Sony WM-D6 Motor runs fast; Repair of CP602 Servo Circuit Unit.

Discussion in 'Tech talk' started by Deb64, Oct 29, 2017.

  1. Deb64

    Deb64 Active Member S2G Supporter

    Messages:
    236
    Trophy Points:
    43
    Location:
    Leicestershire, England
    These notes are intended as a guide to the repair of a common fault with the CP602 Servo Control Unit on the Sony WM-D6 Professional Walkman.
    Note that this repair only applies to the WM-D6; there is a different motor drive circuit on the WM-D6C.

    Common symptoms of CP602 failure are:
    • When play is pressed, the motor runs too fast.
    • The motor will sometimes creep or run very slowly when there are batteries in the walkman and it is switched off.
    This may start as an intermittent fault, with the walkman playing normally and occasionally speeding up. It will eventually get worse, and play fast all the time.

    Before attempting to repair CP602, check for other possible causes of the motor running too fast.
    • Ensure that the 11V DC supply is present and is getting where it is needed. Check that there are no broken tracks on the 11V rail. If there is no 11V supply, check the DC – DC convertor. Mihokm has provided a comprehensive guide to repairing the DC – DC convertor here:
    WM-D6C DC-DC converter fixed
    • Check for dry solder joints on the main PCB, especially in the area around IC601 and CP601. The WM-D6 was made between 1982 and 1984. These units, which are now between 33 and 35 years old, are prone to dry joints and cracked PCB tracks.
    • Check Q603. There is still 6V on the emitter of Q603 even when the walkman is in stop mode

    The most common problem I have found on the CP602 is failure of the NPN transistor, which drives Q603 and / or the FET. I recommend replacing both of these.
    Firstly, you need to make sure that the CX20024 IC on the CP602 is still working. If the CX20024 has failed, then the CP602 cannot be repaired as the CX20024 is no longer available.

    Before removing the CP602 from the walkman, check the following:
    • Is there 2.2V on pin 12 (of the CP602)?
    • Check for a sawtooth waveform at pin 8 of the CP602
    If these are present, then the 20024 IC on CP602 is probably working.

    Before attempting to remove the CP602 module, take plenty of photographs of the PCB so that you will be able to replace the wires and capacitor when re-assembling. Take photos while you’re working too. With digital cameras and phones, photos cost nothing and save hours of head scratching “Where does that wire go?” moments when re-assembling.

    Before desoldering.JPG


    To remove the black potting, place the CP602 in acetone; the potting will soften and come away from the circuit after a day or two. I do not recommend using heat to remove the potting as the amount of heat required to soften the potting will also melt the solder and may damage components on the board.

    I would recommend treating the CP602 as a static sensitive device. When not working on it or soaking it in acetone, store it in an anti-static bag and protect it against static discharges.

    The CP602 board is made from ceramic and the resistors are printed on. They are the darker rectangular shapes on the board.

    NPN and FET.JPG

    CP602 Schematic.JPG

    The FET and transistor will have been glued to the PCB before soldering and will need to be removed carefully using a soldering iron and scalpel blade tip to lift the component while melting the solder on the pins.

    The replacement components I have used are:
    FET: 2SK209
    NPN: 2SC1623-L6

    Both of these are still available. I bought the 2SK209-Y FETs from Digikey and the 2SC1623 NPN transistors from a seller on Ebay.

    I have scratched away the green solder resist between pins 5, 6 and 7. I am wondering if the solder resist is starting to become conductive on these early boards, possibly due to a reaction with the potting compound. I've had several old mute modules fail and I am wondering if the solder resist could be part of the problem. Perhaps somebody who knows a lot more than I do about chemistry could tell me if this is possible.

    Repaired CP602.JPG



    Notes on Soldering SMD components.
    If you're not used to soldering small SOT23 transistors, they are a bit fiddly to work on but you need the right tools, good lighting and magnification and plenty of practice. I find headtorch good for illuminating items I am working on. I would suggest practicing on a scrap board or prototype board before working on your walkman.

    I prefer to solder SMDs using an iron and solder rather than paste and hot air.
    I use a Weller TCP Magnastat iron fitted with a PT S8 0.4mm round tip 425 deg C tip. I also use Fry Powerflow flux and 60/40 tin/lead 22swg thin solder. I also have a pair of high quality (Xcelite) fine point SMD tweezers and a cocktail stick.

