Easiest way to analyze frequencies?

[eightsevenzero]

I just built a distortion circuit and I'm trying to pinpoint the mid/highs emphasis I hear so can even out the EQ to include a fuller sound. There's no dedicated tone control in the pedal, just mainly built into the circuit I/O. What's the easiest way to graphically view frequencies so I can know what I'm dealing with here? I run a mac and don't have spice, also my working knowledge of doing the math to determine what's going on in this circuit is very limited, I usually use my ear. I didn't socket the caps on this one, but I'm going to breadboard up another to plug in the new values after I can see what I'm dealing with currently.

Thanks!

[andy-h-h]

Hello - what are you working on (name/brand) and where is the schematic?   Spice is not as scary as it first seems, although on a Mac it's harder to use than windows. 

[Digital Larry]

Use the spectrogram view in Audacity.

[anotherjim]

You can get it by ear using a graphic equalizer or a parametric eq. For the graphic, the more bands it has, the better. With the parametric, it can help to set gain for full boost and turn the frequency knob to find the ones that really stand out, and then try it with full cut.

[Vivek]

Either you can measure the hardware

or calculate it

or simulate it

If you choose to simulate, there are some fairly easy simulators on the net

example : http://www.falstad.com/circuit/index.html


Please post the schematic of your circuit

[Mark Hammer]

You can get it by ear using a graphic equalizer or a parametric eq. For the graphic, the more bands it has, the better. With the parametric, it can help to set gain for full boost and turn the frequency knob to find the ones that really stand out, and then try it with full cut.

I'm with Jim on this.  A simple one-band sweepable EQ with low Q strikes me as the most practical and speedy solution for finding where the intrusive peaks are.  And after you've done the detective work and evened-out the tone of the distortion, you have an EQ pedal!

The logic of this approach is that one sweeps the resonant frequency of the EQ, with cut set to about 6-9db until the distortion seems to get "calmer".  The drawback is that, unless one has a very precise circuit, you don't know what the actual resonant peak is; you just have a more pleasing-sounding distortion.  But then, even if you did know the specific frequency, what could you do to tame it?

[amptramp]

You can take a waveform expressed as a voltage over a time scale and use the Fourier Transform to change this from the time domain to the frequency domain.  There are Fast Fourier Transform (FFT) apps that run on a soundcard and this can be done cheaply.  The Fast Fourier Transform makes use of certain symmetries in the Fourier Transform to simplify the calculations.

As one bit of music trivia, Jefferson Airplane was the first rock group to use an FFT processor to ensure the sound spectrum was equal at all places in the venue.  At the time, an FFT processor was about the size of a dishwasher and cost about $30,000 to $50,000, about the price of a house at the time.  Now, this is a $100 app on a soundcard.

[EBK]

You could also use a function generator and an oscilloscope.  Sweep the frequency on the input and watch the magnitude of the output.  You should be able to see where the magnitude peaks are and note the frequencies.

[bowanderror]

I've been looking in to this recently as well, and after being stumped by Room EQ Wizard, I stumbled upon a method using a DAW + some free Meldau plugins.

If you scroll down to page 20 in this Eurorack VCF manual (ThreeTom MS-22), you'll find a tutorial on setting up oscillator & white noise signal outputs, as well as an input spectrum analyzer channel. The tutorial is for Ableton Live, but you can use any DAW that supports external VST plugins.

I tried originally to use Audacity, but was only able to get a spectrogram for recorded clips, not live audio inputs. Anyone have any advice?

[j_flanders]

I tried originally to use Audacity, but was only able to get a spectrogram for recorded clips, not live audio inputs. Anyone have any advice?

For recorded clips I use Audacity since it provides a convenient way to export the data.
You can import this data in Excel for example to create a graph. I find this useful because you can have multiple frequency response plots in one graph.
I'm not aware of any way to do spectrum analysis on live input in Audacity.

For live input I use Reaper and the Voxengo SPAN plugin (both are free)
https://www.reaper.fm/
https://www.voxengo.com/product/span/

My signal generator is usually another laptop, running Audacity, generating the tone or noise I need.

But specifically for the OP's question, I usually quickly draw the schematic in Falstad.com and see what effect an R or C  has on the frequency response:
http://www.falstad.com/afilter/circuitjs.html?startCircuit=filt-hipass.txt

[FiveseveN]

For live input I use Reaper and the Voxengo SPAN plugin (both are free)

As the nag screen will dutifully remind you after 30 days, REAPER IS NOT FREE! But it does come with noise generators and "Frequency Spectrum Analyzer Meter", which is resizable, unlike SPAN.

Sound Forge also has a comprehensive spectrum analyzer that works with live sources too.

[j_flanders]

As the nag screen will dutifully remind you after 30 days, REAPER IS NOT FREE!

That's closer to the truth than what I said.
I'm 'still evaluating'...