I love air conditioning. My father owned his own sheet metal & HVAC construction company and I've always thought refrigeration was quite cool. I also need an A/C while living in NYC in the summer. I can not sleep without an A/C going most nights between May and October. But I really don't like taking the A/C out of the window and storing it over the winter.
Enter the cardboard box. I cut several pieces of cardboard to size so that the end product would fit snuggly in my window frame and cover my A/C. I wanted to see if it would sufficiently block the cold from coming in. Here's my research.
I connected an Arduino to three temperature sensors (DS18B20) via phone wire. I placed one sensor outside of my home, one between the AC and the cardboard insulator, and one in front of the cardboard insulator.
The Arduino was run from approximately Nov 1st, 2015 to approximately Jan 1st, 2016.
The software loaded onto the Arduino can be downloaded from this link.
Analysis and Results
The script that was used to analyze the data pulled from the Arduino can be downloaded here. This python script produced the image and two tables worth of data below.
The data can be downloaded too.
|Sensor ID||Mean||Std Dev|
|Comparison||t-Statistic||Two-Tailed p-Value||Average Difference|
|--79 v. --D7||262.15||0.0||1.77|
|--D7 v. --00||1337.87||0.0||6.53|
Welch's t-test was used to compute the t-statistic.
Welp... it's clear to me that this cardboard box does keep the apartment warmer than if I was to leave A/C in the window and forgo the insulator. The there is a clear difference in means between the sensor between the A/C and the cardboard insulator (285F3AFE040000D7) and the sensor in the apartment (28307DFD04000000). This box keeps my place warmer by six degrees Celsius on average.