The printed circuit board of ANAVI Info uHAT has a green solder mask and a gold surface finish. There are a few steps more to complete the manufacturing process. An EEPROM has to be flashed and soldered on each board. It will contain software description of the add-on board following Raspberry Pi Foundation’s HAT (Hardware Attached on Top) specifications. After that each board will go through a quality assurance, and finally each kit will be packaged in a recyclable cardboard box.
Low-volume manufacturing is not an easy task nowadays, especially during a global chip shortage. As usual we will keep you updated. Thank you again for supporting ANAVI Info uHAT!
I2C stands for Inter-Integrated Circuit, pronounced eye-squared-C, and alternatively known as IIC. It is a synchronous, multi-controller/multi-target (controller/target), packet switched, single-ended, serial communication bus. This protocol is suitable for devices wired at short distances, no more than 2-3m. We use I2C in pretty much all our open source products: Internet of Things, mechanical keyboards and Raspberry Pi HATs.
I2C was originally developed in 1982 by Philips. While that makes it 40 years old, it is still a very convenient and widely used bus. There are many I2C sensors and peripherals. It is in pretty much every smartphone, embedded electronics, microcontroller, personal computer and of course Raspberry Pi.
Actually, since the introduction of the famous 40-pin header in 2014, Raspberry Pi single board computers have not one but two I2C buses! We use them both on ANAVI Info uHAT and our other HATs. Th first I2C bus is on pins 3 and 5. On the ANAVI Info uHAT, it is used for the three I2C slots for sensors and the 4th dedicated slot for the mini OLED display.
The second I2C bus is on pins 27 and 28 of the Raspberry Pi and is reserved exclusively for attaching an ID EEPROM. The ID EEPROM contains a software description of the hardware so the operating system on your Raspberry Pi can automatically identify the add-on board.
The I2C bus consists of two signals: SDA (Serial Data) is a data signal, SCL (Serial Clock) is a clock signal. I2C modules also need power, so the dedicated I2C connectors on the ANAVI Info uHAT and our other open source hardware provide two additional pins for VCC and GND. Typically, the VCC for the I2C connectors on our add-on boards for Raspberry Pi are 3.3V.
The I2C bus drivers are “open drain”, which means they can only pull the corresponding signal line at low level. They cannot drive it high. To restore the signal to high when no device is asserting it low, a pull-up resistor has to be added to each signal line. For example, on the ANAVI Info uHAT, we have 4.7K pull-up resistors R4 and R5 connected to SDA and SCL.
Resistor selection varies depending on the devices attached to the bus. In some specific use cases, further adjustment of the resistance value might be required. For systems with lots of devices or longer wires, smaller resistors are better.
How to Enable I2C on Raspberry Pi OS
Raspberry Pi OS, previously known as Raspbian, is the default and recommended Linux distribution for all models and versions of the Raspberry Pi single board computer. By default, I2C is not enabled. There are several ways to enable it, but probably the easiest is using the command-line tool raspi-config to perform few basic commands:
Open a terminal or login remotely via SSH to your Raspberry Pi and type in the following command: sudo raspi-config
Each I2C device must have a unique address. The I2C reference design has a 7-bit address space, although rarely it might be used with a 10-bit extension. The 7-bit addresses range from 0 to 127 (0 to 0x7F hexadecimal). This is a limitation because it is not possible to have two I2C devices with the same address on the same I2C bus. For example, the I2C address on the mini OLED display included in all ANAVI Info uHAT kits is 0x3C. From the software side, this address is used in the example Python 3 script to access the display.
For Linux distributions, including Raspberry Pi OS, there is a package with a heterogeneous set of I2C tools called i2c-tools. To install it on Raspberry Pi OS, open a terminal and execute: sudo apt install -y i2c-tools. Once you have it installed, you can list attached I2C devices by their addresses with i2cdetect. For example, if the HTU21 temperature and humidity sensor module is attached to the Raspberry Pi, the output will be:
The event was organized by OPENNEXT. This is an organization that aims to bring together SMEs and makerspaces across the European Union (EU) to develop new hardware products based on open-source principles. The event brought together representatives of the Technical University Berlin, the Grenoble Institute of Technology, the University of Bath, the DDC – Danish Design Center, and many more organizations involved with open source software and hardware. Lukas Hartmann, the creator of the open source DIY laptop MNT Reform, was also a panel speaker at the event.
Thanks to early backers ANAVI Info uHAT was successfully funded and hit its first stretch goal in a just a couple of days. So we’ll be adding some awesome KiCad and ANAVI Technology stickers. KiCad is the free and open source software we used to design this and other Anavi printed circuit boards.
As a small open source project, ANAVI Info uHAT relies on the community of passionate open source makers. We are near our second stretch goal of $1,000. If we hit it, we will make more video tutorials for all supported sensors.
There is still more than a month until the end of the crowdfunding campaign and we hope more people will jump in and order ANAVI Info uHAT!
AI-Thinker is a leading supplier of IoT wireless products and solutions, including antennas, modules and RF lab service. Since 2017 we have been using AI-Thinker ESP-12E modules with ESP8266 in many of our open source hardware products, including ANAVI Fume Extractor, ANAVI Gas Detector, ANAVI Thermometer, ANAVI Light Controller and ANAVI Miracle Controller.
Now, despite the hard times of global chip shortage, AI-Thinker keeps manufacturing and delivering high-quality modules on time. As a trusted supplier through the years, recently we have purchased from AI Thinker enough ESP-12E modules do fulfill the demand and keep making our open source hardware products.
Furthermore we stocked ESP-C3-12F modules with Espressif Systems ESP32-C3 Wi-Fi microcontroller based on the open standard instruction set architecture (ISA) RISC-V. ESP-C3-12F are pin to pin compatible with ESP-12E.
AI-Thinker is based in Shenzhen, China. The company was founded 10 years ago, in 2012. They also provide LoRaWAN, NB-IoT, Bluetooth and other Wi-Fi modules.
Earlier in January all ANAVI Gardening uHAT kits were delivered to the Crowd Supply warehouse and soon after that Crowd Supply team sent them to their final destination: our valuable and trusting crowdfunding backers! Thank you again for the support.
ANAVI Gardening uHAT is a versatile development board, so please follow the instructions below for safe use:
ANAVI Gardening uHAT should only be connected to a compatible Raspberry Pi with 40-pin header.
Do not expose it to water or moisture, and do not place it on a conductive surface whilst in operation.
Do not expose it to heat from any source; it is designed for reliable operation at normal room temperatures.
Take care while handling the board to avoid mechanical or electrical damage to the printed circuit board and connectors.
Avoid handling ANAVI Gardening uHAT while it is powered on. Only handle by the edges to minimize the risk of electrostatic discharge damage.
In the meantime, there is a work in progress going on the user’s manual which is available at GitHub. We will soon update it. As usual, GitHub pull requests with improvements and fixes to the documentation or the source code examples are always welcome.
We have good news regarding ANAVI Gardening uHAT! All boards have been received almost fully assembled from the local factory and we have sourced all required peripherals.
Recently, we received the last batch of assembled printed circuit boards from the local factory. As you can see on the photo, only the EEPROM is missing. We will flash and solder it in-house.
We have also received additional peripherals, which will be included in ANAVI Gardening uHAT Starter, Advanced, and Developer kits. On the photo you can see the big packages with analog capacitive soil moisture sensor. Each kit will contain a couple of capacitive soil moisture sensors.
The recyclable cardboard boxes for our eco-friendly packaging have also already been delivered. Another local company here in Plovdiv, Bulgaria will print all stickers for us. We expect them next week.
