The EtherCAT Protocol has emerged as an important and dominant standard in the field of industrial automation. As technology advances, the need for faster and more efficient communication between devices becomes crucial. The EtherCAT Protocol, short for Ethernet for Control Automation Technology, is designed to meet these demands. Running on the standard Ethernet interface, the EtherCAT Protocol offers high throughput and real-time performance. Originally designed and developed by Beckhoff, a special group called EtherCAT Technology Group (ETG) was founded in 2003 to enable wider industry participation. Later it was standardized as part of IEC 61158 and IEC 61784-2 standards from 2007. Today the EtherCAT Protocol is one of the most adopted industrial fieldbus standards for automation.
In this article, we will provide you with an in-depth introduction to the EtherCAT Protocol, delving into its various aspects to demystify this powerful real-time Ethernet protocol.
To understand how the EtherCAT Protocol operates, it is essential to grasp its OSI Layer mapping. The Open Systems Interconnection (OSI) model is a conceptual framework that defines how network protocols interact and communicate with each other.
EtherCAT OSI Mapping
At the lower level, the EtherCAT Protocol operates primarily on the Physical Layer (Layer 1) and the Data Link Layer (Layer 2) of the OSI model. At the Physical Layer, the EtherCAT Protocol utilizes standard Ethernet cabling and transmits data in a ring topology. The Data Link Layer, responsible for frame transmission and error detection, leverages the Ethernet MAC. At the higher levels, the EtherCAT Protocol directly operates from the Application layer. Without all the intermediate layers, the protocol overheads are significantly reduced, thereby achieving excellent real-time performance. This lean stack is one of the reasons the EtherCAT Protocol is considered the leading real-time Ethernet protocol for deterministic automation.
The EtherCAT Protocol operates based on a unique principle known as 'processing on the fly.' This principle enables devices to process data in real-time as it passes through them, reducing delays and improving overall system performance. The key to this operating principle lies in the EtherCAT Slave devices, which are responsible for processing the data without the need for intermediate storage. Unlike traditional Ethernet where a complete frame must be received before processing, the EtherCAT Protocol enables processing as and when the frame arrives.
In an EtherCAT Protocol network, the EtherCAT Master Node sends a frame containing data and commands to all the EtherCAT Slave devices simultaneously. Each Slave device processes the data relevant to it and adds its own data to the frame as it passes through. This cascading effect allows for efficient data exchange between devices in a synchronized manner. The EtherCAT Protocol's operating principle ensures high-speed communication, low latency, and deterministic behaviour, making it ideal for time-critical applications and motion control network implementations.
The frame structure of the EtherCAT Protocol is one of the key elements that contribute to its efficiency and speed. At a high level the frame structure is similar to the standard Ethernet frame with EtherType set as 0x88A4 to indicate it is an EtherCAT Protocol frame. The source and destination addresses can be used to transfer broadcast, multicast or point-specific frames.
EtherCAT Frame Format
On the payload front, the EtherCAT Protocol frame consists of two sections: the EtherCAT header and the EtherCAT Datagrams. The header contains essentially the length of the Ethernet datagram without the FCS part. Each datagram has a 10-byte Datagram header, followed by variable-length data and a 2-byte Working Counter. The contents of the datagram are captured in the below table:
| Field | Length (bits) | Description |
|---|---|---|
| Cmd | 8 | Command type |
| Idx | 8 | Index to identify frame |
| Address | 32 | Node Address concerning this datagram |
| Len | 11 | Length of the data in this datagram |
| R | 3 | Reserved, 0 |
| C | 1 | Circulating frame (0 – Not circulating vs 1 being circulated) |
| M | 1 | Multiple EtherCAT datagrams ( 0 - Last datagram vs 1 - Another EtherCAT datagram follows) |
| IRQ | 16 | Event request register of all slave devices combined |
| Data | Variable | Data to be read or written |
| WKC | 16 | Working Counter |
Thus, within a single Ethernet frame, multiple datagram packets are packed and sent to different slave nodes in the EtherCAT Protocol network.
The EtherCAT Master Node is a central component in an EtherCAT Protocol network. It initiates communication with the EtherCAT Slave devices and coordinates the data exchange process. The EtherCAT Master Node is responsible for configuring the network, synchronizing the devices, and ensuring the proper functioning of the EtherCAT Protocol network.
The EtherCAT Master Node can be implemented with a standard Ethernet controller including a regular PC. The EtherCAT Master Node assembles data into the frame and sends it out at the beginning of each cycle. The slave nodes read the data addressed to them, write new data to the frame, and send it downstream. Once the end is reached, the frame is returned to the EtherCAT Master Node which then starts sending the next frame. Embien's product engineering services support teams designing EtherCAT Protocol systems from initial EtherCAT Master Node selection through to full system integration and validation testing.
Choosing the right EtherCAT Master Node software stack is critical to achieving the cycle times promised by the EtherCAT Protocol. Both open-source stacks and commercial EtherCAT Master Node solutions are available, each with different real-time OS integration requirements and determinism guarantees.
