Light Tracker Project: Seeking Algorithm Help

[hisham.i]

Hi every body...
I have a light tracker project, so i will use a photo transistor and pic 16f877 micro controller..
but until now i don't have a complete algorithm and efficient one, of how such a tracker should work.
please send me a detailed algorithm of how such a tracker should work..
thank you

 

[vk6kro]

One way of doing this is to have two phototransistors next to each other on a turntable rotated by a motor, and having a metal or other opaque shield between them.
You would probably need diffusers in front of each phototransistor. This would be plastic as you see in front of fluorescent lights, or it could be blocks of paraffin wax.

Then you would wire the the phototransistors in a Wheatsone Bridge and take the output to a comparator.
The output of this could switch a relay to turn on a motor to turn the turntable (carrying the light sensors) in the direction of the brighter phototransistor.

Instead of the comparator, you could use two A to D converters in a Micro.

To avoid the tracking becoming too active, you could accept an area of similar brightness as being "good enough" and let the motor rest if this happened.
Or you could just take a reading every 5 seconds or so and ignore the time in between.
The motor would last longer if you didn't have it hunting continuously.

 

[hisham.i]

you are right..but the problem i faced with this method is the following:
when the light is behind the tracker and both sensors figure the same intensity of light what should it do? and how can the tracker know that it should rotate the next side?

 

[hisham.i]

you are right..but the problem i faced with this method is the following:
when the light is behind the tracker and both sensors figure the same intensity of light what should it do? and how can the tracker know that it should rotate the next side?

 

[Jcouls29]

We kind of talked about this in this thread. Look for my post and you'll see a video link to a light tracker I made. It consists of 5 photoresistors (LDRs) that keep up with where the light it.

https://www.physicsforums.com/showthread.php?t=223686

-------------
Justin Coulston
TheModernEngineer.blogspot.com

 

[vk6kro]

If you took the two outputs to a micro, you could check that at least one of them is getting a minimum light level. You will know how bright the light is supposed to be.

If both of them are comparatively in the dark, rotate the motor backwards and forwards until you find the light source or it is turned on.

You can't rotate the motor continuously in one direction or the wires will get twisted, so you need sensors to detect when the end of the travel has been reached.