For the solar assignment, I worked with Rob Faludi to augment a project that we already have been working on for quite a while. Our decided task was to use solar cells both as a light sensor and as a power source.
The project is called Botanicalls and it is a system where plants are able to call people on the telephone to express their needs. The existing physical computing setup include a soil moisture sensing circuit, a microcontroller, and a ZigBee radio. There were two issues that we wanted to address:
1. In terms of the of transference of information, these plants are entirely wireless. In terms of power, they are tethered. Because we did not want to deal with the possibility of batteries dying, we have been using wall power for our circuit. This is both inconvenient and a conflict of interest. You should not have to unplug your house plant when you want to move it to a new location. One of the goals of this project is to help people develop successful relationships with their plants. We are specifically targeting people who have trouble keeping up with plant maintenance, so adding the task of changing batteries on top of taking care of the plant does not make any sense. The focus should be on maintaining the plant, not the technology. In addition, this is intended as a consciousness-raising project. We want draw people’s attention back to the natural world and to encourage them to think about the consequences of their actions. We do not want to do this with yet another device that demands the consumption of fossil fuels. In terms of options for alternative energy sources, using solar power for circuitry for a plant makes sense – a plant already has an inherent need for light, so why not use that light to power the circuit? Or in standard SAT format:
photosynthesis:plants::photovoltaics:Botanicalls
2. We’ve started off with soil moisture as the initial need that we are sensing & responding to for the plant. But we’ve been itching to move on to light and actually had some of the code already written for it. We were originally intending to just use photocells, but when brainstorming about this assignment, Rob and I thought it might be a good idea to be efficient in terms of our components and use the solar cells as light sensors in addition to a source of power.
Our existing circuit is powered by a regulated 3.3V, which means we need a source voltage of at least 4.5V. We decided to use 3.6V 50mA solar panels. Here we can see two of them in series held in a window with a fair amount of light generate 5.44V and 104mA:
Here you can see the circuit being powered directly off of 2 sets of 2 panels in series in parallel:
We decided to use this setup to trickle charge a setup of 4AA batteries. Here you can see that the voltage supplied to the circuit when the batteries are not attached is 6.52V:
When the load of the battery pack is added, it drops to 5.24V:
And the analog sensor value, which is measured directly off the panels before the battery charging circuit is 5.86V:
Our only concern is that the sensor readings will be affected by the charge level of the battery. We are going to setup a datalogging scenario so that we can monitor the sensor values and battery charge over time to see if this is actually an issue.
Here you can see the overall setup:
In this image we are using the breadboard to swap out different resistors to see which would give us the ideal range for our analog sensor values.
Our circuit diagrams can be found below – both the diagram for the overall system and a detailed diagram of the part of the circuit that pertains to this project.
For the real life version, the microcontroller does value averaging, so it responds to the amount of light that it is getting over several days. But we also created a version made for demonstration purposes – when the solar panel is covered, a phone call is made requesting that the plant be moved into the light. When the solar panel is exposed to light again, a call is made to say thank you.