    Firstly, having removed the old component, I clean the pads using flux and 1mm solder wick. I then apply a small bit of flux and solder onto one pad and place the component in place by holding it with the tweezers and soldering the pin to the prepared pad. I then apply a small amount of flux, using the point of a cocktail stick, to the remaining pins / pads and then solder each one. You don't need a lot of flux.
     
    Boodokhan and Jorge like this.
  2. Longman

    Longman Well-Known Member S2G Supporter

    Messages:
    3,816
    Trophy Points:
    113
    Location:
    Bournemouth UK
    Great post Deb

    Thie made me think of a common problem with the Roland Juno synthesizers which are a similar age and use similar potting on their DCOs. Many people have fixed them simply by removing the potting from the modules.

     
    Jorge likes this.
  3. Deb64

    Deb64 Active Member S2G Supporter

    Messages:
    236
    Trophy Points:
    43
    Location:
    Leicestershire, England
    It's interesting that other equipment from the same era suffers from similar problems.
    I concour with the comments in the video about not using a heat gun to remove the potting. I did that with my first CP602 (you can see some melted solder in one of the photographs) and I think there is a great risk of damaging the components with the amount of heat which you need to soften the potting.
    I don't think removing the potting alone would solve the problem with the CP602 modules. The ones I have worked on have either the FET or the NPN failed, so I just replace them both now.
    Interestingly there are 5 potted hybrid modules on the D6 (2 x equaliser switching modules, a FG amp unit, the CP602 Servo Circuit unit and the CP301 muting module) but only the CP602 and the CP301 muting module are consistantly unreliable. So far I have not had any trouble with the FG amp unit or any equaliser switching modules.
    In the case of the muting modules, I tried replacing all the transistors on a module once but I still wasn't happy with the sound, which still sounded a bit distorted. This is why I'm a bit suspicious of the solder resist.
     
    Boodokhan and Jorge like this.
  4. Jorge

    Jorge Well-Known Member

    Messages:
    1,754
    Trophy Points:
    113
    Location:
    Malibu, CA
    @walkman archive and Administrators - could you PLEASE make these posts by @Deb64 'sticky'?
    I was trying to find this thread yesterday but failed
     
  5. Valentin

    Valentin Well-Known Member

    Messages:
    1,549
    Trophy Points:
    113
    Location:
    Europe
    I also fixed the CP101 and CP201 from a SONY WM-7 (Dolby B circuit) simply by removing the potting compound thoroughly. http://stereo2go.com/forums/threads/sony-wm-7-no-dolby-nr.6036/

    I measured the voltage at the output of the DC-DC converter when the modules where bad, and it was fluctuating (dropping and coming back up), indicating that maybe the potting compound has became conductive, or conductive when a voltage is applied to it. After soaking them in acetone, I found that the potting compound was very soft around a couple of the transistors (and maybe that part became conductive). The fact is, they work perfectly after depotting them.

    When I did the WM-7 repair, I also found 2 EQ modules from my WM D6C and decided to remove the potting on them as well and measure the transistors and resistors after that. Everything measured ok. Before replacing them, type I tapes sounded dull (like the selector was in Type II position) intermittently.
     
    Last edited: May 6, 2021
    Jorge likes this.
  6. retrocode.kr

    retrocode.kr New Member

    Messages:
    5
    Trophy Points:
    3
    Location:
    Seoul, SOUTH KOREA
    Hello
    I'm a Newbie who's been fascinated by cassettes.

    When I found this article, I felt like I found an oasis in the desert. Because I have a exactly-same-symptom WM-D6.
    I decided to give it a try right away. I replaced the FET, the transistor on the ceramic board and the tantalum capacitor C607.

    But it didn't work.....
    After a close examination, I found that some parts were cracked.
    I'm not sure if it was due to my poor skills when removing potting or because it was soaked in acetone for too long.
    So, if possible, the best thing is to change all the suspicious parts.

    The problem is, as for a newbie like me, there is no way to figure out the values of all the capacitors on the board.
    Would please let me know the capacitor values?


