Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

555 Voltage Controlled Oscillator 1v/octave

  1. Jun 25, 2012 #1
    Hey guys, I'm working on a voltage controlled oscillator to use in a modular synth system and here's the schematic. http://ecelab.com/circuit-vco-555.jpg [Broken] I also have a one octave keyboard that I made and each key is 1/12 of a volt. My question is to what frequency I should tune the oscillator to get 1 volt per octave?
    Last edited by a moderator: May 6, 2017
  2. jcsd
  3. Jun 26, 2012 #2
    Good luck with that....

    I couldn't find any spec for the 555's voltage control but I believe it has a narrow range and is probably not very linear in it's response. The further complication is that you want exponential, not linear, control anyway. And you will want it to be temperature compensated so your tuning doesn't drift all over your key signature.

    I googled "1 volt per octave voltage control oscillator" and came up with some very nostalgic pages about Serge and EMU synths (I also noted a price list for the Serge modules, the VCO setting is at $1600US...oy...I remember buying mine from the factory for about $150 and thinking it was expensive). But this page has a schematic and description that looks pretty good: http://www.cgs.synth.net/modules/cgs48_vco.html
  4. Jun 26, 2012 #3
    What if I used a LM13700? Here's the datasheet if you need it http://www.farnell.com/datasheets/87555.pdf. Do you think I'd be able to get 1v/octave out of it without too much hassle? Ultimately, I'd like to use an IC to keep the footprint small. I'm making a mini modular.
  5. Jun 27, 2012 #4


    User Avatar

    the 1 volt per octave VCO dates back to the original Moogs. i had never thought of the 555 being a very good linear VCO. perhaps you might get away with it. if you refer to


    and use the "astable" circuit, and set R2 to zero and connect R1 to your linear control voltage, maybe you'll be able to make a linear VCO outa this. then to get to an exponential VCO, you might use your LM137000 as a "Logarithmic Current Source", you might get this to sorta work. it might not tune (be true to a perfect some volts per octave curve) well enough.

    for log amps, i had always considered building one out of Analog Devices parts and for a VCO, you should look up circuits for that. i know in the olden days, a company called PAiA made kits with some pretty good and cheap linear VCOs.

    so you need to figure out what makes a good linear VCO and what makes a good exponential converter, and then to figure out how to scale and connect the two.
  6. Jun 27, 2012 #5


    User Avatar
    Science Advisor
    Gold Member

    Way back when, I used Nationals LM324 Low Power Quad Op-amps for the VCOs (and bandpass filters) of a synthesizer. I made the VCOs using a single supply as in their datasheet.


    Instead of trying to manipulate your VCOs output to do the scaling, have you considered altering your keyboard to give the "scaled voltages"? In my opinion this would much simpler and your VCOs good linearity can be realized.

    Attached Files:

    • VCO.jpg
      File size:
      37.4 KB
  7. Jun 27, 2012 #6
    Found another example schematic. It looks pretty much the same as my previous post but uses what appear to be more modern chips: http://www.birthofasynth.com/Thomas_Henry/Pages/XR-VCO.html with an XR-2206 function generator and an exponential control converter.

    The trick, as others have mentioned, is to get a good linear frequency response from the oscillator and drive it with a good temperature compensated exponential converter on the control voltage.

    Way back in the day, I think it was Dave Rossum at Emu, developed a set of VCx chips for their synths. But I can't find any mention of them online now.
  8. Jun 27, 2012 #7

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    to that end you'd probably want a current source to establish the charge rate of your capacitor.

    Have you an understanding of how the 555 operates?
    There's plenty of tutorials around. here's one.
    Changing pin 5 shifts your operating point along the charge curve which is nonlinear.
    But basically, halving voltage at pin 5 will double the rate.
    Pin 5 starts out at 2/3 supply voltage.
  9. Jun 28, 2012 #8
    Thanks a lot guys! I'm starting to understand this better. I didn't know about expo converters previously and I couldn't fathom how a 1 volt change would ever equal an octave. That said though, I think I'm ditching the 555 for the LM13700. It will be a lot smaller with more waveforms options.

    And Dlgoff, yeah, I have considered it. I'm extremely OCD though and the thought of not using the standard makes my stomach upset haha.

    As far as the expo converter, thanks for the link Schipp! I also found this one http://home.comcast.net/~ijfritz/sy_cir9.htm [Broken]. They both seem to have temp compensation, which is nice, but which one do you think would have a smaller footprint? Also, are these expo converters interchangeable with other vcos? In other words, would they work with my 555 vco and my LM13700? I know I'd probably have to turn some pots, but I'm guessing they would work with both right?
    Last edited by a moderator: May 6, 2017
  10. Jun 28, 2012 #9


    User Avatar
    Science Advisor
    Gold Member

    I think a lot of us are like you in this regard. :wink:
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook