I built a solar tracking system with light sensors for a solar cooker. It worked beautifully on a cloudless day but sometimes got confused when there were some clouds. Some weeks later I realized that my glorious invention was trying to find something we know the position of at any time during the day. Considering the time and money spent and how smart I had felt for building that monstrosity made me feel very dumb afterwards.
A non-optical sensor based solution would require other sensors to calculate the relative position of the solar cooker oriented to Earth as well as calculate the elevation. Your solution is turnkey that you just slap down and it does everything needed without external input beyond solar radiance.
Also, a whole bunch of your approach has many other applications with only very minor tweaks. I like what you did.
Yeah, it’s like making a car self-driving without GPS. it’s not dependent on external input that way. If you can make it work by only using cameras and lidar, that’s fantastic work and definitely deserves credit.
If you replaced it with something that requires connectivity, then your first approach may have been a better design if even if it didn’t work out quite right the first time.
If it were able to retain coordinate tables, then after a sunny day calibration it could use that as fallback. And make limited micro adjustments and table updates as-needed.
I built a solar tracking system with light sensors for a solar cooker. It worked beautifully on a cloudless day but sometimes got confused when there were some clouds. Some weeks later I realized that my glorious invention was trying to find something we know the position of at any time during the day. Considering the time and money spent and how smart I had felt for building that monstrosity made me feel very dumb afterwards.
A non-optical sensor based solution would require other sensors to calculate the relative position of the solar cooker oriented to Earth as well as calculate the elevation. Your solution is turnkey that you just slap down and it does everything needed without external input beyond solar radiance.
Also, a whole bunch of your approach has many other applications with only very minor tweaks. I like what you did.
Yeah, it’s like making a car self-driving without GPS. it’s not dependent on external input that way. If you can make it work by only using cameras and lidar, that’s fantastic work and definitely deserves credit.
If you replaced it with something that requires connectivity, then your first approach may have been a better design if even if it didn’t work out quite right the first time.
If it were able to retain coordinate tables, then after a sunny day calibration it could use that as fallback. And make limited micro adjustments and table updates as-needed.