Raspberry Pi Learning Resources

Setting up the Raspberry Pi Weather Station software manually.

You don't need any prior knowledge to set up the Weather Station. There are several steps, but the benefit of setting up manually is that you'll learn about the workings of the sensors and the station as you do it. You'll also be introduced to the command line interface, the text editor nano and the MariaDB database. It's also a great introduction to Linux.

Manual installation

  1. Start with a fresh install of the latest version of Raspbian. You can either use the full Desktop version, or the slimmer 'lite' version.

  2. When booting for the first time, you will be presented with the desktop (or a login prompt if you're using the 'lite' version).

  3. If you're using the Desktop version, from the Menu button in the top left-hand corner choose Preferences > Raspberry Pi Configuration. Otherwise log in and type
sudo raspi-config
  1. We recommend that you change your password.

  2. In the Interfaces tab, enable I2C:

  3. A reboot dialogue will appear. Select Yes.

Setting up the real-time clock

We'll be doing most of the work from the command line. If you're using the Desktop version, open a terminal window using the icon on the menu bar or by pressing ctrl+alt+t.

You'll now be at a prompt:

[email protected]: ~ $

You can type the commands which follow into this prompt.

First, you'll need to download the necessary files:

cd ~ && git clone https://github.com/raspberrypi/weather-station

We've included an install script to set up the real-time clock automatically. You can run this file or, alternatively, follow the instructions below to set up the RTC manually. We recommend using the install script!

RTC setup

First, you want to make sure you have all the latest updates for your Raspberry Pi:

sudo apt-get update && sudo apt-get upgrade

You now need to make some changes to a config file to allow the Raspberry Pi to use the real-time clock:

sudo nano /boot/config.txt

Add the following lines to the bottom of the file:

dtoverlay=w1-gpio
dtoverlay=pcf8523-rtc

Press Ctrl+O then Enter to save, and Ctrl+X to quit nano.

Now set the required modules to load automatically on boot:

sudo nano /etc/modules

Add the following lines to the bottom of the file:

i2c-dev
w1-therm

Press Ctrl+O then Enter to save, and Ctrl+X to quit nano.

For the next steps, we need the Weather Station HAT to be connected to the Raspberry Pi:

sudo halt

Reboot for the changes to take effect:

sudo reboot

Check that the real-time clock (RTC) appears in /dev:

ls /dev/rtc*

You should see something like /dev/rtc0.

Initialise the RTC with the correct time

Use the date command to check that the current system time is correct. If it's correct, then you can set the RTC time from the system clock with the following command:

sudo hwclock -w

If not, you can set the RTC time manually using the command below (you'll need to change the --date parameter, as this example will set the date to the 1st of January 2014 at midnight):

sudo hwclock --set --date="yyyy-mm-dd hh:mm:ss" --utc

For example:

sudo hwclock --set --date="2015-08-24 18:32:00" --utc

Then set the system clock from the RTC time:

sudo hwclock -s

Now you need to enable setting the system clock automatically at boot time. First, edit the rule in /lib/udev/:

sudo nano /lib/udev/hwclock-set

Find the lines at the bottom that read:

if [ yes = "$BADYEAR" ] ; then
    /sbin/hwclock --rtc=$dev --systz --badyear
else
    /sbin/hwclock --rtc=$dev --systz
fi

Change the --systz options to --hctosys so that they read:

if [ yes = "$BADYEAR" ] ; then
    /sbin/hwclock --rtc=$dev --hctosys --badyear
else
    /sbin/hwclock --rtc=$dev --hctosys
fi

Press Ctrl+O then Enter to save, and Ctrl+X to quit nano.

Remove the fake hardware clock package

Use the following commands to remove the fake hardware clock package:

sudo update-rc.d fake-hwclock remove
sudo apt-get remove fake-hwclock -y

Testing the sensors

Install the necessary software packages

Power up your Raspberry Pi and log in.

