Announcement

Collapse
No announcement yet.

Simple switching problem (too hard for me!)

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

  • Simple switching problem (too hard for me!)

    Dear Musical Electronics Experts,

    I have got two groups of 3 x 4066 (QuadSW) of which I need to select one at a time via two pushbuttons. How to do that?

    Since it only switches small 5V currents, I was thinking to just power only the group of IC's (pin 14) via the two pushbutton switches, because only one of the twelve (from 3 IC's) gates of each group will/should be on at a time and the moment one group gets powered and then activate the desired switch. Activation of this switch needs to be only momentarily (a fraction of a second or so to initiate a MIDI software setting).

    So, maybe in this case powering the IC's via the footswitch might be the simplest (non-professional ) solution? It's like turning on the TV via inserting the plug into a power point.
    If you know of a better solution could you please draw a simple circuit for it of how it connects to 4066 so, that I can follow your advice. I'm good at soldering according to a schematic.
    Albeit the scenario looks pretty basic, it turns out to be complicated (for me at least).

    ----------------------
    In case you want to know for what it is : It's for a footswitch controlling the the lowest octave of my (D20) Roland Synth to play/control the 'Live-Styler' accompaniment via a circle of five (accordion button) layout. On the bottom of my page http://www.live-styler.de/home/?Ente...ohannes_Drinda
    is a link to a my JANKO keyboard project, which I build from wood... Some members in this forum might find it as exiting as I do?
    Kind Regards from jjj in Chile
    Last edited by jjj; 02-25-2007, 10:56 AM. Reason: clarify text

  • #2
    The pin 14's require to be powered all the time. Without power the 4066 will probably go "Sort of On" and leak signal through. So, to select one or other of two groups, you need some logic. You could use AND gates, or mickey mouse it with resistors and diodes.

    The good news is that the 4066 needs a very low supply current indeed, only a few microamps or so. That means you can power it off almost anything.

    If you're using it to switch buttons on a Roland keyboard remotely, you should power it off the same supply and ground rails as the keyboard scanning circuitry in the Roland. The reason is that the 4066 can't properly switch signals that are outwith the voltages on its supply and ground pins.

    These limitations are common to practically all analog switches, not just the 4066. If you find it too confusing, you may want to investigate MOSFET optocouplers like the H11F1. These really do work like little ideal switches, and need no power except for that needed to light the LED that turns them on.
    Last edited by Steve Conner; 02-26-2007, 02:12 PM.
    "Enzo, I see that you replied parasitic oscillations. Is that a hypothesis? Or is that your amazing metal band I should check out?"

    Comment


    • #3
      Thanking you, Steve Conner, for your good advice.

      The pin 14's require to be powered all the time. Without power the 4066 will probably go "Sort of On" and leak signal through.
      Oh, something new! I thought when 4066 is off, it's off... but you are right, because it's a chip not a mechanical switch.
      Well, in that case I can wire the 4066 so, that there isn't any signal/ voltage on neither 4066 switch. It's only in the moment I press the footswitch that the desired voltages appear at the designated 4066 switch. I guess that's the solution! (?)
      So, to select one or other of two groups, you need some logic. You could use AND gates, or mickey mouse it with resistors and diodes.
      Interesting, too. Yet, how do I connect AND gates between 4066??
      Optocouplers and Reed SW's work like relays, but take too much space. The 24 switches (of 2 groups) take only 6 x 4066, otherwise it takes 24 individual switches. Beside, I live here in an isolated area in the Chilean Andes and ReedSW and optocouplers aren't easy to obtain, but I got lots of IC's and discrete parts.
      ...you should power it off the same supply and ground rails as the keyboard scanning circuitry in the Roland.
      That's clear and that's what I did. I soldered 5V and ground (from its disk drive) along with all other connections to a new socket on the Synth. I have got the circuit so, it was easy to find it. (Interesting how they positioned the second SW for every key to measure touch sensitivity...)

      Thus, I would be grateful to you if you could just show me how to connect those AND gates between the 4066 so, that I can switch/ activate one group (i.e. 3 x 4066) of two groups at a time. Kind Regards, jjj
      Last edited by jjj; 02-26-2007, 06:29 PM. Reason: adding a solution...

