Wah Probe work-alike

earthtonesaudio · Post by **earthtonesaudio** » 10 Jun 2008, 22:40

I'm about 80% satisfied with this:

http://www.aronnelson.com/gallery/main. ... t.png.html
Basically a "lite" version of the minimum theremin, filtered output, and driving an LED.

That's a 4049B (buffered, not the unbuffered like everyone seems to prefer), and I was able to use the sixth inverter as a "Ghost Dance," using a photocell from radioshack as the variable resistor.
http://www.geocities.com/tpe123/folkurb ... tdance.gif

I'll try to do a video sometime.
Feel free to use, but please give credit if you do.
Enjoy!

-Alex

earthtonesaudio · Post by **earthtonesaudio** » 11 Jun 2008, 00:25

Video:
http://video.google.com/videoplay?docid ... i6DA&hl=en

culturejam · Post by **culturejam** » 11 Jun 2008, 02:31

That's pretty awesome!

Can't wait to see how this progresses.

earthtonesaudio · Post by **earthtonesaudio** » 11 Jun 2008, 12:41

Thanks! So far you're the first person to say anything about it (posted on DIYstomps also). I'm still working on it, and will keep the community posted, 'cause I remember how hard it was to find info on this subject when I first started looking...

~arph · Post by **~arph** » 11 Jun 2008, 14:04

Sounds pretty good, but I seem to hear some clock noise (pitch changing as you move your socks over the foil), but perhaps this is due to the quality of the recording.
Perhaps the clock(s) speed should be increased a bit.

earthtonesaudio · Post by **earthtonesaudio** » 11 Jun 2008, 15:55

Thanks, ~arph,

You are indeed hearing the theremin tone bleed-through. I think the main reason for this is the fact that the inverter used in the wah circuit is on the same chip as the theremin circuit. Could probably be fixed by moving the wah circuit to a separate component, maybe a cm or two separation for good measure.

But you're right, the more elegant (and simplest) solution would be to tune the theremin's pitch out of audio range. I don't know if it can be done and still maintain a usable range on the antenna. I have a couple more designs I'm trying out, but I might return to this one eventually.

On a side note, I think you could even dispense with the LED driver transistor if you wanted, and move the lamp to the feedback loop of the fifth inverter. The only problem I noticed with this arrangement is due to the conduction knee of the LED. Big jump in brightness right at the forward voltage threshold of the LED. An incandescent lamp might work better, but with its own problems.

culturejam · Post by **culturejam** » 11 Jun 2008, 16:51

earthtonesaudio wrote:Thanks! So far you're the first person to say anything about it (posted on DIYstomps also).

Yeah, I saw it at DIYstomp. I guess they're too busy wagging fingers over there to stop and talk about your new design.

marcao_cfh · Post by **marcao_cfh** » 12 Jun 2008, 03:34

earthtonesaudio wrote:I think the main reason for this is the fact that the inverter used in the wah circuit is on the same chip as the theremin circuit. Could probably be fixed by moving the wah circuit to a separate component, maybe a cm or two separation for good measure.

Maybe the problem is that you built it on breadboard. Breadboards can insert lots of "things" into the circuits, and a theremin is a sensible circuit, so maybe the noise and oscillations is due to the breadboard. Try to build it on perfboard or pcb to see the results.

earthtonesaudio · Post by **earthtonesaudio** » 12 Jun 2008, 12:59

Thanks for the input. I have high hopes that I can build something that is insensitive to layout and/or breadboard weirdness.

Last night I made one using 3 opamps. Got a good range using my piece of aluminum foil (about 20cm range), but lame response time. I'll get something good eventually.

earthtonesaudio · Post by **earthtonesaudio** » 13 Jun 2008, 01:46

Got it down to 2 op-amps.
First opamp is configured like this:
http://hyperphysics.phy-astr.gsu.edu/hb ... re.html#c1(Where "C" is your body's capacitance, but put a blocking cap in there, okay?)

Second opamp is configured like this:
http://hyperphysics.phy-astr.gsu.edu/hb ... ar.html#c1

Output drives an LED-to-ground directly.

Freakishly simple. I can't believe I've not seen something like this before.Good sensing even through a shoe! One caveat: for best action, some part of your body should be grounded.

I just happened to use a TL074 quad op-amp, so I'm gonna see if I can use the other 2 op-amps as some sort of state variable filter or something cool like that.

I'll make a schem if anyone asks.

culturejam · Post by **culturejam** » 13 Jun 2008, 02:46

Sweet!

Keep hacking away at it, brother. If you get to the point of vero or PCB layout, I'll volunteer to do a test build.

earthtonesaudio · Post by **earthtonesaudio** » 13 Jun 2008, 16:04

Thanks culturejam. I have to do some more tweaking with the oscillator, cause I think it's running right on the edge of instability. Last night I had it working perfectly, then turned it off and back on... and Nothing. Oscillator stopped oscillating. Some fine tuning required of course.

But, up to that point, it was working well, with no audio bleed-through. The lowest frequency produced by the oscillator was about 33kHz.

If I get it working reliably, I'll do a schem and/or layout for a dual op-amp rather than the quad.

earthtonesaudio · Post by **earthtonesaudio** » 16 Jun 2008, 14:37

Fixed. It's reliable now.

