I am trying to design a solar tracker project for my POE class. I have found online a lot of information about hacking a servo and replace the pot for photo sensors. I want to make a bigger system using the same principles, and without programming. Has anyone ever made a simple solar tracker? How did you accomplish it? Would it be possible to run a motor off of a Jag and connect the photo sensors up to it directly without any programming?
Thanks for your input!
What are the requirements of your solar tracker? What does it need to do? Will it be used for aiming something like a solar array or mirror?
What is your rationale for a no-programming approach? Is it a time limitation? You could easily pick Arduino up in a few days using something like the Arduino Cookbook and learn a lot about programming in the process.
As for the JAG, you would need to generate a PWM signal using something other than a microcontroller. Here is a starting point: http://electronics.stackexchange.com/questions/32533/what-is-the-simplest-way-to-generate-pwm-without-a-microcontroller. If this looks daunting, I’d suggest going back to the Arduino approach.
Sorry I should have been more specific. I am a teacher I am trying to develop an activity for my POE class to use a solar tracker to move a set of Harbor Freight Solar panels to follow the sun. I teach digital electronics as well, and we use Arduino (great system) for that class. I have use an Arduino and PS3 controller to control robots before so that programming would be simple. However, I am trying to avoid programming because I do not have the time in the year to allocate teaching my students to program the system. If I have to I could program the microcontroller myself and have them follow some simple instructions to connect their system to the microcontroller but I want to avoid that if possible.
What’s your time frame for this? This sounds like an awesome activity. Maybe we (growingSTEMS™) can help. Shoot me a PM and we can discuss this in more detail over email.
According to this NI-hosted document, the Jaguars still require an external signal whether using CAN or PWM. (Is it possible to hack the firmware? I don’t know–but even if you could, you’re still programming.)
Seems like you could program up a few Trinket boards to handle the necessary processing; they support up to three analog inputs and two PWM outputs (or two and three; one pin goes both ways). A couple extra wires won’t kill them (just say “MAGIC HAPPENS HERE” in the diagram), and eight bucks won’t kill you if they somehow kill the board.
OK, so no CPU. Actually makes it harder.
Super-simple: A motor that runs at 1 RPD (Rev Per Day). Or at least a system that spins the whole assembly one revolution per day. Runs continuously. Once in a while you should re-set it to point at the sun, because nothing is perfect… (The issue is how to manage wiring to the fixed earth.)
Simple: Use a relay to run a very slow motor. Make an “eye” (see below) to detect the sun’s position. Use an Op-Amp to detect when the sun has passed the “eye” and move the motor a little until it points at the sun again. Feedback loop!
Eye: Just a sensor at the bottom of a tube. More than one sensor, actually: One when aimed “right at” the sun, one when aimed “sun has moved”. Maybe the ‘right at’ sensor is ‘a little further’ so the system moves less often. Like once every 30 minutes. Oh, and a timer of some type to reset to start after, say, 6 hours of darkness. Simple but effective.
More complex: Several sun sensors for an Eye, logic for movement even if there’s no sun, reset timers, and so on. But now we would be better off with software control…
Jaguar? Way overkill, adds way too much complexity. If you don’t want to use a relay and a very slow motor, then get an H-Bridge chip and use a variable duty cycle to control motor speed. Jaguars need too much love and affection to work well from a simple circuit.
hello friend sorry for old thread reply but have you completed the solar tracker project? It seems very interesting so please share some more information
A while ago (for my 8th grade science fair project) I built a solar tracker with mechanical feedback.
It had two bimetallic coils (like you’d find in thermostats, that rotate when their temperature changes), mounted on a shaft and wound in the opposite directions. When the temperature of both coils changed, they just pushed against each other and didn’t move the shaft. However, when one was heated, it would move the shaft in one direction (as long as the other wasn’t heated too).
Then, I constructed shades that would only heat one of the coils if the sun was on one side of the tracker. Basically, it created a mechanical feedback loop, so that if the sun was on one side, one coil would get heated and the tracker would move to the side with the sun until the shades equally heated both coils, and the tracker reached equilibrium.
In the end, it worked pretty well, I saw 14% improvement over a flat mounted panel. All in all, it was a pretty cool project, and definitely different from what most kids built.
Hey! I know this reponse is really late, but this is another 100% programming free solar tracker from the TV show the colony.