Ad Widget

Collapse

Announcement

Collapse
No announcement yet.

Ensoniq SQ-80 losing polyphony voices

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

  • Ensoniq SQ-80 losing polyphony voices

    Quick question to Enzo, or whomever has a thought...

    Just beginning a new CD project (Genre: Adult Alternative / Folk-Rock), and for the first time, I find my trusty old SQ-80 is not running on all 8 cylinders (i.e., polyphony voices). So... Two questions:

    1) If the polyphony is beginning to break down, what do I need to replace? (Note: this is NOT an "over-reaching the available voices" issue. This is truly a "I no longer have all 8 voices" issue)

    2) Where might I find a procedure for this repair? (I have local techs who are very capable, but I'd like to hand them as much info as possible, when I drop the SQ off to them.)

    Much thanks!

    Lumpy

  • #2
    Not too sure what you are asking.

    The 8 'Voices' are actually 8 'Tracks'.

    So, are you having issues saving tracks or calling them up?

    http://www.buchty.net/~buchty/ensoniq/manuals/SQ80.pdf

    Comment


    • #3
      Polyphony, not 'tracks'

      Originally posted by Jazz P Bass View Post
      Not too sure what you are asking.

      The 8 'Voices' are actually 8 'Tracks'.

      So, are you having issues saving tracks or calling them up?

      http://www.buchty.net/~buchty/ensoniq/manuals/SQ80.pdf
      Hi, JPB:

      Thanks for your response. Let me try again. The SQ-80 has "8-voice polyphony". Which means it's capable (when it's operating correctly) of creating 8 discrete sounds at one time. Some of the patches use one voice per note (e.g. piano). Others "LA Synth" on a voice crystal, uses 2 of the unit's polyphonic voices to create a single note. this means if you try to play any more than 4 notes (at one time) using that patch, the extra notes won't play, at that instant.

      I just noticed that pre-recorded sequences that never had voices dropping out (because I use the patches carefully to avoid that), suddenly are. I verified this by playing some chords in real time using the piano-string patch, and sure enough - I had notes in the chord that wouldn't play with the rest of the chordal notes.

      What this means is that I've lost either one or two of the eight polyphonic voices - which tells me one of the electronic components, (possibly on the motherboard) has failed. But I have no clue how to troubleshoot the system to determine what part of the hardware has gone bad.

      Hopefully, that helps.

      Lumpy

      Comment


      • #4
        Crap, I thought my Ensoniq days were behind me. Even as one of their service centers, that company was a constant fight.

        My memory is dim, isn't the SQ-80 more or less an ESQ1 that went to college? Put a different way, this is one you can edit patches and store? AS opposed to the ones with all the stuff on floppies? My first reaction when facing this would be to reinitialize. That will of course erase any custom patch editing you have, unless that is backed up.

        As to your local techs, if they are competent, they don't need us telling them where to start. They may have questions we can answer, but when you take your car in for repair, you don't ask uncle Joe the retired mechanic what you can suggest to the service shop, do you? If they work on synths for s living, then most of what I am about to say will not be new to them.

        The company never issued schematics, even to service centers. Some of us were able, just from skill and experience, to do component level repairs - diagnosing a bad Curtis chip for example. Only years later they offered us a set of schematics for out of production models... for $100. Don't recall if it went as far as SQ80 or not, and I am not at the shop today. But if I recall, there are eight voice circuits based on Curtis chips.

        So, put it on a simple patch you can hear well. play into an 8 note chord. Do all eight notes play? Or do you consistently get only 7? If one is missing, is it consistently say the third one, or does it change? If you can ever get all eight notes, then all eight voice circuits work, and the problem would be in the key assigning or routing.

        If one voice is bad, one way to find which is to get a data sheet on the Curtis chip and determine whether each IC is getting triggers and is generating outputs. No I do not recall wwhich Curtis chip each model used. Curtis made a number of things - VCA, VCF, etc.

        I THINK the Curtis chips were socketed, wwhich means you could take turns removing just one and seeing if the symptom changed. If removing a chip kills a note, then it was probably good. If pulling a chip made no difference, chances are it is bad.

        I still have an ESQ1 board mostly complete, which should have most of the same chips.
        Education is what you're left with after you have forgotten what you have learned.