DWBH · Post by **DWBH** » 23 Jun 2008, 21:41

Hi earthtonesaudio, is the schem on DIYStomp's gallery correct?
Great job

earthtonesaudio · Post by **earthtonesaudio** » 24 Jun 2008, 03:27

The schem on DIYstomps is for "version 1" which sucked. Version 2 rocks, but I haven't made a schem yet. I'll take your interest as a sign that I need to do so. Maybe tomorrow if I have time.

See my post below with the links to the hyperphysics web pages. Make all the R's 100k in the oscillator, that works for me. I also answered in more detail over at Aron's:
https://www.diystompboxes.com/smfforum/ ... ic=68566.0

Good luck. Schem to follow soon.

briggs · Post by **briggs** » 24 Jun 2008, 08:36

That is some great work earth, I would love to see version 2. You should get a DIY project up and get it in the "Ready to Build" section of the forum alongside the other great projects in there!

briggs · Post by **briggs** » 24 Jun 2008, 08:43

Here's the diystomp stuff for people who can't login...

Update: I got it to work with 1/2 of a TL074.

Use this: http://hyperphysics.phy-astr.gsu.edu/hb ... quare.html
And use 100k resistors, and a 100p cap (not attached to ground, but to your antenna). This gives me a range of frequency from about 100kHz to about 30kHz (antenna is about 10cm by 20cm). *Caution* I'm not 100% sure about these values. I just glanced at the breadboard this morning and I think that's what I ended up using.

And then send the output through a low-pass filter into a comparator:
http://hyperphysics.phy-astr.gsu.edu/hb ... ar.html#c1

Take the output of the comparator and drive an LED-to-ground. Adjust the voltage reference on the non-inverting comparator input so that the LED just barely lights up with nothing near the antenna, and gets brighter as you move your hand toward it.

The coolest thing about this is the lowpass filter section. It tunes the response so that the LED gets brighter as you come closer to the antenna, but you could easily invert the response by using a highpass filter instead, or even a BMP tone stack (with a good steep notch) to get inverted response as well as Bright-Dim-Bright as you approach the antenna.

This could work with a dual op-amp, but I have TL074's and that means there are 2 extra op-amps at my disposal for use as wah circuits or whatever.

earthtonesaudio · Post by **earthtonesaudio** » 24 Jun 2008, 12:46

Thanks briggs. I want to make sure I can get a "boxed up" prototype to work before I say anything is verified but I'll put up an unverified schematic as soon as I finish drawing one (not trying to tease people, just really haven't had time). Thanks for posting the text from the other forum, I forgot that people couldn't log in!

...But it really is as simple as I tried to make it sound in that earlier post. I bet if I put up a schem with no values someone would be able to make a better one!

earthtonesaudio · Post by **earthtonesaudio** » 25 Jun 2008, 03:23

Without further ado...

This is basically 2 appnotes connected together via a lowpass filter. First op-amp is a relaxation oscillator, outputs a square wave whose frequency depends on R and C. The other end of "C" is your foot, hand, whatever... So moving closer causes an increase of a few picofarads, and the freq drops down. C is set at 100pF which sets the lowest frequency, somewhere around 33-45kHz.

So the lowpass filter at the output of the oscillator serves two purposes... first, to bias the second op-amp's non-inverting input to the half supply voltage. Second, so that the farther away you move your body from the antenna, the smaller the signal going into the comparator becomes. The divider on the inverting input sets the minimum brightness of the LED. (Remember that it's going into the *inverting* input, so setting the wiper of the pot closer to ground causes the LED to be brighter-basically inverted of what you'd expect). The 470 ohm resistor could be a 10k variable resistor if you wanted to have a control for maximum brightness, but I didn't find the need for one.

The signal coming out of the second op-amp, when the voltage reference is adjusted just right, is a series of spikey pulses at max distance, which increases into a rounded-off square wave at minimum distance... It's almost pulse width modulation, but there's some frequency and amplitude modulation in there as well, so not a true PWM. The 10nF cap to ground at the end is to smooth out the pulses to get some rudimentary control voltage out of it.

It might be possible to use this as an expression input for a circuit that takes a CV input, but I don't have any circuits like that to test it on.

Some more details... 9.1V regulated DC used for power, 100 ohm/10uF supply filtering to the positive power input of the chip... I used 1/2 of a TL074 as the op-amp (and built a nice quacky bridged-T wah filter on the other half-with NO audible bleed through of the oscillator, even though they're ON THE SAME CHIP!)... and a 5mm super-bright LED that I salvaged from a keychain, so unfortunately I have no specs for it. Range is about 10cm with a 10cm*10cm square antenna made of aluminum foil, and it of course is not so sensitive when you stick a shoe in between your foot and the antenna, but I find that I like having a smaller range anyway cause it's a strain to have to lift your foot up too high.

Even though this is just a small collection of ideas lifted from appnotes, I still am quite proud of it, especially the fact that I was able to fit the control and wah circuitry on one chip. My initial goal was to build a Wah Probe work-alike using as few components as possible, and I'm pleased as punch that I've done it using 1/4 of the silicon... And NONE of it is a clone of any pedal that I've ever seen.

I would encourage any and all of you who are interested or curious to try this out, build some for yourselves and let me know how it works out. Not-for-profit use is totally fine with me, but I don't know if it infringes on any existing patents or not, so contact me first before selling any, and we can do a patent search together and then work out some licensing deal. (Here's hoping, anyway). Thanks, and good luck!

So what was I saying...? Oh yeah, without further ado:

Comments and criticisms always welcome.
-Alex

DWBH · Post by **DWBH** » 25 Jun 2008, 13:01

I want you to make me a baby.
You rock dude!