    1719799505954.jpg D6 CP602.jpg
     
  7. Deb64

    Deb64 Active Member S2G Supporter

    Messages:
    236
    Trophy Points:
    43
    Location:
    Leicestershire, England
    From my measurements I believe the values of the capacitors shown are:
    1 - 2n2
    2 - 180pF
    3 - 2n2
    4 - 10nF
    5 - 10nF
    6 - 3n3

    The missing diode was originally a M1MA152AT or a M1MA151AT. These are a single diode package, which is no longer available. Instead I have used the M1MA142WAT1 which is a two diode package. One diode is not used.

    I notice that there is quite a lot of black marks between the pins which connect onto the main board. This could be the remains of the potting or it could be dendrite growth. I don't know much about dendrite growth on circuit boards but there are some articles online about it.
    I have attached some photographs which I took with a microscope. There is an overall picture at x20 magnification and some close ups at x400 magnification.
     

    Attached Files:

    retrocode.kr, Jorge and Valentin like this.
  8. Deb64

    Deb64 Active Member S2G Supporter

    Messages:
    236
    Trophy Points:
    43
    Location:
    Leicestershire, England
    It is also worth removing the remain of the potting from under the pins of the CX20024. You may need to put the CP602 back into some acetone to soften the potting before attempting this. When it is soft you can carefully remove the potting with fine tweezers or a fine wire. I find that doing this seems to help improve the wow and flutter.

    Finally, check all the solder joints of the components on the CP602. I have come across a couple of CP602s which have very poor factory solder joints. It is surprising that they ever worked at all.
     

    Attached Files:

    retrocode.kr, Jorge and Valentin like this.
  9. retrocode.kr

    retrocode.kr New Member

    Messages:
    5
    Trophy Points:
    3
    Location:
    Seoul, SOUTH KOREA
    @Deb64
    Thank you so much for the detailed replay and explanatory photos. I'm really appreciate it.
    I'll order the parts ASAP and put them into action carefully as you explained.
    I look forward to seeing good result and will post a detailed review as soon as the result come out.
     
    Jorge and Cassette2go like this.
  10. Jorge

    Jorge Well-Known Member

    Messages:
    1,754
    Trophy Points:
    113
    Location:
    Malibu, CA
    I remember having a real hard time soldering to those lands:
    L1100840sm.jpg

    ended up with a 'bungee', to be on the safe side:
    L1100883sm.JPG


    I've learned my lesson: Never Again! My next D6 goes straight to @Deb64, whatever the shipping costs'n'dangers..., I do not even want to know what those horrible traces and printed resistors are :confused:
     
  11. retrocode.kr

    retrocode.kr New Member

    Messages:
    5
    Trophy Points:
    3
    Location:
    Seoul, SOUTH KOREA
    @Jorge

    Yes, you're right. It would be great difficulty to solder those tiny components.
    I hope it would be interesting challege worth a try.
    Wish me a good luck. Thanks! :nwink:
     
    Jorge likes this.
  12. Deb64

    Deb64 Active Member S2G Supporter

    Messages:
    236
    Trophy Points:
    43
    Location:
    Leicestershire, England
    Jorge likes this.
  13. retrocode.kr

    retrocode.kr New Member

    Messages:
    5
    Trophy Points:
    3
    Location:
    Seoul, SOUTH KOREA
    Well, to make it short, I didn't succeed in conclusion.
    I've replaced capacitors and diode, but the motor still rotates fast.
    I'm sorry that you guys helped me a lot, but I didn't get good result.
     
  14. Chang Song

    Chang Song New Member

    Messages:
    13
    Trophy Points:
    3
    Location:
    North CA, US
    It might be simple issue like white wire detached from the post around C610 - I recently fixed a CP602 but it ran still fast. I had hard time debugging and it turned out that white cable had the plastic sleeve hanging on to the post but internally was disconnected.
     
    Last edited: Oct 9, 2024
  15. Chang Song

    Chang Song New Member

    Messages:
    13
    Trophy Points:
    3
    Location:
    North CA, US
    CP602_wire_diagram.png
    I recently had to debug on the CP602 work (and reconstruct pin 2 using external 47kOhm resistor due to loss of plate on the board) and made this wire diagram. Hope it is useful for anyone who needs to reconstruct CP602.

    Note 1: Resistors are embedded on the board
    Note 2: Service manual has CX20024 pin wrong around pin 11 and pin 12 connections
    -- CP602 pin 11 -> CX20024 pin 8 (not 9)
    -- CP602 pin 12 -> CX20024 pin 9 (not 7)
     
    Valentin and Deb64 like this.

Share This Page