I'm in the process of creating a device which ultimatly will take an input signal, i.e. a guitar signal, and detect the frequency and ultimatly turn this into a CV (Control Voltage) for an analogue synth. This will allow me to control analogue synths via a guitar.

The first part of my project is simply converting a frequency into a voltage. For this I am using a LM331 chip, right now I have this breadboarded up as per the data sheet instructs in its simple F/V conversion.

Figure 6
http://www.national.com/ds/LM/LM231.pdf

Right now the output is not changing with frequency but remaining at about 13v (I am supplying it with 15v) I am ordering a new chip incase this one has somehow been damaged so hopefully this may recifty this problem.

One problem I am expecting even once I recieve some sort of fairly linear response is delay. There are also going to be many other problems I expect.

I'm hoping someone (or maybe even a few!) will hopefully take some sort of interest in the project and it's progress and maybe even be able to provide some advice when I need it as I am by no means an expert in this area!

One quick question I have been trying to find an answer to for a little while is what is the difference between a LM331N & LM331AN? Don't seem to be able to find any difference?
Thanks for reading!

 

The circuit you mentioned with the LM331 and the datasheet you provided are not the same. The LM231 is a voltage to freq. converter, the circuit you want, I think, is the LM331. Can you verify this and possibly post your schematic or circuit pictures.

Is your circuit as seen below?

iONic

 

Ionic,
Take a look at Figure 6 in the datasheet. It shows a frequency to voltage converter, which is opposite the normal function of the IC; a voltage to frequency converter.

tjs2010,
Delay is one problem; it may be in the millisecond range though. Polyphony is another; if you play more than one note at a time (ie: chords), you'll get somewhat of a mixed-bag output - it'll be somewhere between the highest note played and the sum of the output frequencies.

Check to make sure your grounds are all good.

 

In Fig. 6 of the data sheet (and reading the description of the circuit) it refers to a square wave ("pulse input at Fin is differentiated by a C-R network" etc.) as the input stream. If you just run the guitar signal into this "straight" you're not going to get much. You need to turn the guitar signal into a square wave first. You can do this a few ways; clip the bejeezus (technical term) out of the signal by running it through a gain stage set to something like x100(0) or maybe through a comparator (where you can then set the "trip" point as well). This kind of circuit has been done a bunch of different times, I'd tool around Google.

Mike T.

 

Hook your guitar to your audio input on your computer and try one of the free oscilloscope software programs out there. A picture is worth a thousand words, and the picture you'll see is what you're being warned about. It won't be a sine wave or a square wave - it'll be a shotgun blast. Unless you're tuning a single string.

But, you might be able to pick out "patterns" - just as the Fourier transforms do - and have something respond to that.

 

Few synthesizers attempt to reproduce the exact sound from binary data. Some very expensive Yamaha units DO synth a piano nearly perfectly, even accounting for sympathetic string vibrations with DSP.

The majority of keyboards and computer MIDI cards use many samples of an instrument and replay them at the requested pitch from input by keyboard.

 

One thing to keep in mind is that hitting the same note on a different string excites the higher harmonics. When you hit a string there is the fundamental frequency and the higher frequency components to worry about. For instance the next harmonic is at exactly one octave higher then the fundamental. If the note is 440Hz (A), then there is also a 880 Hz frequency component overlayed at a lower amplitude.

http://en.wikipedia.org/wiki/Harmonic