temperature_sensor_esp32_mcp9808

temperature_sensor_esp32_mcp9808

This is the repository for data and analysis for my blog post Quantifying the Thermal Benefits of Replacement of my House's Front Door.

Highlights of key directories or files of interest are

• esp32_wifi - the build directory for the ESP-IDF code project for the ESP32 microcontroller-based temperature sensors.
  • README.md - a rudimentary README file for the microcontroller.
  • main.c - key source file for microcontroller.
• temperature_logger.py - Python script to query ESP32 temperature sensors and log data.
• horemheb - Python package (named after an Egyptian pharoah) of tools for analysis and plots produced in this blog.
• 100_400_plot_temperature.ipynb - plot raw temperature data.
• 200_200_analyze_temperature_over_params.ipynb - Analysis of data, application of Newton's Law of Cooling.
• 300_200_optimize_cooling_de.ipynb - Analysis of dynamical cooling model.
• 100_400_analyze_LVK_weather_data.ipynb - For reference, load and look at the temperature and humidity measurements from the LVK airport over the duration in question here. I didn't apply this data to the analysis, but it is available. It is possible that there is a correlation between cooling and wind speed or direction.

---

1/22/2025

An ESP32S3 board reads temperature information from an MCP9808 temperature sensor via I2C.

---

The set up of this project follows that of labrador_classifier.

Some basic setup:

micromamba env create -p "./.venv_temp_sense" "python=3.13"
micromamba activate "./.venv_temp_sense"
micromamba install "uv"
uv pip install zeroconf requests # for web access and mDNS hostname lookup
uv pip install rich
uv pip install ipykernel ipywidgets # for notebooks
uv pip install pandas matplotlib
uv pip install scipy

---

From prompts, I created a 3.13 python environment with idf_tools.py install-python-env.

Then, as per the "Getting started with the ESP-IDF" Evernote, the following commands created a VScode session that did create a build/ directory:

. ~/stash/esp/esp-idf/export.sh
export IDF_XTENSA_GCC="$(which xtensa-esp32-elf-g++)"  # <- not sure if this is crucial, but it works
cd ~/Work/Animata/Esp_projects/temperature_sensor_esp32_mcp9808/esp32
cursor .

---

Set up package

Created a package to manage the loading and fitting of this data. Named package after Egyptian pharoah Horemheb:

cd horemheb
uv pip install -e .

---

March 20, 2025

Downloaded humidity and wind data over this period of time for LVK from the National Weather Service. Click Advanced Options and in the pop-up window click Permanent Chart to select the item to be plotted and Gather Historical Data to select a date range. Then, from the resulting plot, click the bars in the upper right and the click Download CSV from the pull-down menu.