Solar Power on the Raspberry Pi

Here are some more results from today from the RasPiConnect screen attached to the Raspberry Pi and Arduino on Project Curacao in the Caribbean.

The first RasPiConnect screen below shows the status of the Raspberry Pi Solar Power system.  The screen yesterday showed the status of the Arduino Battery Watchdog system.

This screen was generated with RasPiConnect on the Raspberry Pi located in Project Curacao in the Caribbean. The Charts are generated using MatPlotLib and all the software is written in Python. The Raspberry Pi utilizes a MySQL database to store all the data incoming from sensors connected directly to the Pi as well as the incoming data from the Arduino Battery Watchdog which has many sensors attached to the Arduino I/O pins and the I2C bus. The block diagram from the Raspberry Pi power subsystem is below:

The most interesting part of this screen today is that you can tell by the solar panel voltage on the chart that the Raspberry Pi batteries (separate from the Arudino batteries) have been fully charged by the sun 5 out of the last 8 days. Very good news. Note the current power coming from the battery is -142ma. This data comes from an IN219 high side current sensor with the library fixed to allow negative currents.

Main Project Curacao RasPiConnect Screen

The below screen shows two interesting things. First of all, the log file shows the Arduino turning on the Solar Power system to the Pi. Why does it need to do that? It does this because it turns on the Wind Power system at night (no sun!) to trickle charge the battery. We feel a little sorry for the Arduino because it is a futile gesture because the Wind Power turbine was destroyed back in March. We are putting a new turbine up in September with a few improvements and a vibration sensor. One thing we can do with a vibration sensor is if the turbine is running very fast and vibrating a lot during the day, we can turn the Wind Power system on which will slow the Wind Turbine down. We don’t know how much it will slow down yet, but that is something we will measure. The second interesting thing is that the outside humidity sensor (AMS2315) is showing 1% humidity. No way. It should be about 66% or so. The inside humidity sensor (DS311) is showing 56%. Why is the AMS2315 way off? We think it is because when we tilted the box back, the sensor is exposed directly to the sun hence baking the sensor. The outside temperate (34.3C) still seems correct.

More results next week.  Part one of the latest results was published here.