EtherCAT, the abbreviation of Ethernet for control automation technology, is an fieldbus system based on open Ethernet. The aim of the development was the adaptation of Ethernet to automation applications requiring short data update times (also known as cycle times).
What is EtherCAT?
Typical automation networks have a short data length per network node with a smaller payload than the min. payload of an Ethernet frame. The use of one frame per network node per cycle leads to the use of a small bandwidth and consequently, a weak overall performance of the network. EtherCAT approaches the matter differently: namely by ‘processing on the fly’.
The EtherCAT applications read the data addressed to them while the telegram runs through the application. Input data also are inserted while the telegram is running through. The frames are only delayed by a fraction of one microsecond at each network node, and many network nodes - normally the entire network - can be addressed with a single frame.
The EtherCAT protocol is optimized for process data and is transmitted directly within the standard IEEE 802.3 Ethernet frames by using the Ethertype 0x88a4. It can consist of several sub-telegrams. Each one serves a certain memory region of the process mapping.
The data sequence does not depend on the physical order of the nodes in the network; addressing can be done in any order. Transmission, multicast, and communication between slaves are possible and must be performed by the master application. When IP routing is required, the EtherCAT protocol can be inserted into UDP/IP datagrams.
Short cycle times can be achieved as the host microprocessors in the slave applications are not involved in the data processing. The entire process data communication takes place in the slave controller hardware. Normal network update rates are around 1 to 30 kHz. However, EtherCAT also can be used with slower cycle times.
Because of use of double Ethernet levels, the slave control devices close an open port automatically and send the Ethernet frame when no downstream application is found. Slave applications can have two or more ports. Because of these characteristics, EtherCAT can support nearly any physical structure. As such, the bus or line structure that is known from fieldbuses is available for the Ethernet.
As 100BASE-TX Ethernet levels are used, the distance between two network nodes can be up to 100 m. Up to 65535 applications or units per segment can be connected. If such a network is connected in ring-shape (requires two ports on the master unit), it is possible to cut down on the use of cables.
A distributed clock mechanism is used for synchronization. This leads to very a low flutter of less than 1µs, even when the communication cycle is fluttering. This is equivalent to the IEEE 1588 precision time protocol standard. For this reason, no special hardware is required in the master unit, and insertion into any software or standard Ethernet MAC is possible.
The typical process in order to introduce a distributed clock mechanism is started by the master. This is done by sending a message to all slaves at a specific address. On receipt of this message, all slaves change the value of their internal clock twice, once when the message is received and once when it returns. The master can read all of the changed values and can calculate the delay for each slave. This process can be repeated as often as required in order to prevent zero values. Total delays are calculated for each slave, depending on its position in the slave ring. This is uploaded to an offset register. The master finally triggers a write-read process for a message at the system clock, which causes the first slave to carry the reference time, and all other slaves will adjust to this. To maintain the synchronization of the clocks after initialization, the master or the slave periodically must send a message in order to determine any speed change between the internal clocks of the respective slaves.
The system clock is a 64-bit counter with a base unit of 1 ns (starting on January 1st 2000 at 0:00).
The improvement of the protocol by Safety over EtherCAT provides for secure communication and control communication in the same network. The security log is based on the application level of the Ethernet system without influencing the lower levels. It has been tested according to IEC 61508 and meets the preconditions of the safety integrity level (SIL) 3.
Gateway applications are used to integrate fieldbus components into an EtherCAT network. Other Ethernet protocols also can be jointly used. The network is completely transparent for the Ethernet applications. All Internet applications also can be used in such an environment: integrated web servers, e-mail, FTP transfer, etc.
EtherCAT Technology Group
The ETG is an international user and consumer organization, and it has its headquarters in Nuremberg. It was founded in 2003 and has offices in Tokyo (Japan), Beijing (China), Seoul (Korea) and Austin, Texas (USA). In June 2010, it had over 1350 member companies from 50 countries. The ETG sees itself as a forum for end-users from different fields. The ETG offers information concerning the system and its applications as well as organizes technological training groups. It also has technical / marketing committees and distributes information at trade fairs.