Our crowdfunding campaign recently ended very successfully but you can still order our open source Gardening uHAT for your Raspberry Pi and be among the very first owners!
Raspberry Pi Pico is a tiny and fast development board by the Raspberry Pi Foundation built using the brand new RP2040 32-bit dual ARM Cortex-M0+ microcontroller. The major advantage of Raspberry Pi Pico is the affordable price as it is available for about $4 (without taxes and shipping).
In this video tutorial you will learn how to get started with MicroPython using the open source Thonny IDE on Raspberry Pi Pico. Thonny runs on Mac, Windows and Linux distributions, in the video it is used on Ubuntu. The video includes Pico unboxing, MicroPython installation guide, blinking LED example, MicroPython REPL demo and conclusions.
The video tutorial was sponsored by PCBway which provide high-quality prototyping services. On the photo you can see prototypes of green printed circuit boards with white silkscreen following Raspberry Pi specifications for micro Hardware Attached on Top (uHAT).
ANAVI Gardening uHAT is a low-cost, open source Raspberry Pi add-on board that helps you develop smart solutions for monitoring and growing plants.
ANAVI Gardening uHAT supports multiple sensors for soil moisture, temperature, humidity, barometric pressure, and light. Getting started is easy: just plug it into a Raspberry Pi with your bare hands and follow the instructions in the user manual. No soldering is necessary, and no tools are required.
Raspberry Pi is a famous series of small single-board computers (SBCs) developed in the United Kingdom by the Raspberry Pi Foundation in cooperation with Broadcom. This is a step by step tutorial for using Raspberry Pi and capacitive soil moisture sensor with Microchip MCP3002 analog-to-digital converter (ADC) and a Python script for detecting the soil moisture in percentage.
Capacitive Soil Moisture Sensor
Capacitive Soil Moisture Sensor v1.2 and v2.0 measures the volumetric content of water inside the soil and retrieves the moisture level by capacitive sensing rather than resistive sensing like other sensors. The benefit of using a capacitive soil moisture sensor is the lack of corrosion and longer lifespan.
Wiring
Unlike Raspberry Pi Pico, the recently released microcontroller, all versions and models of the Raspberry Pi single-board computers do not include an analog-to-digital converter (ADC). This tutorial explains how to use Microchip MCP3002 with Raspberry Pi.
Microchip MCP3002 is a 10-bit resolution, dual channel ADC with SPI hardware bus. It can be connected to any Raspberry Pi single board computer version and model, including Raspberry Pi 4 and Raspberry Pi 0. However, this tutorial is not for Raspberry Pi Pico microcontroller. For more details about the wiring of Microchip MCP3002 a Raspberry Pi single-board computer have a look at my previous tutorial.
Alternatively, the easier option without a breadboard an a bunch of cables, is to use a dedicated Raspberry Pi add-on board with built-in ADC. Using the free and open source tool KiCad we designed ANAVI Gardening uHAT exactly for this purpose. It has dedicated pins for connecting a couple of capacitive soil moisture sensors. The prototype has been created thanks to PCBWay. This is a lead-free prototype printed circuit board with 2 layers, green solder mask and white silkscreen. PCBway offers a huge variety of colors and even flexible PCB.
Software
Flash Raspberry Pi OS, the official Debian based GNU Linux distribution by the Raspberry Pi, on microSD card and boot it. On the Raspberry Pi, open a terminal and using the raspi-config tool enable SPI as shown in the video. Reboot the Raspberry Pi.
Python3 script for reading data from a couple of capacitive soil moisture sensors through Microchip MCP3002 ADC is available at the anavi-examples repository in GitHub. The script relies on popular Python libraries spidev and RPi.GPIO. Open a terminal and run the following commands to clone anavi-examples and run the script:
git clone https://github.com/AnaviTechnology/anavi-examples.git
cd anavi-examples/anavi-gardening-uhat/soil-moistore-sensors/python/
python3 soil-moistore-sensors.py