The term Ethernet refers to the family of local-area network (LAN) products covered by the IEEE 802.3 standard that defines what is commonly known as the CSMA/CD protocol. The original Ethernet was developed as an experimental coaxial cable network in the 1970s by Xerox Corporation to operate with a data rate of 3 Mbps using a carrier sense multiple access collision detect (CSMA/CD) protocol for LANs with sporadic but occasionally heavy traffic requirements. Success with that project attracted early attention and led to the 1980 joint development of the 10-Mbps Ethernet Version 1.0 specification by the three-company consortium: Digital Equipment Corporation, Intel Corporation, and Xerox Corporation.
Data terminal equipment (DTE) - Devices that are either the source or the destination of data frames. DTEs are typically devices such as PCs, workstations, file servers, or print servers that, as a group, are all often referred to as end stations.
Data communication equipment (DCE)- Intermediate network devices that receive and forward frames across the network. DCEs may be either standalone devices such as repeaters, network switches, and routers, or communications interface units such as interface cards and modems.
Ethernet provides the users with two core benefits : Existing field bus systems have by now reached a complexity, which requires an increase in the bandwidth of communication systems. Ethernet may form a future-proof basis for expected increases in performance
· Specialised network systems cover specialised tasks. This requires adequate user competence in handling numerous networks and functions like automation, vertical communication, image processing, drive control, and safety
A universal field bus system, which can be represented by Ethernet, is also considered an appropriate means in meeting the most extensive level of convergence.
From the viewpoint of the IT department, in the world of manufacturing, the factory contains terminals that are connected to a central IT system. Machines are handled as PC workstations, each one being assigned to a telecommunication outlet (TO), in compliance with international standards for applications-independent structured cabling to ISO/ICE 11801.
Compared to the office, the requirements of automation technology do not allow the unmodified integration of machines into existing office networks. The following requirements in industry have been identified:
· Permanent availability of the network on a time scale <10 ms
· Network areas secured by means of suitable devices (security, safety)
· Creation of physical subnet within individual machines
· Installation of diverse topologies (line, ring, for example)/flexible cascading options
· Definition of real-time areas
· Alternative and combined use of different transfer media (fibre optic cable, copper cable, wireless)
· Direct connection of two IP67 devices (transfer segments without patch cables)
· Transfer segments with different media (for cable carriers, for example) and a large number of connectors
The office network is only used for integrated communication, than for control tasks without the plant or machine. The situation is the same in a field bus subsystem, which controls all vital parameters of the process. The integration of data required for production control is implemented via network node. This integration of the machines can be implemented via the standard office network, and takes place in current systems at a layer above the field bus subsystem in a separate network infrastructure.
Field play at office
An obvious approach is the physical separation by forming two network hierarchies.
- Industrial building networking: networking of the plant building for bi-directional, vertical integration of production data
- Unit networking: networking within a unit for the purpose of process control
The separation of these two networking applications as such does not fully solve the problem of heterogeneous networks from the user perspective, but nevertheless represents the starting point to solutions providing an expedient application-specific structure.
Industrial building networking
By comparison with the office world, industrial buildings differ in terms of their ambient conditions. The differences may seem less weighty as long as the new factory floor does not contain any machinery. The aim is to create an infrastructure, which is optimally suited to meet future requirements of the shop floor. This requires a global pre-wired system, which is suitable for all applications.
Production cell networking
In contrast to the office network, an automation network does not consist of identical units, but rather of diverse devices, which are distributed within the plant according to their application. There, the switch cabinet of a tooling machine may accommodate up to ten nodes on a single (m2), in contrast to only one node on an area of 100 m2, for example, in a sewage plant. This requires a certain flexibility of the network, and the topology must be adapted to the relevant application. In addition to the line, star and ring topologies, complex tree structures are also implemented in these areas. Automation applications require a cascading depth of 20 nodes (layers) or more, as are usual in a line topology. Such conditions are not encountered in the three-layer model of the office building.
A production unit in the automation system requires structures with a significantly higher flexibility and complexity, rather than the rigid, horizontal hierarchy model to ISO/IEC 11801. The integrated network used in office networking, combines the demand for a uniform, repetitive application for the computer connections. By contrast, industrial applications have good reason to set the focus on topologies deviating from the horizontal hierarchy. Reduced wiring effort, for example, forms a vital criterion in the users assessment. The decisive factor is not the aspect of saving cable material, but rather the reduced space requirements in the switch cabinet. The benefits of an increase in installation space by discarding the installation of large cable ducts are apparent. This result is guaranteed by the interconnection of all network nodes with a central switch that supports the connection to 24 V power supply units and to 400 V mains.
A layout consisting of several hierarchy layers also supports an electrical segmentation, resulting in autarchic distributed units, which remain operational in the event of disturbances and failures. Such a structure allows the conversion and expansion of the installation in one plant section, while the other section continues production. In this way the installation reflects the automation topology. The modular layout of automation units has become widely accepted. Various function groups are merged to form complex plants. This trend continues within stand-alone machines. Here too, users benefit from the cost-effective and efficient implementation of a modular structure of high granularity. Expansions are possible by simply adding on modules. This aspect is of particular importance in the assembly area - for example, in the automotive industry - in view of the necessary, continuous adaptation of the infrastructure to plant requirements, due to the production of new models.
Network components need The demands made on network components are structured according to function and mechanical design. As office components are not able to cope with the mechanical conditions of an industrial environment, network components must be adapted to withstand tough ambient conditions. This also involves the adaptation of connectors and housings, and their combination with the functionality of network components, which are suitable for industrial utilisation. Industrial networks can only be operated if they are safely isolated from the office network. Such separate networks, as well as networks of the same type, always require the implementation of defined gateways or network nodes. In this way, the user can easily integrate a new machine into the network system by cascading the various machine modules and creating more than ten additional layers in the hierarchy. However, network management may argue that such actions require the integration of ten distributed switches, in addition to the master switch, the assignment of corresponding IP addresses, and the local replacement of faulty switches. However, troubleshooting network problems definitely requires a local unit diagnosis. Although manageable SNMP-compatible switches would support such operations using a central network utility, maintenance staff usually does not consider this an acceptable approach.