Updated: Apr 1, 2019
This week, I came back from Spring Break to find that there were issues with the water quality monitoring system that I developed for the SenPi project. Whenever the water quality sensor was removed from the water at any point during its operation, it would go haywire and the program would stop working until the Raspberry Pi was rebooted.
After spending a few hours attempting to fix the problem, it looks like this is an issue with the sensor's native software. Back when I was first brought onto this project, we had a lot of trouble trying to access the sensor -- it wouldn't respond to any of the messages we sent it, and the computer refused to believe that it even existed -- until Dr. Blanchard pulled out a beaker full of water and dropped it in.
The manufacturers of the sensor admitted in an email that their sensor does not function in any capacity when it is not fully submerged. I'm currently looking for a workaround; although I've been able to handle this error and disconnect the sensor if it starts freaking out due to being dry, I haven't been able to figure out how to reconnect to the sensor after doing so. Currently, after a dryness event, the sensor connects until I send it a message, at which time it disconnects itself. I might have to resort to telling the pi to reboot itself if the sensor encounters a dryness error, although this could be problematic if the sensor gets lifted out of the water during field operations -- especially if we plan to attach two sensors to the same Raspberry Pi.
In other news, the current remote-power setup for the SenPi module, after a 3-hour field test, had used less than a quarter of the total charge on its power pack. I believe a burnout test has been planned for this week using a Garridae-class vessel.