[Noisebridge-discuss] seeking circuit idea: frequency to voltage

[jim]

   i love this discussion. best i can tell, the 
problem definition should be improved: what is 
the nature of the signal you want to measure, 
what of that signal do you want to measure, and 
what's the time constraint of the measurement? 
   for example, if the signal is voice, we can 
figure there's a fundamental frequency, there 
are harmonics, and there's noise. a bar graph 
indicator that is allowed to be imprecise with 
a simple, hobbyist-level circuit, can probably 
only usefully measure something in the domain 
of power (area under the curve, an integration 
type of measurement). seems to me at best it 
might successfully reflect the fundamental 
frequency of the voice, and that could be dicey 
in the case of a rapidly changing inflection 
(some one speaking excitedly), tho' it seems 
like one might be able to make it useful for 
singing, where the fundamental is consistent 
for many, many milliseconds. 
   hoping for more from the experts, 
jim 




On Fri, 2009-12-11 at 10:32 -0800, Christoph Maier wrote:
> On Thu, 2009-12-10 at 20:55 -0800, Michael Shiloh wrote:
> > a student wants to build a bar graph type display that will show 
> > frequency, not voltage. the frequency is in the audio range.
> > 
> > i know there are chips to do this, but i'd like to find a circuit using 
> > more basic and common components, like a 555, 741, 2n2222, etc.
> > 
> > precision and linearity is not an issue. it's merely a visual element.
> > 
> > any ideas?
> 
> Is independence from the _shape_ of the input signal an issue 
> (i.e., that a sinewave [=flute] should give the same effect 
> as a rectangle [=highly overdriven <whatever>])?
> 
> What do you want to have shown if you feed in a _mix_ of frequencies,
> i.e., a chord??
> 
> Jonathan's comments are the correct answer of someone highly Skilled In
> The Art, 
> under the implicit assumptions 
> that you actually want to measure frequency with as little
> cross-sensitivity as possible, 
> and that you want to measure the amplitude of each frequency
> separately. 
> 
> The other suggestions so far trade off simplicity against
> cross-sensitivity. 
> 
> Christoph
> 
> 
> _______________________________________________
> Noisebridge-discuss mailing list
> Noisebridge-discuss at lists.noisebridge.net
> https://www.noisebridge.net/mailman/listinfo/noisebridge-discuss
>