Help Modding My Polyphonic Synthesiser

Discussion in 'The Projects Forum' started by apprenticemart2, May 15, 2013.

  1. apprenticemart2

    Thread Starter Member

    May 7, 2013
    67
    7
    Help Modding My Polyphonic Synthesiser with a 4051 multiplexer/demultiplexer.

    I have a Teisco SX-400 four voice synthesiser.

    It has 1 x CV/Gate inputs that goes to all 4 voices.(CV = CONTROL VOLTAGE for note value/frequency).
    All voices are played at the same time(stacked) from one CV/Gate combination.(Gate = NOTE ON, duration that the note is held)

    I would like to modify the circuit so that I can control each voice independently with 4 x CV inputs instead.

    My attempts at a circuit description follow.The schematic I have is difficult to read so I have redrawn it as best I can.

    Internally generated CV's enter at R5 100Ω, and are buffered? by a 4558 op-amp(would like to understand more about the op-amp section).

    They enter a 4051 which demultiplexes out depending on control signals at A, B, and C, to give 4 individual signals that then go off to a TL084 sample and hold circuit? and off to the voice circuits VCF VCA etc after that.

    Can I add another 4051 that works in reverse, and have this switched from the same control signals at A, B, and C to multiplex four inputs to one output which then enters at R5 100Ω, and from then on uses the same circuits as originally intended? (this is represented by U3 in the schematic)

    Thanks in advance
     
    Last edited: May 15, 2013
  2. apprenticemart2

    Thread Starter Member

    May 7, 2013
    67
    7
    Well, I was able to get something working all by my lonesome. I'm currently testing my modifications, but everything seems to be working as expected.
    I'm just updating this post with my solution to help inspire others who get stuck. If you persevere for long enough the answers eventually come.
     
  3. wayneh

    Expert

    Sep 9, 2010
    12,151
    3,058
    Sometimes! ;)
     
  4. tubeguy

    Well-Known Member

    Nov 3, 2012
    1,157
    197
    A famous quote:

    "Never give up, never surrender" :cool:
     
  5. wayneh

    Expert

    Sep 9, 2010
    12,151
    3,058
    A man's GOT to know his limitations.

    - Dirty Harry
     
    tubeguy likes this.
  6. tubeguy

    Well-Known Member

    Nov 3, 2012
    1,157
    197
    Laughed out loud !
     
  7. apprenticemart2

    Thread Starter Member

    May 7, 2013
    67
    7
    OK so I was premature in my celebrations here.
    I think I'm getting some sagging in the voltages I input(don't know why this would happen), but I haven't tested/measured this effect yet(will see what I can find)

    I haven't put in a unity gain buffer on my inputs. If I add a buffer (TL084?)Will this fluff up the voltage again if it tries to drop, or do I need to try to duplicate the little op-amp circuit that's in my schematic above(is native to my synthesiser) for each of my 4 control voltage channels?

    My synthesiser plays fine from itself and the control voltages generated by my CV sequencer all check out fine but I'm not getting the correct frequency of note.

    What should happen is 0 volts will give me a C2, then +83.33mV will give me the next note up, then +166.66mV for the next note, then +250mV for the next and so on, but it will play the same frequency of note twice or thrice in a row then the fourth note will be at the correct pitch, then 2 or 3 notes at the same pitch again and then it will jump up to play the correct pitch.

    Any ideas anyone?
    What does the op-amp circuit that's in my schematic actually do?(some kind of buffer?)

    Cheers
    Martin
     
  8. tubeguy

    Well-Known Member

    Nov 3, 2012
    1,157
    197
    I used 4051's quite a while ago, but I have an idea. .
    If the supply voltage is 5 volts I believe the on-resistance is relatively high - like 1k.
    It is in series with the other resistances which adds up to about 2 k in series with 2.2uf caps. Maybe that causes enough lag in charging the caps to cause this issue.

    If you can power the 4051's from 15 volts I think it will lower the on resistance to something like 80 ohms. (From memory :rolleyes:)

    Other idea: put a buffer directly after the 4051 output.
     
  9. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    U4a thru U4d are opamp channels wired as unity gain buffers/voltage followers. You should buffer your inputs as well. Note that U1a, U1b is used as a bipolar voltage clamp to prevent the CV IN's voltage from exceeding the thresholds set at the junctions of R1, R2 and R3, R4 less the drops across D1, D2.

    The TL084 isn't a terrible opamp, but it's a very old design (70's) and there are newer, faster, lower-noise opamps readily available quite inexpensively. Take a look at the MC33078 dual and MC33079 quad low-noise opamps from ONSemi; the input offset voltage is typically about 1/10th that of the TL08x family. Since that offset error will be copied directly to the output, it will directly impact the frequencies your synth generates.
     
  10. apprenticemart2

    Thread Starter Member

    May 7, 2013
    67
    7
    This is a great help guys. I will add a buffer on my inputs and see what improves.

    Thanks for the circuit description SgtWookie. There's some terminology in there that I can now google and learn more about.

    Cheers
    Martin
     
  11. apprenticemart2

    Thread Starter Member

    May 7, 2013
    67
    7
    EDIT:I'm looking learning about voltage level shifting at the moment.Just found some stuff.

    Just to consider a different direction/solution to my problem.

    Can anyone suggest a way (IC/circuit) to change my 0volt to +5volt control voltage inputs range to give a range out of -2volts to +3volts?

    (so 0volts in would equate to -2volts out, 1volt in would be -1volt out, 2volts in would be 0volts out etc.)

    I don't know if modern PIC's(PIC16F1783 for example) can give a negative to positive voltage range from their DAC out.



    Cheers
    Martin
     
    Last edited: Jun 16, 2013
Loading...