Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Simple light tracking device

  1. Mar 23, 2008 #1
    Hi everyone!

    I wanted to share my final year electronics project on PIC-Microcontroller. It is a simple light tracking device. The project theme to demonstrate the working of a Sun tracker using a PIC micro controller.

    The sun tracker solar panel has a controller interface that can adjust the angle of the panel in the direction of the highest incident radiation. I want to demonstrate this using a simple photo sensor that will move in the direction of a moving light source (can be a simple torchlight moved by hand). The components I have used are:
    1)Photo sensor (phototransistors)
    2)PIC microcontroller - 16F877 (has a built-in A/D conversion module)
    3)unipolar stepper motor
    4)ULN2003 (Darlington Transistor Array -for driving the stepper motor)

    Photo-Transistors are used as sensors for generating an analog voltage proportional to intensity of light falling on it. The PIC16F877 has a built-in Analog-to-Digital-Converter (ADC). The analog voltages from the Photo-transistors are given as Inputs to the Analog Port pins of the PIC. The Photo-transistors are mounted on the Shaft of the Stepper Motor. The Stepper Motor requires that its stator windings should be energized in a programmed sequence to cause the motor to run in a given direction with a required speed. The software includes the positioning of collectors through stepper motor & data acquisition and processing in the Microcontroller. I have written the code in MPLAB IDE. Here are the pictures of the PCB and the main working project.

    This the PCB inside the device. The soldering of components took about 2 days.

    This is the wooden box inside which the motor and the circuit has been placed. The sensors are mounted on the ring. This took the longest time to complete.

    Here is the video of the device in operation. Let me warn, the quality isn't that good since it was captured using my cell phone cam. I recommend watching it under low resolution.

    Feel free to comment and provide suggestions for improvement. :smile:
    Last edited by a moderator: Sep 25, 2014
  2. jcsd
  3. Mar 23, 2008 #2
    I've seen a solar array about 1 yard square. It had a passive heat pipe that would orient the panel. If one side of the pipe were in shade, and the other in sunlight, the freon(or whatever) would move to reorient the panel normal.

    It's hard to beat ingenious simplicity.
  4. Apr 2, 2010 #3
    I guess I saw your project on youtube also :)
    I am doing a light detector also, can you tell me the chips you put on your board?

  5. Apr 2, 2010 #4
    There were only two "chips": The PIC microcontroller and the current driver (ULN2003). That's the advantage of using a higher configuration microcontroller. There are plenty to chose from.

    You get a built in ADC and other cool peripherals. The other important task apart from the circuit design is the program code.
  6. Apr 2, 2010 #5
    I saw this done yesterday at our first year design showcase. The guy used 2 photo resistors on either side of a panel sticking up from the middle and a simple circuit with a servo. Could track East/West. Granted, it was more of a solar cell idea.
  7. Apr 4, 2010 #6
    I did this exact same thing with a PIC microcontroller about a year ago. It works a little different and I never integrated it with a motor.

    http://picasaweb.google.com/lh/photo/hR0TkviIOQVlHE_UHL3NnA?authkey=Gv1sRgCJXKmJS94Iqofg&feat=directlink [Broken]

    This link is to a Picasa video I made about a year ago demonstrating it. I decided to output polar coordinates for it based on my specs and application. I also used photoresistors since I could easily obtain them. The nice thing about this one is that it's relative to north at all times so directionality can be adjusted pretty simply (most nowadays are anyways). Anyways, I figured I'd put this up to show how I did it.


    Justin Coulston
    Last edited by a moderator: May 4, 2017
  8. Apr 4, 2010 #7
    That was very cool, Justin! Thanks for sharing :smile:
  9. Apr 4, 2010 #8
  10. Apr 4, 2010 #9
  11. Apr 8, 2010 #10
    what is the algorithm for such a tracker which allows it to track the light?
  12. Apr 8, 2010 #11
    I think it's if one signal is at a higher voltage than the other, rotate the dish until they're equal
  13. Apr 8, 2010 #12
    but the sun is very high with respect to the difference between the position of the sensors so i think the difference of the intensity will not be high difference...
  14. Apr 11, 2010 #13
    Depends on what you want to track. If you want to track the sun, get a solar panel if you can afford it.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook