Diy stompbox size Chromatic tuner.

[ubaid88]

Hey i found a great Chromatic tuner. It is bit difficult but great. Have a look at this.

from http://schickt.de

TheGIMMICK is a DIY PIC16F84A-based chromatic guitar tuner. The PCB is roughly 40x60mm so the whole thing can be built into a regular sized stomp box.



Schematic:
http://www.schickt.de/data/gimmick/theGIMMICK_sch_0.73.pdf

Pcb Transfer:
http://www.schickt.de/data/gimmick/theGIMMICK_brd_0.73.pdf

Component layout:
http://www.schickt.de/data/gimmick/theGIMMICK_qv_0.73.pdf

Bill of Material:

Pic Program:
http://www.schickt.de/data/gimmick/theGIMMICK_0.9g.hex


More here:
http://schickt.de/index.php?entry=entry080606-004307

[wavley]

Very cool

[pepi5290]

hello guys,i'm a student and i will build the chromatic tuner for my exam,i wuold need the list code of the program.It's very important for me .
Pls responds
Thx all.

[Taylor]

If you mean source code, he has not released it. The Hex code is there, though. I built this project last year. It works, but it's  fairly jumpy.

[Ice-9]

That looks like a cool project , I wish i knew more about pic code though.

[pepi5290]

This file hex is very bad, because my project does not work well

[egasimus]

^ Do your own homework

[ayayay!]

Why can't I find the output of that thing? 

[slacker]

It ain't got one.
If wanted to hear the guitar whilst tuning you'd have to add a buffer in front of it and take the output from that.

[ayayay!]

Oh so just switch the input(s)  Got it. 

[Galego]

I've done my own code for that same schematic. The results aren't that great. Since the PIC isn't receiving the unaltered guitar signal, but a 'square wave' version, the resulting harmonics when you play a string on a guitar really mess up things. Basically, all you need to do is to count how long it takes between two High or Low states of the input pin, try to do some averaging or some other clever way of filtering your results and figuring out what that time relates to in frequency.

Here's a video of when it was still on the breadboard:

[Taylor]

I think taking the fundamental extractor from the OC2 and replacing the comparator in this tuner with it would make the tracking much better.

[defaced]

The results aren't that great. Since the PIC isn't receiving the unaltered guitar signal, but a 'square wave' version, the resulting harmonics when you play a string on a guitar really mess up things.

Interesting.  In the commercial tunes I've used, I've always gotten much better results with a highly distorted tone than a clean signal.  I wonder what's different. 

[Galego]

I think taking the fundamental extractor from the OC2 and replacing the comparator in this tuner with it would make the tracking much better.

Please, do explain. With pictures if you can

[Taylor]

The lower part of this schem is the fundamental extractor.

http://www.aronnelson.com/gallery/main.php?g2_view=core.DownloadItem&g2_itemId=9447&g2_serialNumber=2

There's a sallen-key lowpass, then precision rectifiers split the upper and lower halves of the waveform and (as I read it) do a kind of comparison to see if they both cross zero within a cycle, or only one. If only one does, this means that there was a harmonic tripping the comparator instead of the fundamental. That's my analysis anyway, which might be clarified by someone else but I think it's roughly correct.

So, this kind of circuit does a good job of throwing away harmonics - better than a plain old comparator anyway.

[Skrogh]

Only problem with the pitch-tracker from the OC-2, is that it is quite unstable especially on open strings...

As I interpret the OC-2 pitch-tracker, signal goes through the sallen-key lowpass, then two passive lowpass', the signal is then split into a positive and negative half, each are sent through a logarithmic slew limiter, working as an envelope-tracker. These envelopes are then compared with the output form the second lowpass and the process Taylor described about the signal having to cross both the negative an positive envelope, is applied. What happens when the OC-2 "jumps" is that the signal decays too fast to hit the envelope, thus the effect cuts out.

This is a bigger problem on open strings, since they have a more power full attack, and faster decay (I believe). They tend to set the envelope high, then "duck" and avoid it.
I can see if I can find the LTspice-simulation I made some time ago, if it has any interest? It might be possible to change the rise and/or fall time of the envelope to compensate.

[Marcos Camara]

What is the frequency of the crystal Q1?

[scratch]

4MHz according to the BOM found here ...

http://www.schickt.de/data/gimmick/theGIMMICK_bom_0.74.txt

[Marcos Camara]

Scratch,

thanks!

[Marcos Camara]

I found the circuit very unstable (varying too). I tried changing the feedback resistor of the 1st stage of the TL072 putting a 10K resistor in series with a 1M pot. The result was a slight improvement but not significant. I tested several values ​​for C1 in order to cut harmonics, but the idea was even 68nF.
Someone managed to stabilize the circuit?
thanks.