Project: Light Sensor and Charging Circuit

Discussion in 'The Completed Projects Collection' started by nomurphy, Feb 19, 2008.

  1. nomurphy

    Thread Starter AAC Fanatic!

    Aug 8, 2005
    567
    12
    I had to repair my outdoor light by reverse engineering the circuit board.

    It failed because water/moisture got onto the top of the PCB and rusted component leads, the bottom was just fine. I had to clean it in alcohol and then replace a couple badly rusted components. I then used a parylene conformal coating on the board to help prevent further issues. I silicone sealed around the solar panel and detector to help keep out the water.

    It uses a couple AA rechargeable batteries, a solar panel for charging during daylight, and a light detector to detemine if the LED should be on (night) or if it should be charging batteries (day).

    I find it's kind of an interesting circuit, so I'm posting it here.
     
    • Lamp.pdf
      File size:
      32.5 KB
      Views:
      1,264
  2. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    My solar garden lights have a continuous white LED and others have a fading RGB LED.

    They use a single Ni-Cad battery cell, one transistor to sense light to turn it on and off and two transistors as a stepup voltage generator.
     
  3. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Both are interesting circuits. I fiddled around a bit with both of them with the idea of generating a boost circuit for running an LED from a single cell that was between 1 and 1.5v, but neither experiments were completely successful. Audioguru's circuit required more than 1.5v, and nomurphy's circuit was extremely dependent upon the battery's charge level.

    After experimenting and a number of searches, I stumbled upon Al Dutcher of Al Labs' design for a single cell driver that will light any LED, in EDN's "Design Ideas".

    Al's design is really quite simple and elegant; two NPN transistors (2N3904 are used, but you can substitute others), five resistors, a 1N4148 diode (you can use other silicon diodes), an inductor (this controls the frequency of the circuit; a 330uH is a decent compromise for 50kHz operation, but you can use many other values) and the LED of your choice.

    The LED may not light at it's maximum brightness, but it WILL light up if it's still good. Additionally, the intensity of the LED will appear very linear regardless of the battery's charge condition.

    Attached:
    1) BoostAnyLED.PNG - An image of a simulation I ran on Linear Technology's SwitcherCad
    2) BoostAnyLED.zip - A ZIP file containing the .sch file for SwitcherCad
     
  4. nomurphy

    Thread Starter AAC Fanatic!

    Aug 8, 2005
    567
    12
    First of all, I think you mistated ownership, because the one I posted requires two AA batteries.

    Second, it's not my circuit. I simply posted the inner workings of a Chinese manufactured outdoor light purchased at a local Mega-Lo-Mart type hardware store.

    During the day it charges, and at night it lights. If the days are short and the nights are long (such as in winter), then the batteries may discharge and the LED will be off before sunrise. I've never bothered testing it to see how well it charges vs. amount of light to the solar panel.
     
  5. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    Most of my solar garden lights use a single 1.2V Ni-Cad battery cell.
    Most use a fading RGB or a white LED that require about 3.5VDC.

    I have 24 solar garden lights (many were for free from my Electrical utility Company)
    and only one still lights at sunrise now after charging for 6 hours in full sunlight and after shining for 12 hours all night. Some shine for only 2 hours at night now.
     
  6. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    I didn't mean to imply that you were the originator of the circuit, as you stated that you had simply traced out and captured the schematic of the circuit.

    What I should have stated is that the circuit that you posted did not meet my design requirements of running via a single 1.1v-1.5v cell, although that does not mean that your effort was unappreciated. ;)
     
    Dead Crow likes this.
  7. tmmizwan

    New Member

    Jan 12, 2009
    5
    0
    I NEED TO KNOW HOW TO MAKE THE BRIGHTNESS OF LIGHT VARY WITH SURROUNDING BRIGHTNESS. WHEN THE SURROUNDING NOT ENOUGH BRIGHTNESS, SO THE LIGHT WILL PRODUCE MORE BRIGHTNESS. ACTUALLY THIS CIRCUIT USE SOLAR PANEL. AND MY PROJECT IS SOLAR STREET LIGHT.
     
  8. Solar1

    Member

    Jan 12, 2009
    15
    0
    Last edited: Jan 12, 2009
  9. Gadgetmaul

    New Member

    Aug 18, 2013
    1
    0
    Actually I started a company doing the exact same thing. It took 10 thousand dollars of research and development to come up with the solution of a self sustainable street light.
    I did the research and development. I will tell you just a bit of private information. You need a higher wattage panel in order to charge up in less than an hour in full sun Or in a couple hours on rain and overcast. Final product we have sells for 4000.00 each with orders of 10,000 per month so the solution is expensive but the light is grid free and last over 25 years without maintenance. :).
    They also produced lumens over 20,000 from the emmiter heat free. Lux is way better than sodium a 60 lux on a 120 degree area.
     
Loading...