At the command line, type the following:

sudo apt-get install i2c-tools python-smbus telnet -y

Test that the I2C devices are online and working:

sudo i2cdetect -y 1

You should see output similar to this:

     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f
00:          -- -- -- -- -- -- -- -- -- -- -- -- --
10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
40: 40 -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
60: -- -- -- -- -- -- -- -- UU 69 6a -- -- -- -- --
70: -- -- -- -- -- -- -- 77                         
  • 40 = HTU21D, the humidity and temperature sensor.
  • 77 = BMP180, the barometric pressure sensor.
  • 68 = PCF8523, the real-time clock. It will show as UU because it's reserved by the driver.
  • 69 = MCP3427, the analogue-to-digital converter on the main board.
  • 6a = MCP3427, the analogue-to-digital converter on the snap-off AIR board.

Note: 40, 77 and 6a will only show if you have connected the AIR board to the main board.

Now that the sensors are working, we need a database to store the data they produce.

Database setup

Now you'll set up your Weather Station to automatically log the collected weather data. The data is stored on the Pi's SD card using a database system called MariaDB. Once your station is successfully logging data locally, you'll also be able to upload that data to a central Oracle Apex database to share it with others.

Install the necessary software packages

At the command line, type the following:

sudo apt-get install -y mariadb-server mariadb-client libmariadbclient-dev
sudo apt-get install -y apache2 php5 libapache2-mod-php5 php-mysql
sudo pip3 install mysqlclient

If you make a mistake, use the cursor UP arrow to go back to previous lines for editing.

Please note that this will take some time.

Create a local database within MariaDB

Enter the following:

sudo mysql

You will now be at the MariaDB prompt MariaDB [(none)]>. First, create a local database account for the Pi user and assign the necessary privileges. You should also choose a password for this account.

create user pi IDENTIFIED by 'password';
grant all privileges on *.* to 'pi' with grant option;

Then, create the database:

CREATE DATABASE weather;

You should now see Query OK, 1 row affected (0.00 sec).

Switch to that database:

USE weather;

You should see Database changed, and your prompt should now be MariaDB [(weather)]>.

If MariaDB doesn't do anything when it should, you've probably forgotten the final ;. Just type it in when prompted and press Enter.

Create a table to store the weather data

Type the code below, taking note of the following tips:

  • Don't forget the commas at the end of the row
  • Use the cursor UP arrow to copy and edit a previous line, as many are similar
  • Type the code carefully and exactly as written, otherwise things will break later
  • Use CAPS LOCK!
  CREATE TABLE WEATHER_MEASUREMENT(
    ID BIGINT NOT NULL AUTO_INCREMENT,
    REMOTE_ID BIGINT,
    AMBIENT_TEMPERATURE DECIMAL(6,2) NOT NULL,
    GROUND_TEMPERATURE DECIMAL(6,2) NOT NULL,
    AIR_QUALITY DECIMAL(6,2) NOT NULL,
    AIR_PRESSURE DECIMAL(6,2) NOT NULL,
    HUMIDITY DECIMAL(6,2) NOT NULL,
    WIND_DIRECTION DECIMAL(6,2) NULL,
    WIND_SPEED DECIMAL(6,2) NOT NULL,
    WIND_GUST_SPEED DECIMAL(6,2) NOT NULL,
    RAINFALL DECIMAL (6,2) NOT NULL,
    CREATED TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,
    PRIMARY KEY ( ID )
  );

You should now see Query OK, 0 rows affected (0.05 sec).

Press Ctrl+D or type exit to quit MariaDB.

Set up the sensor software

Begin by downloading the data logging code. You can skip this step if you have set up the real-time clock.

cd ~
git clone https://github.com/raspberrypi/weather-station.git

This will create a new folder in the home directory called weather-station.

Start the Weather Station daemon and test it

A daemon is a process that runs in the background. To start the daemon we need for the Weather Station, use the following command:

sudo ~/weather-station/interrupt_daemon.py start

You should see something like PID: 2345 (your number will be different).

