Announcement

Collapse
No announcement yet.

Microkorg CPU trouble

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Microkorg CPU trouble

    I have 3 of these units in, each with a blown power supply which is common. The fix is discussed in this thread: http://music-electronics-forum.com/t15415/
    Schematic is attached there as well. In that thread, the CPU is identified as a common failure as well as the power supply components.

    On the first unit, I've restored the power supply by replacing IC5, C31 and FU1, after which it booted up and ran. Everything works except for the 5 control knobs and the pitch bend and modulation wheels. Could this be a bad CPU? I would expect it not to boot or run at all if that were the case, but I can't find any other potential fault.

    The knobs and wheels feed a CV directly to the CPU's built in D/A section. I have verified these voltages are present at the correct pins on the CPU, also the relevant ground and reference voltages for the CPU analog section. So it seems there is a fault in this section of the CPU that allows it to otherwise run but not respond to changes in the analog inputs.

    Is there anything I could be missing here?

    Finally, I need a replacement part if the CPU is shot. The original is a Rensas H8S/2320, which I cannot source. Digikey has a similar parts H8S/2321, 2322 and 2390. The 2322 seems identical except for 8K internal RAM rather than 4K. Is there any chance this would work as a replacement? If not, where can I find the original?

    Thanks for any help
    Jon

  • #2
    Possible need to reinit the thing?


    Have you run the diagnostic programs?
    Education is what you're left with after you have forgotten what you have learned.

    Comment


    • #3
      I did the test mode (write+3) and followed the walk through you posted in the earlier thread and the knobs wouldn't respond there either. Also tried the shift+7 reset and wheel calibration, also no dice. Is there another reset I'm missing?

      Got the H8S/3022 today, but I'll have to cut the legs off the old one so I really want to exhaust the possibilities first.

      Comment


      • #4
        Cut the legs off? Look up Chip Quik. Works great

        Chip Quik: Easy removal of surface mounted devices



        No, looks like you have done the tests. Only thing I see all the analogs have in common is the reference voltages and the IC.
        Education is what you're left with after you have forgotten what you have learned.

        Comment


        • #5
          Once repaired though, will the circuit have the same tendency to fail as it did before, or are the recommended replacement components upgrades or different in some way to address the problem? With a on-off failure you can generally attribute it to 'just something that happens' but with multiple failures of the same type it points to a design fault or under-specified component.

          Comment


          • #6
            I've fixed a couple of Microkorgs, and, based on that experience, I regard them as very cool toys. I think they were designed from the get-go as disposable electronics, not as musical instruments designed to last indefinitely, and it's hard to overcome the limitations of that design philosophy.

            Comment


            • #7
              So it was the CPU. I tried the H8S/2322 from Digikey and it didn't work. I then ordered the H8S/2320 part #HD6412320VF25 from UTSource which is always a crap shoot as to whether the part works or not. It did work. However not perfectly, and that may be due to my imperfect SMD soldering.

              The original problems of no analog function were solved, however it had 2 weird flaws: The middle LED segment was missing on all 2's and 3's on the display, the resonance knob didn't work and the tempo knob controlled tempo and resonance at the same time. I guess I could call it circuit bent.

              After some cleanup and reflowing of the solder, the LED problem was gone but 2 other new problems appeared - the 6 button stopped working (1-7 and 8 were fine) and it would no longer respond to the write+1 test mode or any other bootup sequence, it would only boot normally.

              On further investigation, a couple of the data lines (T1 to T10 on the schematic) were shorted to ground within the CPU. So I'm going on the assumption that these chips are good and I screwed it up. I'm getting more chips and trying again.

              Any sane person would not do this, but I have 2 more in for the same problem already. If I can get one right, maybe the rest will be a breeze.

              Comment


              • #8
                Or pay someone local who is properly equipped to install the ICs on your boards.
                Education is what you're left with after you have forgotten what you have learned.

                Comment


                • #9
                  If you're having problems with soldering the ICs you may want to use some additional flux to improve wetting and break the oxide bag that forms over the molten solder that gives it a tendency to bridge. I use solder specific to SMD rework - it has a lower melting point and contains silver which both reduces leeching and improves flow.

                  Comment


                  • #10
                    Just a thought,
                    Are you certain the Processor you got from UT Source has a program in it? Many of these processors are generic and if not sourced from the manufacturer of the unit they're installed in, come unprogrammed. BTW a few years ago you could still source this processor from Roland. Perhaps you could connect with someone who is a dealer & can procure it for you...

                    ...good luck...glen

                    Comment

                    Working...
                    X