|
LEDs are essentially photodiodes. This concept was introduced way back in the 70s but has been forgotten by most of us.
The principle behind using leds as photodiodes is that:
A led in reverse bias can be modeled as a capacitor in parallel with a current source. The current sourced by this current source is directly proportional to the light incident upon the led. If the capacitor is charged, the current sourced by the current source will aim to discharge it. Hence, higher the light intensity, larger the current sourced and faster the capacitor discharges. If we can determine the time taken for the capacitor to discharge, we can get a fair measure of the intensity of light incident upon the led.
Algorithm to monitor the light intensity incident upon the led:
- Connect a led between 2 pins of a micro-controller and reverse bias it. I.e. the pin connected to the anode is kept at logic 0 and the pin connected to the cathode is kept at logic 1. So the capacitor is charged to 5V.
- Now remove the reverse bias and, set the pin connected to the cathode as an input pin. What happens now is that the current sourced by the current source discharges the capacitor in time inversely proportional to the light intensity incident upon the led. This time can be easily inferred by the time it takes the pin connected to the cathode to reach logic 0!
This algorithm gives us a reasonably accurate idea about the light intensity incident on the led under observation.
Fig: Using Leds for photo-sensing.
References: 1) http://www.merl.com/papers/docs/TR2003-35.pdf
2)" LED used as a light sensor " – By D.V Gadre and Sheetal Vashisht |
