Sun tracker via GPS

Discussion in 'The Projects Forum' started by Essadaoui_Red, Mar 8, 2012.

  1. Essadaoui_Red

    Thread Starter New Member

    Feb 17, 2012
    we try to design and build a sun tracker carrying a plate photvoltaîque, piloted by GPS,

    GPS must locate the position of the sun and gives an order to move towards this position throughout the day,
    but we really do not know where to start :confused:

    all your suggestions and ideas will be welcome,

    thank you in advance for your helps,
  2. ErnieM

    AAC Fanatic!

    Apr 24, 2011
    As a GPS is a GLOBAL POSITIONING SYSTEM, and not a SOLAR POSITIONING SYSTEM it really isn't going to be much help here, is it?

    Not that it is entirely useless. Your longitude can be used to convert from Coordinated Universal Time (UTC) to local time. (In local time the sun is due south at noon.) Your latitude can be used in combination with a sun chart or formula to compute the angle of the sun above the horizon.

    Unless your mirror is moving around your position is a constant.

    The tool you really seek is called a CLOCK.
  3. spinnaker

    AAC Fanatic!

    Oct 29, 2009
    Correct but this assumes the device is not moving. If it were moving then a GPS could be utilized.

    I really haven't looked at one of these charts in detail but I would imagine it could be fairly large if you wanted to cover an extended area. Then perhaps a formula could be used to obtain optimum positioning.
  4. Markd77

    Senior Member

    Sep 7, 2009
    GPS won't tell you the direction that you are pointing in. You need a compass for that, or just optical sensors that find the brightest point. You can get the time from GPS, but if you aren't moving hundreds of miles, you can keep track of time more easily.
  5. Essadaoui_Red

    Thread Starter New Member

    Feb 17, 2012
    thank you friends,

    let's me explain you more,
    we try to build an universal sun tracker, means that it work in any where (africa,USA, UE...), the system should localise it's position via GPS, and calculate the position of the sun...and orients th PV to the sun...
    the problem is in the program that calculate the position in real time ??? and how can we transform these informations to control signal ?

    thank you
  6. davebee

    Well-Known Member

    Oct 22, 2008
    If you want to know where the sun is, why not use a light-sensitive detector?
  7. Essadaoui_Red

    Thread Starter New Member

    Feb 17, 2012
    Thank you for your reply. You are right, but the use of light-sensitive detector is widly spread, and we want to explore the possiblity of traking the sun by GPS.
  8. spinnaker

    AAC Fanatic!

    Oct 29, 2009

    Actually it can if you are moving. Some GPS devices will provide velocity east and velocity north from that head can be derived.

    Code ( (Unknown Language)):
    3. double heading = atan2(velEast, velNorth);     // Returns the angle, in radians, between -pi and pi.
    4.     heading *= 180.0 / M_PI;                      // Convert from radians to degrees.  Now heading is in the range [-180, 180]
    5.     if (heading < 0.0)
    6.         heading += 360.0;                         // Get rid of negative headings.
  9. spinnaker

    AAC Fanatic!

    Oct 29, 2009
    You don't need the complexity of a GPS.

    You could use a calculator such as this

    Then down load the data to your tracker for the users location. It would be up to the user to know his location and the direction to setup the tracker and initialize it. They could do this with Google Earth, Google maps and many other utilities. They could also use a GPS if they have one. They only other thing they would need is a cheap compass.

    They would enter their position on the PC using your positioning software and this would down load a positioning table to your tracker. You could even use an SD chip to make it easier.

    The user would then initialize the tracker at it's start position, set the date and time and the tracker would position itself based on that locations table.

    All you would need is an MCU with RTC, some stepper motors and maybe limit switches and the associated hardware.

    There is no need a GPS other than making setup slightly easier for the user.
  10. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    So what stops the clock drifting?

    A GPS gives the benefit of second-accurate timekeeping, timekeeping even during GPS blackout (storms etc) so it's a RTC. It also gives LAT and LON.

    Provided he doesn't mind the cost of a GPS module it's a good solution.

    Otherwise he needs an RTC anyway, and to manually enter the LAT and LON, then use a light sensor etc and some tricky long term averaging code to correct clock drift.

    And if it has one light sensor, with all its problems of cleaning and fading, why not have two light sensors and just make it a light tracker? They give more power output than clockwork tracking anyway as they track the brightest region of the sky.
  11. spinnaker

    AAC Fanatic!

    Oct 29, 2009
    Form what I have read, accuracy is not all that critical for positioning. Blackouts could be handled via battery backup.

    Though I admit the GPS would be a nice feature for the reasons you stated. If you really wanted to be cool then you would add an electronic compass, then it could be self initializing.
  12. Georacer


    Nov 25, 2009
    I too believe that light-sensitive elements are used in solar tracking in solar farms.
  13. ErnieM

    AAC Fanatic!

    Apr 24, 2011
    If you wish a truly inexpensive device with world wide applicability you could replace the compass and the GPS with a stick and a ruler.

    1) Place stick in ground at dawn.

    2) Monitor shadow. Shadow will begin to move and get shorter, then stay about the same length, then begin to get longer again.

    3) Find the point in #2 where the shadow is shortest. Mark that point.

    4) Measure height of stick and length of shortest shadow and compute the ratio.

    You get three important data items from this simple experiment:

    A) The line from the point marked in #3 to the stick is pointing true north.

    B) The ratio found in #4 is the latitude.

    C) The local time when the shadow is shortest is 12:00 noon.

    Thus all the clock needs for setup is a simple reset button. If anyone is monitoring the equipment the "shortest shadow" measurement may be repeated every several months to keep time errors to a minimum.