This repository consists of the code and hardware design for my entry into the Silicon Labs Low Power Contest 2015.
This entry is a component of a larger project called Sensaura which provides hardware and software solutions to implement low power sensors communicating over a wireless network protocol and publish values to a MQTT message bus.
This project consists of a prototype implementation of a SensNode processor module using a EFM32 Zero Gecko starter kit and a simple sensor backpack to test the power consumption and functionality of the design.
The goal is to develop a small, cheap, wireless sensor board with a well defined API, footprint and network protocol that can be used as the basis for a range of customised sensors.
The sensors are intended to be run from batteries (the ideal design should operate for 6 months on a single AA cell) or other disconnected power sources which allows for deployment in areas without a fixed power supply available.
Sensor data is gathered through an adapter connected to an IP connected host, formatted as JSON and published to an MQTT server for consumption. This design allows maximum flexibility without requiring the sensors themselves to be connected to the internet.
To achieve the desired goal of operating on a single AA cell for a period of 6 months power usage must be carefully carefully controlled both in the design of the sensor and peripheral circuitry as well as by utilising the low power features of the EFM32 CPU.
The processor module is anticipated to spend the majority of it's time in sleep mode, only waking up at periodic intervals to read sensors and transmit results. In the case of the EFM32 this will set the processor to energy mode 3 (EM3) which allows it to wake up on RTC or pin change events at the discretion of the firmware.
As the design is generic and intended for use with other ARM processors as well not all low power features of the EFM32 (such as LeSENSE) can be utilitised due to difficulty exposing those features in a generic way.
Due to time constraints the implementation of the prototype is not yet complete, some functionality is untested or only partially implemented.
The hardware components consist of the Silicon Labs EFM32ZG-STK3200 Development Kit, a breakout board matching the final board footprint with a NRF24L01+ wireless module fitted and jumper leads to connect to the IO pins on the development board and a simple sensor backpack for testing.
The sensor backpack consists of the following components:
- Si7020 Temperature/Humidity Sensor
- TSL2561T Light Sensor
- A soil moisture sensor made of two metallic probes configured for resistive measurement.
Two of the components use an I2C interface for communication, the moisture sensor requires a digital output to control the current flow for measurement and an analog input to take the reading.
Both boards are single sided PCBs that were milled with a small CNC machine. One of the goals of the project is to ensure that processor and sensor boards can be made using resources generally available to the hobbiest - milling or etching single side boards that can be hand soldered is a core requirement.
To allow data being transmitted on the NRF24L01 network to be captured and used on a PC a simple Arduino based adapter was also constructed. This simply uses an Arduino Nano to report any packets received on the NRF24L01 network over a serial port connection and was assembled on a breadboard.
As the SensNode is intended to be processor agnostic I use a simple API that provides generic access to GPIO, analog and timing functions. For the purposes of the prototype I implemented a shim layer that implements the SensNode API using the EMLIB Library. The actual sensor firmware was then implemented with calls to the SensNode API.
Implementing a sensor like this is relatively simple in concept - the main application loop simply reads the values of the various sensors and transmits them over the NRF24L01 network and then puts the processor to sleep for a specified period of time before repeating the process.
For this particular sensor there are no unpredictable events (such as motion sensors) so the wake up process is purely driven by the real time clock. The values measured in this case are also fairly slow to change (there would not be a significant change over a duration measured in minutes) so it can spend the majority of it's time in sleep mode.
As I mentioned earlier the project is not yet complete. Time constraints and some unexpected difficulty with the NRF24L01 interface prohibited a full implementation from being completed in time for this submission. As a result any power profile readings taken at this stage would not be of much use.
On the positive side however the project has exposed some improvements that can be made to the overall design;
- The main issue I had with the NRF24L01 module was the stability of the signals on the SPI bus. Most of these are caused by the use of jumper wires to connect it to the development board - a common problem when working with external peripherals.
- Due to a misreading of the datasheet for the CPU I assumed that the LEUART could only be driven by the 32.768kHz clock and only supported 9600 baud transmission. As a result I used the two USART channels to implement I2C and serial communications and provided a bit banged SPI interface. This further complicated the interface to the NRF24L01.
- In a design such as this the power consumption of the CPU is only a part of the total picture - power usage by external peripherals needs to be managed correctly as well. One enhancement I intend to make to the design is to provide the processor sleep state available on a GPIO pin - this will allow external circuitry to shut down if needed when the processor goes into sleep mode which will further reduce the total power consumption.
Development on this prototype will continue with the eventual goal of moving to a EFM32HG108F64G CPU on it's own dedicated board.
This prototype is part of a much larger project which is moving somewhat slowly with only a single person working on it part time. If you find the design concept interesting I encourage you to take a look at the Sensaura project in more detail - I would welcome any feedback or assistance.