINT : Put a minidisc and make STOP-REC-STOP-REC... to up and down the head and untwist theses wires step by step. Once the head is flat against the disc when the head is down in recording mode like this : Then try record few seconds and play what you recorded. If you break tiny metal wires (orange) = get a new OWH, you failed.

There is a worm screw yes, but in his second video at 40 seconds you can see that his deck is not trying to move the OPU at all before ejection. It ejects directly. As ck10 said he was able to move the OPU in service mode. So nothing is blocking mechanically and no motor problem too. Diagrams are in the SM of minidisc.org : https://www.minidisc.org/part_Sony_MDS-JE510+S38.html

If you don't want to disassemble more, the only solution is to try unlocking it or cleaning it (with contact spray) by top like I explain here : https://video.latavernedejohnjohn.fr/videos/watch/44346503-e4e8-427e-bece-0b6750bf7336 But with the risk to ruin the OPU which do not like liquids inside obviously if you don't protect it and do a bad job.

In fact in Sony bulletin about this clearly says that ALL SWITCHES OF THE MECHANISM are concerned and to be replaced (normally). 99% of time, we (myself too) resolve the CHUCKING IN one and that's it, problem solved. But others can be faulty over time. Replace this one is difficult for a novice. I only changed theses when someone do sh** job before me. See pictures : https://www.jonathandupre.fr/articles/17-minidisc-md/233-reparation-platine-md-sony-mds-je510-massacree

The plastic sheet is only a part of the Sony fix. The second part is to remove some metal to allow more space to the OWH. It's not your case but I explain it for visitors. But on some versions (picture you posted before) you can see the fix is not present (no dent in metal) : Thats why it's written :

OWH = Over Write Head = Fragile metal/plastic piece in white you are destroying progressively... :-) So you have to understand why your Optical Pickup Unit (with the OWH screwed above) is not returning correctly to the center. Is a little piece/screw is blocking ? Maybe missing some grease on worm ? Week motor ? Low tension (volts) send to motor by electronics ? I don't know. MDM-3 is prone to grab/twitst the OWH. The symptom is no recording and erase disc because OWH is not plane to the disc anymore (twisted). And you need tact to make it come back to his original shape flat against the upper side of the disc. It's the first time I see the MDM-3 not returning correctly the OPU. As I said to Kevin before. If the software allows you to eject, it means that the signal comes to the CPU/µc that the OPU is parked is received, but you can see physically that it's not the case (by hurting the OWH plastic). That's why I supposed a bad/faulty LIMIT IN switch.

I corrected my word "leveller" -> Lever (as Sony spells it in SM). Yeah I think his OHW will die soon if he continues. :-) It's a miracle it still alive. I persist, the problem for me not is loading or up/down gears but the OPU is not fully pushed to the left before ejecting -> sled problem. That's why I put colors to better understand : BAD : GOOD :

Yes Kevin, the service mode can be used to handle manually the sled part of the mechanism. Like almost all other models << and >> buttons can be used on front panel to move the optical block left or right. Just be careful with ends. IMHO if his deck allow ejecting in this position, there is a problem with the LIMIT IN switch. That's explains also that his deck was late to start the disc until 12 seconds in his video : too long time.

You can safely remove this metal plate (black screw) for the moment for your tests.

Did you put the lever correctly ? It seems that your optical pickup does not fully return to the limit park (center) before eject. The gap in normal use :

Put this OWH in the right way first. Poor recording head.

Not strange, normal, there is a MDM-2 inside this unit guys. So laser power is manually adjusted with a potentiometer (RV) on the BD board, no chance to adjust it with the service menu. If the lens is already cleaned (first basic thing to do), simply make this test : Power off/on the unit. Put your JA3ES on the side. Insert a well-working recorded MD and play it. If the readability is clearly better (quick track access, no OPU bizarre noises, no dropouts on music), great you win a worn-out KMS-210A OPU classic symptom. Search the web. That will not tell you if the disc a well played in back-end. Go into service mode and do a CPLAY mode test (IN, MID and OUT) with a full well-working recorded MD, it will give you live C1 error value between 0 and 7350. Observe values for 10 seconds in the 3 zones/areas (INSIDE, MIDDLE and OUTSIDE) and made an average. Now power off/on the unit and do the same thing with your JA3ES on the side. Same, observe values for 10 seconds and made an average. With the comparison, deduce.

And cleaning the lens manually ?

