Solar Tracker circuit.

Discussion in 'The Projects Forum' started by Burnit0017, Feb 15, 2010.

  1. Burnit0017

    Thread Starter Member

    Feb 15, 2010
    101
    3
    Greetings, I am new to the forum and I am working on a solar tracker. The circuit uses two photo resistors in a voltage divider to change the state of a comparator which will turn on a nand gate 10k oscillator and then using a voltage doubler to turn on a H fet bidirectional motor control. The circuit will only work if the light sensors are placed back to back and then mounted vertically on the tracker.
    The links shows a project for mounting the sensors using the technique .
    http://www.youtube.com/watch?v=iH49Xd_NV3A
    If both sides of the input sensor are turn on the H fet configuration will fail. Not a good thing. If anyone has a solution to resolve this problem please let me know. Feedback welcome.
     
  2. russpatterson

    Member

    Feb 1, 2010
    351
    16
    Looks like a cool project. So it has 4 sensors so it can turn and adjust the tilt of the panel?

    Here's one idea to solve your problem. Put the 4 photo sensor into a microcontroller A/D then just write a few lines of code to determine which FET's get turned on when. You would eliminate a couple dozen parts from your board and it should also be cheaper.
     
  3. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Interesting, indeed. ;)

    You really don't need the voltage doubler circuits for the lower MOSFETs. As long as you can get Vgs to 10v, they'll be turned fully ON.

    You might consider using a 4093 quad NAND instead of the 4011, as the Schmitt-trigger inputs allows you to create an astable multivibrator using just one gate, a capacitor and a resistor for feedback - and it can't get "stuck" like the 4011 circuit you used could.

    Rather than have the C-D sensors detect the difference in light between right and left, have them sense the difference between front and back intensity.

    If the back is brighter than the front, the array should be rotated until the front is brighter than the back. From that point forward, adjust until the right/left sensors have equal brightness.

    There is currently no hysteresis feedback for right/left; this can result in excessive "seeking" or oscillations.
     
  4. Burnit0017

    Thread Starter Member

    Feb 15, 2010
    101
    3

    Greetings, I have not checked the price of a micro controller in a while. The parts I am using are all left overs from other projects. The software solution will make the project easier. Thanks and best regards.
     
  5. ELECTRONERD

    Senior Member

    May 26, 2009
    1,146
    16
    You'll have to program your μC in the C language or ASM (assembly). A suitable programming device would be the PICKit2. This can be ordered from Digi-Key for $35, assuming you don't include the dev board.

    Austin
     
  6. russpatterson

    Member

    Feb 1, 2010
    351
    16
    The cost of the MCU itself would be ~ $1.50 - $3.00. Here's a link to a project that is similar to yours. I haven't read through the whole thing but his circuit looks similar. http://www.redrok.com/electron.htm#led3x
     
  7. Burnit0017

    Thread Starter Member

    Feb 15, 2010
    101
    3
    Greetings, the circuit showen is for one motor. The project will use two motors. One to track during the day and the other to track during the seasons. Because of the back to back mounting of the light sensors both input sets will turn off when pointed directly at the sun. I plan to use a small PV panel to power the circuits and motors. I am trying to mill the pcb's using a cnc router. The project in the video is not my it is just to illustrate how the sensors will be mounted. Thanks for the input and best regards.
     
  8. Burnit0017

    Thread Starter Member

    Feb 15, 2010
    101
    3
    Greetings, I added an Inverter and AND gate to the input of the B and D oscillators to prevent Q3 and Q4 from turning on at the same time. I am hoping this will prevent damage of the circuit caused by possible auto head lights or the increased lumens from lighting strikes. Feedback welcome.
     
    Last edited: Feb 16, 2010
  9. Burnit0017

    Thread Starter Member

    Feb 15, 2010
    101
    3
    Greeting, the project is moving forward. I have started to mill the PCB's using a cnc router. I will be fun to see if all the pad locations can be soldered. CMOS chips are used for all the logic and oscillating circuits and a small solar panel will be used as a power supply for both circuits and motors.
     
  10. joulian

    Active Member

    Aug 27, 2009
    53
    0
    have u completed this project.
    Can you post a pic of it and any updated schematics?
     
  11. Burnit0017

    Thread Starter Member

    Feb 15, 2010
    101
    3
    Greetings, I have spent the last few weeks gathering all the parts. I am currently learning how to use TinyCad and FreePCB so I get a Grebber file to generate the G-code needed to mill the circuit boards. Thank you for keeping pointed in the right direction and I will post some pictures as the project develops.
    Best Regards
     
  12. Burnit0017

    Thread Starter Member

    Feb 15, 2010
    101
    3
    I had to order a new router for the cnc mill to solve a spindle run out problem. The pic's show new router installed.
     
Loading...