Group Actions
Engagement
  • Author Author: ralphjy
  • Date Created: Date Created
  • Views 656 views
  • Likes 4 likes
  • Comments 2 comments
Related
Recommended

Solar SPEC - What's Next?

ralphjy

I learned a lot putting together the Solar SPEC unit for the Heat Is On Challenge.  I still have some work to do finishing the SPEC unit, but I've also thought of some other solar related projects to work on.

 

1) VI Curve tracer - this is actually a circuit that I started to build to characterize the reference cell in the SPEC unit, but I ran into some power supply issues and did not complete

2) Sun tracker/locator - this is something that would require an IPcamera with pan/tilt capability, a position reference, and a bunch of software

 

The curve tracer is just an implementation problem that I will fix with new hardware (larger power supply).

 

The sun tracker/locator is something that I'd like to do for fun.  I know that there are lots of apps available that will provide the sun's location, but it would be nice if I could log that info along with the irradiance.  That could just be a software exercise to extract data from the internet, but it would be neat to use a more direct approach.  The first approach would just to sweep the camera and process the frame brightness to locate the sun.  I don't think that the image sensor would be damaged if it only had direct exposure for a short time (maybe someone has tried this?).  There would have to be a position reference attached to the camera to provide an absolute reference (maybe an IMU?).  An alternate approach would be to attach a solar cell to the camera and use its output for feedback rather than the camera image.  This would require less processing but needs an additional board like an 8266.  The solar cell probably has less direction sensitivity also.

 

I'll see if I can find an image processing library that will just extract average brightness from the center of a frame capture without taking too much processing time (I'll try python PILLOW first).  Capturing frames on an IPcamera is straightforward, but most of the camera APIs don't allow you to specify a specific pan/tilt increment to move.  It might take some playing around, but I'd like to do it just for proof of concept.

 

I know that I could just build something, but I'd like to re-purpose an older P/T camera that I have and I could also use the camera to take snapshots of the sky conditions.

  • in reply to genebren

    No, camera isn't necessary.  Photo-diode is probably the most straightforward implementation and I could get better (easier to control) tracking if I built my own pan/tilt base.  But, I do like the way things are packaged in the camera and I would be able to use the camera normally also.  My wife has said that the neighbors probably wouldn't like the idea of a camera on the roof (particularly one that is scanning).  They have been polite enough not to complain about my drone so far...

     

    What would be really neat is if I could somehow get the camera to transmit the photo-diode and position data.  I have to admit I've never opened one up to see what processor they're using.  Probably too much work though I'm sure the capability is there.  It is nice that the WiFi and web server are built in.

     

    Hadn't thought of using the battery charger as the variable load.  Might be fun to try.

  • Cool plans.  Is a camera necessary for a sun tracker?  You could mount your reference cell to a pan/tilt  base and attempt to find the peak.  I have seen a lot of trackers built from photo-diodes, again looking for a peak signal.

     

    On the VI curve tracer, on though that I had was to use a li-ion battery charger with programmable charge current setting to act as a variable load.  That way you can command the current and measure the voltage.

     

    Good luck and have fun!

    Gene