I repaired this deck for someone. The CXA2523AR was replaced because unable to switch between read and write laser power. Always stuck at 4.5 mW on my LPM (APC pin shorted to ground) The BH6511FS was replaced because partially broken : focus/sled/tracking OK but the spindle motor does not turn, PWM commands from CXD DSP arrived to it but no power given to motor. Motor was tested OK with an external PSU.

Read will never fail if you increase the hex value only by 3 in service mode as I said before. It begins to be a problem for circuits (RF amp) much further in hex values. RF signal will be higher but very far from the saturation level of the RF amp input. Anyway, It's not a good solution to solve the real cause of a bad disc reading but it's clearly safe if it can help someone to back an unit to life if doesn't have a LPM, in my opinion, but like said before it's dirty. You are not have to be agree with me.

It's critical in the way that if the read power is too low the disc will not be detected / TOC not read well / music dropouts and bad mecha noises. But you know if that the read power is sufficient already when you but the disc in at start : if the TOC is read when you put the disc in, why it cound't read again at the and of a recording process ? Anyway, TOC is not read at the end, TOC is not "modified" on disc, old TOC is just fully erased and replaced (overwritten) by the complete modified one from the RAM of the deck, no need to re-read it. In case of bad read power laser you can get "Retry" or "Retry Error" messages during recording.

IOP value is an indication about laser diode degradation. If you want to calibrate the laser a LPM is needed. IOP value doesn't give a real idea of the light emitted laser power. I have the chance to have one. But I think you can't do a good job without. It's the relation between the laser diode consumption in mA given by the IOP measurement + the laser light measured by the LPM in mW that said if the laser uses the right amount of energy to produce the right amount of emitted light energy on the other side. If there is derivation between the two (more than the 10% tolerance indicated in Sony SM's), is the fact that the laser diode begins to fail. Adjusting record power needs more accuracy than adjusting read power that it can be adjusted roughly (for thoses who doesn't had a LPM) in service mode. Changing LDPWD for reading (0.7-0.9mW) slightly (increase or decrease by 3 hexa values) can be done by users safely without LPM if it can solve some reading problems, It's dirty but safe for discs and device because the jump was tiny : 0.1mW and the laser is in low-power mode. Changing LDPWD for writing (7mW) absolutely needs a LPM because changing the value directly give a jump of + or - 0.3mW on the LPM for each hexa value, and as it run in high-power mode, it's unsafe because it can make the laser works higher that the IOP rated value written on the label, and more near of the limit that the diode can support = life duration shorted and overheated recorded disc more than the 180°C Currie point needed.

Seems to be that something had been misunderstood. Laser calibration values are stored in BD Board yes. But the IOP value is not a calibration value. The IOP value entered manually in the service mode doesn't help the deck to configure the new installed laser. It doesn't care at all. It's just a comfort feature to get the actual IOP value of the installed OPU without removing the whole mechanism to read the tiny label on the OPU itself when comparing what you have on multimeter (connected IOP output pin on the 6P service connector). It's just a value (stored in NVRAM/EEPROM) that you can read and write (when you put a laser in place), but the value itself was not used at all by BD board and circuits to configure/manage the laser. Laser calibration values are stored in BD Board yes, so - If you swap both matched BD-Board + OPU, you have nothing to do. - If you swap one part of the two, you have to redo all the adjustments to apair new BD-Board + OPU couple. Not just only the IOP. It will works if adjustements are not "re-done"; but both components are just working "out of specs" and some problem can occurs about reading and writing discs like recorded discs can be badly readed on other devices because badly recorded. Hope it was more clear.

Yes, 260A and 260B are fully interchangeablee. Even 260E can be interchanged with them but needs to be configurated (calibrated) differently about the laser power in service mode. KMS-xxxx had nothing to do with MDLP or not, the only fonction of the OPU is to read the disk and erase the disk by hot burning in rec mode. MDLP depends of DSP features (CXDxxxx capability) and firmware (= SYSCON) too of course.

Plenty of MDS-S38, S39, B5. But I'm not in England.

If it's a digital multimeter : put your multimeter in diode mode function. Make touching the two probes to have contact and you should here a bip or buzzer and read 0 on LCD display when contact. Put probes on both side of the ribbon cable and test each pin to check if continuity between each side.

Check cables first, it does not cost anything. What is your version in service mode ?

Maybe bad mechanical switches or bad white flat wire cable somewhere too.

Very strange. It's like if the eject button is interpreted by the software like a REC button. It can arrive if the switch is bad because it can make some electrical glitches on signal. Or a resistor. But as you bought the unit from eBay you didn't know anything about the past of this device ? Possible that some parts like microprocessors had been mixed with a B5 ?