A continually running process is required to monitor the rain gauge and the anemometer. These are reed switch sensors, and the code uses interrupt detection. These interrupts can occur at any time, as opposed to the timed measurements of the other sensors. You can use the telnet program to test or monitor the daemon, with the following command:

telnet localhost 49501

You should see something like this:

Trying 127.0.0.1...
Connected to localhost.
Escape character is '^]'.
OK

The following text commands can be used:

  • RAIN: displays rainfall in ml
  • WIND: displays average wind speed in kph
  • GUST: displays wind gust speed in kph
  • RESET: resets the rain gauge and anemometer interrupt counts to zero
  • BYE: quits

Use the BYE command to quit.

Set the Weather Station daemon to automatically start at boot

Use the following command to automate the daemon:

sudo nano /etc/rc.local

Insert the following lines before exit 0 at the bottom of the file:

echo "Starting Weather Station daemon..."

/home/pi/weather-station/interrupt_daemon.py start

Press Ctrl+O then Enter to save, and Ctrl+X to quit nano.

Update the MySQL credentials file

You'll need to use the password for the MySQL root user that you chose during installation. If you are not in the weather-station folder, type:

cd ~/weather-station

then:

nano credentials.mysql

Change the password field to the password you chose during installation of MySQL. The double quotes " enclosing the values are important, so take care not to remove them by mistake.

Press Ctrl+O then Enter to save, and Ctrl+X to quit nano.

Automate updating of the database

The main entry points for the code are log_all_sensors.py and upload_to_oracle.py. These will be called by a scheduling tool called cron to take measurements automatically. The measurements will be saved in the local MySQL database, and they will also be uploaded to the Oracle Apex database online if you registered.

You should enable cron to start taking measurements automatically. This is also known as data logging mode:

crontab < crontab.save

Your Weather Station is now live and recording data at timed intervals.

You can disable data logging mode at any time by removing the crontab with the command below:

crontab -r

To enable data logging mode again, use the command below:

crontab < ~/weather-station/crontab.save

Please note that you should not have data logging mode enabled while you're working through the lessons in the scheme of work.

Manually trigger a measurement

You can manually cause a measurement to be taken at any time with the following command:

sudo ~/weather-station/log_all_sensors.py

Don't worry if you see Warning: Data truncated for column X at row 1 – this is expected.

View the data in the database

Enter the following command:

mysql -u root -p

Enter the password (the default for the disk image installation is tiger). Then switch to the weather database:

USE weather;

Run a select query to return the contents of the WEATHER_MEASUREMENT table:

SELECT * FROM WEATHER_MEASUREMENT;

After a lot of measurements have been recorded, it will be sensible to use the SQL where clause to only select records that were created after a specific date and time:

SELECT * FROM WEATHER_MEASUREMENT WHERE CREATED > '2014-01-01 12:00:00';

Press Ctrl+D or type exit to quit MySQL.

Upload your data to the Oracle Apex database

At this stage, you have a Weather Station which reads its sensors and stores the data at regular intervals in a database on the SD card. But what if the SD card gets corrupted? How do you back up your data? And how do you share it with the rest of the world?

Oracle has set up a central database to allow all schools in the Weather Station project to upload their data. It is safe there, and you can download it in various formats, share it, and even create graphs and reports. Here's how to do it.

Register your school

You'll need to register your school and add your Weather Station. Come back here when you have your Weather Station passcode.

Update credential files with your Weather Station details

Add the Weather Station name and password to the local Oracle credentials file with the commands below. This allows the code that uploads to Oracle to add it to the correct Weather Station.

cd ~/weather-station

nano credentials.oracle.template

Replace the name and key parameters with the Weather Station Name and Passcode of the Weather Station above. The double quotes " enclosing these values in this file are important, so take care not to remove them by mistake. The Weather Station name must match exactly and is case-sensitive.

Press Ctrl+O then Enter to save, and Ctrl+X to quit nano.

Rename the Oracle credentials template file to enable it:

mv credentials.oracle.template credentials.oracle

Checking that data is received

Manually trigger an upload with the following command:

sudo ~/weather-station/upload_to_oracle.py

Log into your school's Oracle Apex account and go to 'Weather Measurements'. You should see the station readings:

You can download your data in various formats and also make charts using the menu:

What next?