I’ve done some writeups of how I made the Concrete Cricket design (there’s a bit about that here and here), but I’ve been asked about gutting the MP3 players, so here’s that! It’s stupid-simple — you just wire up two sides of the Play button, run wires to either side of a reed relay, and control the relay through an Arduino. I know some people (notably Michael Chladil) have also soldered wires to the indicator LED and watched when it went low or high through an analog read, but I’m not nearly that fancy. I was hoping to have a nearly idiot-proof, mostly self-running system, and this works well for that.

The story, in pictures, follows.

MP3 shell:

After taking the guts out:

Wiring up the switch:

After that, I wired up connections to the audio out and battery in (though you could just as easily keep things the way they came):

Reed relay (the 250 ohm, 5v variety from your local friendly Radio Shack) in use:

Using an Arduino and the aforementioned relay to turn MP3 player on and off:

On a breadboard:

Perfboard everything all nice:

And everything all wired up nicely (and all wires covered in heat shrink):

Oh, and because I’ve been asked about this as well, a stupid-simple LM386 Op-Amp circuit to go along with this:

It’s painfully cheap on Electronic Goldmine, but you can find ‘em at Radio Shack if you’re in a bind.

========================

/*
Concrete Crickets Arduino Code
Michael J. Dory
www.doryexmachina.com
Written 04.11.07, revised 10.30.07
(thanks to code from Tom Igoe and Rob Faludi)
*/

int RangePotPin = 5;    // Analog input pin that the rangefinder is attached to
int RangePotValue = 0;   // value read from the pot

int MP3Pin = 12;    // the pin that the relay controlling the MP3 is attached to

int blinkyPin = 13; // the pin that the reference LED is attached to

int onOffPin = 11; //the pin that the speaker attach/kill switch relay is on.

void setup() {
// initialize serial communications at 9600 bps:
Serial.begin(9600);

// set pins for output
pinMode(blinkyPin, OUTPUT);
pinMode(onOffPin, OUTPUT);
pinMode(MP3Pin, OUTPUT);

// initialize both pins at low
digitalWrite(MP3Pin, LOW);
digitalWrite(blinkyPin, LOW);

delay(100);

// turn MP3 Player on, blink pin 13 for reference
digitalWrite(MP3Pin, HIGH);
digitalWrite(blinkyPin, HIGH);
delay(5000);

// take both pins low again
digitalWrite(MP3Pin, LOW);
digitalWrite(blinkyPin, LOW);

delay(100);

// blink LED three times quickly to show mp3 player is on
digitalWrite(blinkyPin, HIGH);
delay(100);
digitalWrite(blinkyPin, LOW);
delay(100);
digitalWrite(blinkyPin, HIGH);
delay(100);
digitalWrite(blinkyPin, LOW);
delay(100);
digitalWrite(blinkyPin, HIGH);
delay(100);
digitalWrite(blinkyPin, LOW);
delay(100);

// start playing the MP3
delay (2000);
digitalWrite(MP3Pin, HIGH);
digitalWrite(blinkyPin, HIGH);
delay(500);
digitalWrite(MP3Pin, LOW);
digitalWrite(blinkyPin, LOW);

// blink three times to show that the MP3 player is playing and setup is complete
digitalWrite(blinkyPin, HIGH);
delay(300);
digitalWrite(blinkyPin, LOW);
delay(300);
digitalWrite(blinkyPin, HIGH);
delay(300);
digitalWrite(blinkyPin, LOW);
delay(300);
digitalWrite(blinkyPin, HIGH);
delay(300);
digitalWrite(blinkyPin, LOW);
delay(300);

// end setup
}

void loop() {
RangePotValue = analogRead(RangePotPin);   // read the pot value
Serial.println(RangePotValue);             // print the rangefinder value back to the debugger pane
delay(10);                                 // wait 10 milliseconds before the next loop

if(RangePotValue <= 125) {                // if the rangefinder detects something in range
digitalWrite(blinkyPin, HIGH);          // turn the light on
digitalWrite(onOffPin, LOW);            // turn the speakers off
Serial.println(”detected”);             // print “detected”

delay(100);
}
else {
digitalWrite(blinkyPin, LOW);          // if the rangefinder doesn’t detect anything
digitalWrite(onOffPin, HIGH);          // leave the speakers connected
Serial.println(”not detected”);        // print “not detected”
}

}

…with a bit of a nod to Mouna, RSG and yours truly, amongst others. I’m highly flattered, and a bit confused, but really glad the festival’s getting the attention it deserves. It was an amazing time — by far the best festival I’ve been involved in, and likely the best I’ve even been to in recent memory. A full write-up is due, but in the meantime:

[read]

How pimp, seriously?

Thanks a ton to Tom Jenkins and Rita Chipalkatty for their amazing eyes and design help on this — it looks GREAT.

Made this bugger yesterday. It’s all heat-shrinked and electrical taped, so it’s MUCH cleaner than its predecessors — moving along well!

IMG 0044, by doryexmachina.

Looking much cleaner, too. I’m trying to heatshrink everything for this one, so we’ll see how it goes, but it’s looking much cleaner so far.

This one has a few advancements over the one I finished last semester. First and foremost, it’s much cleaner and better laid out, and secondly it has pots to control the sensitivity of the distance sensor and the volume of the speaker output. Those right there make a big difference in terms of stability and deployment.

The board design and layout is worlds cleaner than the last one, as evidenced above, and has so far worked 100% faithfully (also an advancement of the prior prototype). So far it’s been living in a coffee cup, and it looks pretty sneaky:

It’s been good out in the wild as well. It’s easy to go install, because it looks like a coffee cup as long as nobody sees the sensor, and if you keep your thumb over the sensor, it won’t chirp anyway, so it’s easy to be sneaky. Louder than I thought it might be, too — a really solid version so far, I think.

And closer-up, indoors footage:

More on the other two (or maybe one, at this point — I’m not sure I’ll be able to fit five in the space I have at Conflux) as progress is made!

This is just a breadboard prototype — it’s still just a pile of wires sitting in a takeout container, so visually it’s nowhere near there yet, but you can hear the song just fine from half a block away. We’ll see how it goes as I start putting them into their proper containers, but for now you can get a pretty clear idea of where I’m going with this.

I took a really good look at my first boards for the concept I presented back in May, and spent some time getting a more power-efficient, louder, generally less buggy design together. I’m actually really happy with how it’s coming along so far, and I had a chance to street-test one such design this week. That test involved just the mp3 player, op-amp circuit (now a bit louder), two speakers (as opposed to the first prototype’s single speaker), and a 9v battery in a takeout container. I’m classy!

It was surprisingly loud. I could hear it just fine within 10 or 15 feet, which is the ideal range I’d had in mind, and half a block away, with car s and people passing, I could still make out tones. That’s better than I’d hoped for — it may attract people from further away, which is great, but it’s still not so quiet that people right near it will miss it, and not so loud that it will get unwanted attention from the authorities or bother people living nearby.

I’m also sure it worked because an old couple called the cops on me. I’d attribute that less to the project and more to the crazy dude running around with a beeping box of exposed wires, taking pictures and dashing back and forth to hear how loud it was (me, of course). However, police attention = successful graffiti!

By all means, if you’re in the area, please do come on out to Conflux. I’ll hype that more as time goes on, but so yous knows, it’s September 13 – 16 in Williamsburg — tons more info found on their site.