Strange LED Sensor like behavior  (Video atta

Hi,

Using a 74HC595 shift register to control 24 LEDs, I attached the clock pin to pin 13 (NONE PWM) updated my code and noticed that the LED were actually reacting to my movement! Whenever I moved my hand over the LED grid they would light up, when I remove my hand they would go back to being off.

From the code found on (I changed it some to do Row Column scanning to drive more LEDS! :smiley: )

//Pin connected to ST_CP of 74HC595
int latchPin = 8;
//Pin connected to SH_CP of 74HC595
int clockPin = 12;
////Pin connected to DS of 74HC595
int dataPin = 11;
//int dataPin = 13; <---strange shit happens with datapin set to 13

I've uploaded some Video of the strangness here:
http://207.58.160.243/~ff/arduino/led/

On ledz_0002.wmv I change camera angle at about 30 seconds to show I am not touching any wires to trigger this behavior.

I think it might having something to do with this

http://projects.dimension-x.net/technology-and-projects/ledsensors/
LEDs As Sensors

Any clue anyone?
thanks! :smiley:
-ff

On the Arduino NG's, the LED on pin 13 sometimes causes it to behave a bit strangely. Can you move the clock to another pin?

mellis,

I did move it back to the normal pin 11 and things work fine BUT i'm still curious to know if/why the LEDS are reacting to my hand and body movement.

-ff

Nice! I like that a lot! :smiley:

Do you have the arduino on a surface such as a cushion ? if so that might be why it reacts to your body's motion else it might be poor wiring.

also: do you have the 1uf capicator shown in the schematic in your circuit? this will reduce noise from the circuit; maybe the capicitance or magnetic field of your body is affecting the circuit...????

This thing with LEDs being sensitive to incoming light was interesting to me. I did some simple experiments today where I hooked up a yellow LED to my digital multimeter. When sunlight was shining on it, I got about 1.5V of potential measured. When I tried to measure the current, the meter always said 0, so it was definitely less than 0.1 mA. The voltage was much less when not in direct sunlight: about 0.1 V was a typical value.

So the bottom line is, if you want to use an LED as a light sensor, it's a good idea to feed it into a high-impedance input, perhaps even an op-amp configured as a negative feedback voltage follower.

  • Don