This week, I spent my time in two areas. First, I put together a bill of materials and ordered components for three more hardware nodes for the final demo, as well as for fuller system testing. Fortunately, most of the parts (all of them but the thermal sensor) came in the day after I ordered them, so I was able to assemble the rest of the hardware nodes, including soldering male-to-male headers on the battery converter boards and level shifter boards. The thermal sensor is pretty trivial to wire up, so I’m not worried about having to integrate those in a timely fashion.
The other thing I worked on was integrating the thermal sensor into the hardware design. I first went through hardware specs to figure out how it wires up, and realized that the pull-up resistors built into our level shifter board were sufficient enough in creating a voltage difference between the data and clock lines for the thermal sensor I2C and no additional resistors were needed. Most recently, I have spent time writing code to actually turn on and obtain readings from the thermal sensor. I am currently working on that now, and expect to have that up and running by class on Monday, which is within my planned schedule.
Next week, I plan on finishing up the thermal sensor integration, hopefully assembling the rest of the hardware nodes once the thermal sensors get in, and then I will run battery and power consumption tests to make sure our hardware is in spec with our use case requirements. If this is not the case, the risk mitigation plan is to increase the power efficiency of the code, either by using lower power libraries/using resting states more effectively or by limiting the uptime of the PIR sensor, which according to my calculations (and pending unit testing) should be the heaviest power consumer of any individual component.