How much power can you get out of an LED?

Thread Starter

hp1729

Joined Nov 23, 2015
2,304
Yes, if you shine a light on an LED it generates a voltage. How do you determine how much power you can get out? If I put a resistor in parallel with it of 100K ohms the voltage drops to almost zero volts. 1M I get about 0.2 V. 10 M I get 1.2 volts. No resistor I get 1.4 Volts.
I am using an 8 mm diameter white LED.
With the 10 M I calculate about 0.12 microamps at 1.2 V. So am I looking at about 0.144 microwatts? Not very efficient.
If I use colored LEDs do I get color sensitive readings?
Has anyone else played with this idea?
 

Johann

Joined Nov 27, 2006
190
No harm in experimenting! That's what makes our hobby/profession interesting and that is how new things are often discovered. At least you know that the LED is in a ready package and might serve as an optic sensor in one of your projects!
 

Thread Starter

hp1729

Joined Nov 23, 2015
2,304
No harm in experimenting! That's what makes our hobby/profession interesting and that is how new things are often discovered. At least you know that the LED is in a ready package and might serve as an optic sensor in one of your projects!
Such was the objective, yes, thank you.
 

ErnieM

Joined Apr 24, 2011
8,045
If you need a color sensor take a peek at a company called ams (ams.com), formerly TAOS. They make several light and or color sensors that are quite useful.

The only drawback using their products is AFAIK they only come in surface mount packages. Welcome to 2016.
 

Papabravo

Joined Feb 24, 2006
13,924
Yes, if you shine a light on an LED it generates a voltage. How do you determine how much power you can get out? If I put a resistor in parallel with it of 100K ohms the voltage drops to almost zero volts. 1M I get about 0.2 V. 10 M I get 1.2 volts. No resistor I get 1.4 Volts.
I am using an 8 mm diameter white LED.
With the 10 M I calculate about 0.12 microamps at 1.2 V. So am I looking at about 0.144 microwatts? Not very efficient.
If I use colored LEDs do I get color sensitive readings?
Has anyone else played with this idea?
I guess it might be more efficient than trying to harvest useable energy from the RF background. There we are talking femtowatts. A femtowatt is 1e-15 watts.
 
Top