        Comment


        • #5
          Here is link to the SQ-80 schematic.

          Index of ./ensoniq/sq80/

          Comment


          • #6
            Looks close enough to an ESQ1. I'd do the experimants I described. The analog pages in the link include the Curtis chips and the SSMs.
            Education is what you're left with after you have forgotten what you have learned.

            Comment


            • #7
              Enzo - Thanks! Initializing the system may not work, but it's a great start

              Originally posted by Enzo View Post
              Crap, I thought my Ensoniq days were behind me. Even as one of their service centers, that company was a constant fight.

              My memory is dim, isn't the SQ-80 more or less an ESQ1 that went to college? Put a different way, this is one you can edit patches and store? AS opposed to the ones with all the stuff on floppies? My first reaction when facing this would be to reinitialize. That will of course erase any custom patch editing you have, unless that is backed up.

              As to your local techs, if they are competent, they don't need us telling them where to start. They may have questions we can answer, but when you take your car in for repair, you don't ask uncle Joe the retired mechanic what you can suggest to the service shop, do you? If they work on synths for s living, then most of what I am about to say will not be new to them.

              The company never issued schematics, even to service centers. Some of us were able, just from skill and experience, to do component level repairs - diagnosing a bad Curtis chip for example. Only years later they offered us a set of schematics for out of production models... for $100. Don't recall if it went as far as SQ80 or not, and I am not at the shop today. But if I recall, there are eight voice circuits based on Curtis chips.

              So, put it on a simple patch you can hear well. play into an 8 note chord. Do all eight notes play? Or do you consistently get only 7? If one is missing, is it consistently say the third one, or does it change? If you can ever get all eight notes, then all eight voice circuits work, and the problem would be in the key assigning or routing.

              If one voice is bad, one way to find which is to get a data sheet on the Curtis chip and determine whether each IC is getting triggers and is generating outputs. No I do not recall wwhich Curtis chip each model used. Curtis made a number of things - VCA, VCF, etc.

              I THINK the Curtis chips were socketed, wwhich means you could take turns removing just one and seeing if the symptom changed. If removing a chip kills a note, then it was probably good. If pulling a chip made no difference, chances are it is bad.

              I still have an ESQ1 board mostly complete, which should have most of the same chips.

              Enzo: I'll try initialization first, and go from there. The system battery I had replaced in 2007, so I don't think that's at issue.

              Meanwhile, you can hear some some samples of Lumpy's work at: Lachlan MacLearn - www.lachlanmaclearn.com

              Comment


              • #8
                The battery can be perfectly healthy and the memory can still be corrupted. There are other ways this can happen. Static discharge comes to mind, and even soft errors. Cosmic rays pass through our bodies and our gear all the time. Through the earth itself for that matter. They often do so without actually doing anything, but if one happens to strike an atom in a RAM IC, it can change the state of a bit, which of course the computer accepts as data. No harm done other than the data. Thus a soft error. Reinit cures that.

                When I was an Ensoniq repair center, back when ESQ1 was a very popular board, reinitializing was a common repair. Guys would bring them in with a complaint like all the patches now had modulation on them. Or the pitch tuning had shifted on all the patches.
                Education is what you're left with after you have forgotten what you have learned.

                Comment


                • #9
                  Originally posted by Enzo View Post
                  Looks close enough to an ESQ1. I'd do the experimants I described. The analog pages in the link include the Curtis chips and the SSMs.
                  Which ic's are the infamous "Curtis Chips"?

                  Comment


                  • #10
                    Sorry, shop lingo.

                    Curtis Electro Music. The chips are CEM something on the schematic. Find the SSM chips in the schematic and look to their right for CEM something.

                    I have some paper data sheets on them somewhere if we can't find it online. Or were the pins identified on the drawing? (I guess I could look)
                    Education is what you're left with after you have forgotten what you have learned.

                    Comment


                    • #11
                      Apparently, Mr Curtis was a busy guy.

                      http://www.synthtech.com/cem/cemdata.html

                      CEM (Curtis) SSM Hardware Chips used in these Synthesizers - welcher dsp wo?

                      Comment

                      Working...
                      X