Industrial ethernet 101

Transmission Control Protocol/Internet Protocol (TCP/IP) is to ethernet as spices are to food. You rarely get the best experience from one without using the other. That's because the TCP/IP suite is the key, behind-the-scenes technology that brings the internet and intranets to life - just as spices bring out the flavours in our food. As a result, TCP/IP is inextricably linked with ethernet in the business and consumer worlds.

But what many people don't understand is that using ethernet and TCP/IP only guarantees that ethernet devices can coexist within a network; it does not guarantee that these devices can communicate effectively. For devices to be able to understand and operate with each other, a common application-layer protocol is needed.

The TCP/IP architects understood this problem and designed these protocols to work with multiple application layers. To date, there are many used with the TCP/IP suite - for example, protocols for file transfer (File Transfer Protocol (FTP), email (Simple Mail Transfer Protocol (SMTP) and World Wide Web HyperText Transfer Protocol (HTTP).

[image]

Now that ethernet has infiltrated the plant floor, networking organisations and automation vendors are promoting several new application-layer protocols. These protocols are intended to make ethernet suitable for the rigors of industrial automation.

Introduced in early 2000, EtherNet/IP is becoming the front-runner of many of the industrial ethernet solutions. Part of the reason is that it is based on open technology, using the same application layer - the Common Industrial Protocol (CIP) - as DeviceNet and ControlNet. This approach offers myriad benefits to automation users and vendors alike: low product-development costs, ease of use, simple product and network integration and multi-vendor interoperability.

Further, more than one vendor or organisation backs EtherNet/IP. ControlNet International (CI), the Industrial Automation Open Networking Alliance (IAONA) and the Open DeviceNet Vendor Association (ODVA) recognise and/or support this networking standard. As a result, EtherNet/IP delivers interoperable ethernet products from a large community of vendors.

What is EtherNet/IP?

It's important to understand the basic technology behind the name before delving too far into the specifics. EtherNet/IP is best defined by deconstructing its name. The 'EtherNet' portion refers to commercial off-the-shelf Ethernet (IEEE 802.3) - not a modified, proprietary version - and the TCP/IP suite. 'IP,' on the other hand, stands for Industrial Protocol and is what distinguishes this network.

Unlike many industrial ethernet options, EtherNet/IP uses an already proven, open protocol at the application layer (ie, CIP). So EtherNet/IP is CIP deployed on an ethernet TCP/IP network, just as DeviceNet is CIP deployed on a Controller Area Network (CAN).

In terms of how it works, EtherNet/IP uses TCP/IP to send explicit messages - those in which each packet not only has application data, but also carries the meaning of the data and the service to be performed on the data. EtherNet/IP also leverages standard User Datagram Protocol/Internet Protocol (UDP/IP, part of the TCP/IP suite) transport services, which provide higher performance and multicast functionality for real time - also known as implicit messaging. As it leverages both TCP/IP and UDP/IP protocols to encapsulate networked messages, companies can use EtherNet/IP for both information and control applications.

EtherNet/IP - here and now

EtherNet/IP is almost three-and-a-half years old. In that relatively short time, this network has seen a significant amount of interest and activity. A year after CI and ODVA unveiled the specification, Rockwell Automation launched the first wave of EtherNet/IP products, which centered around the Allen-Bradley ControlLogix, ProcessLogix, PLC 5 and SLC 500 controller families. Since then, many other vendors have either developed EtherNet/IP products or announced plans to do so.

No piece of technology is capable of addressing every need. That holds true for industrial networks as well. EtherNet/IP is bound by the inherent limitations of ethernet.

But EtherNet/IP has enough features and functionality to make it work for a wide variety of applications. As a result, it's more versatile than most networks. Because EtherNet/IP provides both control and information capabilities, it also gives companies the ability to collapse their communications architecture.

The before and after

A traditional application would look something like the following. An information network links a programmable logic controller to an upper-level, MES-type system. The controller is also connected to a second network - a dedicated I/O bus - to link all the peripheral devices on the factory floor. The end user would use the I/O network to modularise the construction and wiring of the system, gather remote data and manage the speed of production.

In comparison, the controller could now be linked to an EtherNet/IP network for both information and I/O using a flat network architecture. Depending on the particular application, this setup can offer not only better performance than the traditional application, but additional functionality as well, such as:

• New services like voice and video - eg, installing a camera to watch the plant floor and product flow remotely.
• Variable topologies - eg, setting up all the programming terminals on a virtual LAN, which would allow engineers to connect the terminals on a single network, yet segregate and isolate multiple control systems at the same time.
• IT integration - eg, accessing the MES system from the plant floor to review and refresh in real time maintenance scheduling information.
• Remote management - eg, programming the control system to send a message to a maintenance pager when alarms occur.
• Internet connectivity - eg, accessing an online troubleshooting guide to adjust an inoperative sensor.
• Flexible security - eg, restricting access to a port on a switch, which results in restricted access to a specific I/O block.

These additional capabilities provided by ethernet give manufacturers a significant advantage, assuming EtherNet/IP can meet all of their industrial network requirements. These capabilities are not, and cannot be, available on a fixed-purpose industrial network. Moving to this type of architecture marks a philosophical and operational shift for most companies - one that opens a realm of new possibilities to an innovative user.

And after the after?

Like all newly developed technologies, EtherNet/IP is not static. So the organisations and their member companies responsible for its growth, including Rockwell Automation, are sure to add features that will help companies evolve with changing business requirements, such as safety, synchronisation and wireless utilities. But when it comes to new technology, no one can accurately predict what lies on the horizon.

What is industrial ethernet? The ethernet network is a local area network (LAN) protocol developed by Xerox Corporation in cooperation with DEC and Intel in 1976. Ethernet uses a bus or star topology, and supports data transfer rates of 10 Mbps (standard) or 100 Mbps (using the new 100Base-The version). The ethernet specification served as the basis for the IEEE 802.3 standard, which specifies the physical and lower software layers. Ethernet uses the CSMA/CD access method to handle simultaneous demands. It is one of the most widely implemented LAN standards. It is designed to deal with connecting open systems to communicate with other systems. It consists of seven layers: a complex structure is partitioned into a number of independent functional layers. Each layer provides a set of services by performing some well-defined sets of functions. These services are provided by the layered-specific functional entities. Services at a layer can only be accessed from the layer immediately above it. Each layer uses only a well-defined set of services provided by the layer below. Protocols operate between peer entities in the different end systems (peer-to-peer protocol rules). Physical layer The physical layer is responsible for passing bits onto and receiving them from the communication channel. This layer has no understanding of the meaning of the bits, but deals with the electrical and mechanical characteristics of the signals and signalling methods. Data link layer Data link layer is responsible for both Point-to-Point Network and broadcast network data transmission. It hides characteristics of the physical layer eg, transmission hardware from the upper layers. It is also responsible for converting transmitted bits into frames. It transmits the frames into an error free transmission line by adding error control and flow control. Network layer Network layer is responsible for the controls of routers and subnets operation. It also handles the formation and routing of packets from source to destination with congestion control. Transport layer Transport layer is a kind of software protocol to control packets delivery, crash recovery and transmission reliability between sender and receiver. Multiplexing between transport and network connections is possible. Session layer Session layer provides dialogue control and token management. Presentation layer When data is transmitted between different types of computer systems, the presentation layer negotiates and manages the way data is represented and encoded. Application layer This top layer defines the language and syntax that programs use to communicate with other programs. For example, a program in a client workstation uses commands to request data from a program in the server. Common functions at this layer are opening, closing, reading and writing files, transferring files and email messages, executing remote jobs and obtaining directory information about network resources.