Arduino Refrigerator Monitor
The Idea
As a testament to my knowledge and understanding of circuits and programming in Arduino, I wanted to create something that could incorporate the sensors from my Arduino Starter Kit for a practical application. Thinking about some of the inconveniences around the house, I figured to start with the fridge and hone in on the reasoning for the freezer burn in my ice cream.
The Process
Starting with some preliminary research on the functionality and background of the BMP280 sensor, thermistor, and photocell I was able to grasp the limitations and general setup code of each component.
I then moved forward with wiring the breadboard to the allocated pins to read data and added 10k ohm resistors that were required based on my prior research for the photocell and thermistor.
After ensuring the stability of my breadboard configuration, I proceeded to the Arduino IDE and initiated the coding process. I focused on implementing the essential functions to convert voltage readings from individual sensors into corresponding measurements, including temperature in degrees Celsius, light intensity as a percentage, pressure in pascals (converted from bar), and humidity expressed in grams of water vapor per cubic meter of air. There was a lot of troubleshooting btw.
Once the data inputted into the serial monitor started to reveal realistic readings I threw it into the freezer to get behind the conditions that might have caused freezer burn on some of my food.
The Analysis:
The graph below, depicts the data taken from various sensors located on the breadboard of an ItsyBitsy microcontroller consisting of a thermistor, Bosch BMP pressure sensor, and a photocell. The change in curvature and path of the data is directly correlated to different environments the sensors were subjected to over a 125-second interval. The blue line represents light intensity, whereas the orange line, grey line, and the yellow line represent thermistor temperature, a BMP temperature reading, and pressure change respectively.
The controller started on the countertop and collected data for roughly 5 seconds before being lifted to be placed on the floor. This is denoted by the rapid spike from 13 to 17% in light intensity caused by the closer proximity to the lamp above. It is also worth noting that this response to being lifted occurred faster on the photocell than on the pressure sensor. This means that the photocell is more sensitive and quicker at detecting atmospheric changes compared to the pressure sensor. Once the controller made its way to the floor at the 9-second mark its apparent that the pressure rose substantially to a maximum of 7.1 Pa given the change in height of roughly 0.5m whereas the light intensity remained low and noticeable fluctuations. This would be attributed to a human error of me swaying over the sensor while momentarily blocking the light with my head. It is important to note that both temperature values have remained relatively constant at 22 °C.
The next environment the sensors were tested was on top of a fridge which measures 1.9m in height from the floor. Even in a fridge, the temperatures remained unchanged unlike the light intensity and pressure change which now measured at 25% and -9.4 Pa at equilibrium. It can be said that the change in light intensity is directly correlated to its closer proximity to the lamp whereas the pressure drop was due to an increase in elevation as expected.
The microcontroller was then placed inside the freezer located at the top of the fridge. Given that this freezer was at the top of the fridge compared to the bottom the change in pressure changed only slightly to now -13 Pa. This is physically strange as the pressure should have increased given that the height has dropped by approximately 0.25m from the top of the fridge. This means the pressure reading at the top was not at equilibrium and did not get an accurate reading given its slower response time. During this time, the light intensity plummeted rapidly to 6.53% along with the thermistor temperature at a slightly slower rate which is now situated at 10°C. The BMP Temp reading did drop but at a substantially slower rate. From this scenario, it’s clear I didn’t allow the sensors to each equilibrium as it would be expected after a certain time the temperature readings would both read a value below 0 °C.
After this test, the microcontroller was removed to equilibrate on the countertop. The pressure returned to the initial readings along with temperature whereas the light intensity did not. This can be attributed to me distancing myself away from the microcontroller this time to sense more light after noticing the first error from before.
The Result
Based on the data from the controller readings, it was observed that higher humidity levels increase the chances of freezer burn. This occurs when moisture in the air condenses on the food's surface, forming ice crystals that subsequently undergo sublimation, causing dehydration. Pressure changes, such as abrupt opening or closing of the freezer door, can lead to temperature fluctuations, condensation, and dehydration. Temperature changes also play a role in freezer burn as they can cause sublimation of ice crystals and affect the air's moisture-holding capacity.