Temperature logger and display
This project uses the raspberry pi along with other electronics to measure the temperature, plot it in a website and show it on-demand on a LED alphanumeric display.
Hardware used
- Raspberry Pi 3 Model B
- Temperature sensor DS18B20
- Light Dependent Resistor (LDR)
- Capacitor (1 μF)
- 0.54" Alphanumeric display and backpack from Adafruit
- Wires and resistors for setting them up.
Setup
The setup involves connecting the electronics, enable the appropriate interfaces on the raspberry pi and running the python scripts.
The wiring of the circuits and connections to the raspberry pi is not covered
here.
Guidelines can be found in:
- CamJam Edukit 2: Sensors for the temperature sensor and LDR circuits
- Adafruit for the LED backpack.
The interfaces used are the 1-wire for the temperature sensor and the I2C for the alphanumeric display. Guidelines on how to enable them are also found in the links given above.
The last step is running two python scripts with information given in the following sections. Because of dependency on some libraries, python 2 should be used.
Logger
The start logging run python temp_logger.py. This creates a file in the logs
directory with the date time and temperature value.
Alternatively, you can use the script temp_log.py which will log a single
measurement of temperature in the same file as the above logger. It's purpose is
to be used in a cronjob.
To plot temperatures use the gnuplot script plot_temp.gp which will output the
files in the webpage directory.
To access the website online you should make a symbolic link from the location where
websites are published in your system.
e.g. /var/www/html in Debian
You can also create a cron job to run the gnuplot script and plot the new values
of temperature.
Display with hand wave
The code in temp_wave uses the LDR to detect a hand wave over the circuitry
and display the temperature.
To initiate, run python temp_wave.py.
The LDR is an analog device and to convert its response to digital, we use a circuit that charges a capacitor. We measure the time to charge the capacitor which depends on the amount of light.
This method produces an unstable signal that makes it a challenge to find an effective way of detecting the hand wave. Alternatively, an ADC can be used to convert the analogue signal or a motion sensor which would provide more reliable methods of detecting a hand motion.
Nevertheless, this project uses a few simple criteria to detect the hand wave within the current setup. It calculates the rolling average of the time-to-charge and the cumulative difference of each value with the previous. The temperature display is triggered when their ratio increases above a threshold and if the increase has happened smoothly.