Print PDF
Favorite
Email
Hi, my name is Colin. I am an avid guitarist, and I am always looking for cool new sounds. I'm a big fan of the guitarist Paul Gilbert, and I heard him play this crazy effect in one of his songs, so I decided I would try to emulate it. The result was a very long endeavor, but in the end a successful one.
I have created a digital loop pedal. The pedal itself is based upon the PIC16F877 microcontroller, and is complete with 256kb of RAM for storing sound. The unit achieves true bypass with the help of a relay, and runs on a standard 9V power supply.
If you want to build a pedal like this, be warned- it is a daunting task. The complexity of this project reached the upper limit of what my brain can autoroute on a prototyping board, and was very frustrating many times when it didn't work as planned.
In the end however, I learned a lot. I think this is a great project to try, but only try if you are really dedicated. For those interested, I would suggest you design your own, using mine as a template to know what works and what doesn't, rather than copying directly- though if you wish to copy, you are more than welcome.
I have created a digital loop pedal. The pedal itself is based upon the PIC16F877 microcontroller, and is complete with 256kb of RAM for storing sound. The unit achieves true bypass with the help of a relay, and runs on a standard 9V power supply.
If you want to build a pedal like this, be warned- it is a daunting task. The complexity of this project reached the upper limit of what my brain can autoroute on a prototyping board, and was very frustrating many times when it didn't work as planned.
In the end however, I learned a lot. I think this is a great project to try, but only try if you are really dedicated. For those interested, I would suggest you design your own, using mine as a template to know what works and what doesn't, rather than copying directly- though if you wish to copy, you are more than welcome.
Remove these ads by
Signing UpStep 1Audio-only demo
This is an audio-only demo I recorded.
| « Previous Step | Download PDFView All Steps | Next Step » |
Is the PIC 16 bit ? (or a DSP chip?) Because the sample quality is much better then most 8-bit audio projects I've heard. Even more amazing 'cause your using a serial RAM...
The PIC is programmed via a PicKit2. You can google that to see what it means. As for layering, it would require memory that is twice as fast and twice as long, but it would be possible with a faster PIC. I don't know what overdubbing means.
Preamplification is achieved by a two stage operational amplifier design, pretty straightforward. Actually the design presented here is terrible, but any operational amplifier when used correctly would suffice.
The cost I don't recollect, I imagine it was less than $40.
Sometimes you can hide low res sampling by piping into reverb, etc., but this really does sound great...
I never considered the reverb idea, perhaps I should think about that. Thanks for the advice!
Well done, I look forward to seeing more of your work!
Barnaby
CAN U SEND TO ME SCHEMATIC PLEASE
IF U WANT,U CAN SEND TO ahmad_dimarzio@yahoo.com
tq b4
I was trying to make a similar project but i can only get a maximun sample time of 625Hz.
Thanks.
I'm working on my own digital pedal based on the PIC18F26J11 (Just what I had laying around). I'm not sure what effect I plan to try to create, but It's fun to have a go.
At the moment, I am trying to get the chip to act as a buffer, all I want it to do is read the analog into a variable, and then output it to the DAC (Mine uses 4.7k and 10k resistors, which I know gives a lower quality signal). I have a lot of trouble with noise, and have been using 100nF capacitors in places to see what I can do.
I have found I should have one series with the input, series with the output, and then one to bypass the output to ground.
I have tried the opamp circuit you have presented, but I have no idea what it's purpose is, how it works, or if mine even worked the way it was supposed to.
Could you please let me know how it was supposed to work, or even just tell me how the opamp is configured (-ve voltage amplifier?). I guess it is supposed to change the guitar signal into a 0 - 3.3v representation of the signal?
Cheers,
MM
You are correct, the op amp is there only to amplify the input guitar signal to something a little more usable for the ADC, since an unmodified guitar signal usually has a very small amplitude.
From what I remember (this was a long time ago) I just biased the guitar input up to something like 1.7 V, then amplified by a pretty large gain. In this particular configuration the gain is negative, but that's alright, it just changes the phase of the signal.
If you are going to try to do something similar, probably you should design the amplifier yourself though, looking at my old schematic, there are a lot of improvements that could be made to it that would improve the sound quality. Like I said, it was made a long time ago, I didn't understand too much about circuits back then :P
Let me know when you finish your project, I'd love to hear more about it.
Anyway, good luck (and thanks for making your question specific, you wouldn't believe how many people come on here and ask me for "help building the pedal" without specifying what they don't understand about it!)
A bias means to set the upper limit right? Sorry, I'm new to opamps. At one point, I was inverting the signal before the output, and found it made no change at all.
I'm looking at building a flanger effect at the moment, as it looks doable. I need about 20mS worth of recording, and then mix that into the live feed.
It's great you posted your instructable, as it has been a stepping stone. A friend suggested I make a digital pedal, and I leapt right in knowing all about ADC and DAC. The problem came with noise, and was solved with capacitors.
Cheers,
MM
Here is a link to another circuit I designed (much better quality amplification actually) which shows the use of bias:
http://www.getlofi.com/?p=2945
if you look at the left/right channel amplification, the 1uF capacitors serve to prevent DC from flowing through from the inputs. Effectively, it just adds a DC voltage to the input signals so that they can be properly amplified.
Anyway, good luck with that. 20 ms of recording time should be totally possible, even at high sampling rates, so looks like you are on the right track. Let me know if you have any other questions.
I spent a whole lot of time learning, and in the end I designed my own opamp circuit. There is a 10k input resistor and an 8.2k feedback resistor. The +ve input is at 1.03v through the use of a 10k and a 22k divider. My simulation predicted this will create a 0v to 3.3v wave from a -1.8v to 1.8v input.
I still have two 100nF caps on the output, one to ground and one to output. These seem to cut down on noise dramatically.
It was an excellent experience.
Cheers,
MM
I will say this though, a capacitor would not work at all.
What kind of DAC do you use?
Take a look at Step 5 under the DAC section for a link to the schematic I used, or refer to my notes in Step 7 for a poorly-drawn schematic that I put in my notes.
And how did you write the code so that the playback frequency was exacly the same as the recording frequency?
Thanks!! Awesome stuff
I got the playback and recording frequencies synchronized by the use of interrupts and calibration, essentially. Again, read the code if you want an example of how I did it (though I am very confident there are smarter ways of doing it)
Would the song be Get Out of My Yard?
From,
Owen
how many samples / second?
Maybe I'll make something like this, but using the Btc Alghoritm of roman black to encode and decode audio.
If you can encode or decode a byte of audio in under, say, 60 or 70 us, you could make it work though. It probably would have been smart of me to write this whole thing in assembler- it would be much faster code in the end.
www.romanblack.com/BTc_a
Yups! Sorry for so many questions ;)