      Comment


      • #4
        [IMG][/IMG]

        Comment


        • #5
          A revised version of my switching problem

          I try to keep the musical details to a minimum and just focus on the problem:
          I have got two groups of 3 x 4066 (QuadSW) of which I need to select one group at a time via two pushbuttons.
          Only 5V (low currents) switching is involved and only one of the twelve gates (from three IC's) of each group should be on at a time (for a fraction of a second).
          My idea is to switch the 5V via one pushbuttons (of two in a footswitch) for each group to pins 1,3,8,10. Only the chosen 4066 switch will then be on, even though all 12 (4066) IC's switches (from one group) will be powered at the same time,(but non-of the other gates).
          Could you please check if I'm 'ticking' right? I forgot to mention that only one of the 12 gates in each group at a time receives 5V from an accordion bass button/ switch and that's when I press the footswitch to obtain the desired output to control Live-Styler software via a pentasonic layout (which is the 120 button accordion bass part).

          It's the same as pressing 2 piano keys of a Synth (in the lowest octave at the same time) briefly. If, e.g. the piano key C and A# is pressed (A# the nearest black key to the left from C in the lowest octave, used only for Live-Styler) to switch Live-Styler accompaniment to minor chords).
          To switch to a C-7th chord the nearest white key (B) needs to be activated, etc. That can be quite complicated and that's why I chose to do it via a circle of quints (accordion button) accompaniment. Together with my Janko keyboard I'll have the easiest to learn & play MIDI instrument. If you are musically creative... here's the chance to make it happen for you, too!

          Comment


          • #6
            Mea culpa! I totally confused the issue, because I somehow mistook the in/outputs.
            The gates is not the inputd of a 4066 switches! 1,3,8 & 10 are the inputs.
            That opens a totally knew oportunity... for me. Now I have to rethink my plan. Gee, I have been blind!
            I have been out of my electronics hobby for some years now and that's why I have to catch up again.

            Here's a relay version of what I try to do. I think that'll make it much clearer:

            Comment


            • #7
              Progress ('Regress'?) Report:
              I desoldered the old circuit and rebuilt this one:

              My joy was short-lived... When I press the acordion button C and then the (Minor) footSW all gates are activated and I get a chorus of all 12 sounds! Does it mean all 6 IC's are 'kaput' or is there leakage?
              What to do? Maybe I have to use opto-couplers or reed relays? I'm left with is this basic diagram with relays:
              http://img63.imageshack.us/img63/11/bbbpg6.jpg
              All I want is that when I press an accordion button (one at a time) and momentarily the footSW, another, specific accordion button/key will be activated (to activate that Live-Styler software). Maybe there's a much simpler way?
              Last edited by jjj; 03-08-2007, 08:52 PM.

              Comment


              • #8
                I'm not much of a designer so probably can't help with the most elegant circuit solution, but in my repair experience I find the 4016 & 4066 to be a common point of failure so consider them a bit fragile.

                NEC makes a 4-section darlington optocoupler - maybe that would be a bit more robust.

                http://www.mouser.com/search/Product...551-PS2502-4-A

                Edit: I think I see the problem now: If you draw out the full schematic with all the keys hooked up I think you will find that all the keys are actually connected to each other. When you press "C" it also energizes "A#", which also energizes its associated minor key, and so on all around the circle. The keys don't know the difference between being physically pressed and electrically energized. I don't even see a way of curing that with steering diodes.

                Maybe some kind of "first takes precedence" delay logic could work, but I'm guessing it would be problematic in actual real-world playing situations.

                Must be somebody around here who is good with switching logic...
                Last edited by Mark Black; 03-08-2007, 07:38 PM.

                Comment


                • #9
                  Thx Mark Black,

                  http://img63.imageshack.us/img63/11/bbbpg6.jpg

                  Note: The above pic only shown one connection; that of the C key. Each group (Minor & 7th) has 12 different connections adding another key.
                  Looking at the simplicity of this relay circuit I guess it's less complicated to just use/ build a bank of 12 electro-mechanical, easy to operate switches.
                  I harvested a quantity of gold/steel alloy, high-tensile contact wire from an old Dr. Boehm electronic organ with little boards. From that I can easily build a dual 12-pol switch controlled via a push bar. Regards, jjj

                  Comment

                  Working...
                  X