Today’s cars are electronic devices – and are already converging into communication devices. Actually everything, down to the smallest chip, communicates with each other inside the vehicle. This can make life difficult for design engineers, since they need to design systems where one electronic device needs to communicate with others to operate. This is true in today’s industrial automated world as well as in our cars. One of the most important communication protocols in this area is the Controller Area Network (CAN) protocol which is one of the most successful and as we think, one of the most important bus protocols worldwide.
Before CAN was introduced, every electronic device was connected to other devices using wire. This worked fine, as long as the functions in the system were limited. However today’s systems have dozens of control units for managing devices, sensors, and actions. CAN helped to reduce the wires, e.g. in cars. This is one of its main advantages – especially with electronic systems becoming more prevalent in cars.
CAN provides a mechanism which is incorporated in the hardware and software by which different electronic modules can communicate with each other using a common cable. Thus it enables efficient communication between various electronic control units (ECU) which are embedded in today’s vehicles such as engine management systems, ABS, gear control, lighting control, air conditioning, airbags, central locking, a.s.o. All the devices which communicate are connected to the same two wires in a CAN bus. The protocol is based on a non-destructive arbitration mechanism which grants bus access to the message with the highest priority without any delays. Its error detection mechanisms include the automatic disconnection of faulty bus nodes in order to maintain the communication between the remaining nodes. The transmitted messages are not referenced by the transmitter node address or the message receiver but by their heading identifier. This identifier specifies the message priority within the system. CAN is so popular because of its communications being relatively immune to external interference and to single control unit faults. This reliability is among the properties that has made it the current standard in difficult environments, with wide temperature ranges, and greatly varying environmental situations.
The CAN in Automation Association is driving the development and constant improvement of the CAN bus and Spansion has been a member of this non-profit organization since summer 2013. About 560 companies are currently members of this international users and manufacturers‘ group. Representatives of CiA actively support the international standardization of CAN protocols, and CiA members develop specifications to be published as CiA specifications.
Spansion brings its expertise in the MCU area thus continuing the long-standing activities of Fujitsu Semiconductor who joined CiA 2 decades ago. We will be working in a task force to drive the standardization of the CAN FD (Flexible Data Rate) standard which is an improved and faster version of the CAN protocol. We will aim to contribute to new revisions of ISO11898 (protocol) and ISO16845 (test).
CAN FD provides more payload per message; up to 64 bytes instead of just 8 bytes beforehand. At the same time CAN FD increases the bandwidth by switching the payload section to higher speed. Some demonstrators already presented FD-networks running at a bus speed of 15 MHz which is 15 times the maximum speed of the legacy CAN. First deployments to mass production vehicles however aim at a bus speed of 4 MHz during payload transmission.
By this the range from 1-5 Mbit of which there was no cost efficient automotive network in the past, will be covered most efficiently. The CAN FD brings its advantages to this: it uses the existing networks adding more bandwidth and helping to prolong the lifetime of the CAN technology: CAN FD users can integrate this faster protocol into their existing networks without additional components for the physical layer. It is merely necessary to integrate a new communication IP which comes as embedded function of the MCU. Spansion will deploy ist first CAN FD MCU in 2014. In fact this MCU will carry multiple instances of the mCAN IP.
We are looking forward to the collaboration!
Wolfgang Wiewesiek studied electrical engineering at TU Braunschweig Germany where he also started his professional career at the Institute for Data Processing Units developing space-born control units for international scientific research projects in 1987. He joined NEC in 1997 and became the CAN specialist of the company. He holds a patent on ‘CAN mirror mode.’ In 2004, he joined Fujitsu as marketing manager for automotive microprocessors and network specialist. In Aug 2013, he became an employee of Spansion through the company’s acquiristion of the analog and microcontroller business of Fujitsu Semiconductor Ltd.