In an EtherCAT Protocol network, the EtherCAT Slave devices are connected in a daisy-chain topology, forming a series of nodes. Each node in the network is referred to as an EtherCAT Sub Node. These Sub Nodes are responsible for processing the data relevant to them and passing it along to the next Sub Node in the chain.
The Sub Nodes receive the EtherCAT Protocol frame from the previous node, extract the relevant data, and add their own data to the frame before passing it to the next node. This process continues until the frame reaches the EtherCAT Master Node. The EtherCAT Sub Node requires a specialized controller to achieve the 'processing on the fly' capability. While such controllers were previously available only with dedicated FPGAs and ASICs, many modern industrial MCUs/MPUs now support them as built-in peripherals. Embien's FPGA design services enable custom EtherCAT Sub Node implementations on FPGAs for applications where off-the-shelf slave controllers do not meet the performance or I/O requirements of the target motion control network.
The EtherCAT Protocol operates the Ethernet line in full-duplex mode whereby one pair of lines is used for sending the data frame from the master to all the slaves, and another pair to transmit the return data from the last slave. Thus, the EtherCAT Bus Topology is a logical ring structure irrespective of the actual network topology being employed. This flexibility in EtherCAT Bus Topology allows devices to be connected as line, tree, star — or any combination of them. As each slave daisy-chains the line in the EtherCAT Bus Topology, there is no need for network switches, eliminating the precious delays associated with typical Ethernet switches. Even in case of a line break in a ring-based EtherCAT Bus Topology, the terminating nodes on both ends transmit the Ethernet frame back to the master, allowing the master to quickly identify the precise location of the break. The EtherCAT Bus Topology also provides redundancy options for cables and even masters, enabling higher availability of the automation network.
The flexibility of EtherCAT Bus Topology is a key advantage for machine builders, because the same EtherCAT Protocol can be wired in a simple line for compact cells or extended into a tree structure for multi-axis systems without any protocol-level changes.
Synchronization is a critical aspect of the EtherCAT Protocol, especially in applications where precise timing is required. The EtherCAT Protocol achieves synchronization through Distributed Clocks, which allows all devices in the network to operate with a common and accurate time reference. Distributed Clocks in the EtherCAT Protocol work based on the concept of a Distributed Clock Domain, where each device has its own clock that is synchronized to a common reference clock. The EtherCAT Master Node takes the role of the Clock Master and distributes the reference clock signal to all the Slave devices. This is achieved via a mechanism where each node adds the timestamp first during reception of the downstream message and a second time when the return message is being transmitted from the same node. The EtherCAT Master Node uses this information to calculate the delay for each node. With this, the EtherCAT Protocol achieves deviation of less than 1 µs, which is equivalent to the IEEE 1588 Precision Time Protocol standard (PTP).
The EtherCAT Protocol offers various profiles that define the functionality and behavior of devices in an EtherCAT network. These profiles ensure compatibility and interoperability between devices from different manufacturers. EtherCAT Protocol profiles cover a wide range of applications, including motion control, I/O devices, safety systems, and more.
Each EtherCAT Protocol profile defines specific parameters, functions, and behavior that devices need to adhere to. These profiles ensure that devices from different manufacturers can seamlessly communicate and work together in an EtherCAT Protocol network. Some of the common profiles include the Servo drive profile (SERCOS), CAN Application Protocol over EtherCAT (CoE), Ethernet over EtherCAT (EoE), File Access over EtherCAT (FoE), and ADS over EtherCAT (AoE). The CoE profile is especially relevant for motion control network applications, as it maps CANopen device semantics onto the high-speed EtherCAT Protocol transport.
EtherCAT-Tunneling TCP/IP is a feature of the EtherCAT Protocol that allows EtherCAT frames to be transmitted over standard Ethernet networks. This feature enables the EtherCAT Protocol to extend its reach beyond the boundaries of EtherCAT networks and communicate with devices in the internet realm.
With Ethernet over EtherCAT, TCP/IP frames are encapsulated in EtherCAT Protocol packets and transmitted over Ethernet networks. With the EtherCAT Protocol as the primary communication mechanism, the real-time nature of the system is preserved without compromising the performance expected of the network. This brings popular internet protocols like TCP/IP, MQTT, HTTP, HTTPS, and VPN onto the industrial automation network, making the EtherCAT Protocol a true convergence platform for both OT and IT traffic. The high-speed and synchronized nature of EtherCAT makes it suitable for a wide range of industrial and embedded application domains.
The EtherCAT Protocol stands out as a leading real-time Ethernet protocol for industrial automation, achieving sub-microsecond synchronization through its on-the-fly frame processing architecture. The flexible EtherCAT Bus Topology and the coordinating role of the EtherCAT Master Node together make it the go-to choice for demanding motion control network applications in robotics, CNC, and semiconductor manufacturing equipment.
Embien's product engineering services cover the full EtherCAT Protocol development lifecycle — from slave controller selection and hardware design to firmware integration and system validation.
Embien's FPGA design services enable custom EtherCAT Protocol slave implementations on FPGAs, delivering the deterministic processing performance that real-time automation demands.
A case study of Embien deploying an industrial environmental monitoring system with real-time data acquisition and EtherCAT Protocol-based device integration.