Learning about modern network communication technologies

Communications plays a vital role in most industrial automation solutions and Rockwell Automation's Integrated Architecture offers a variety of options.

Rockwell communication technologies can be categorized by the primary network technologies (those that are actively used in modern implementations of a Logix system) and legacy network technologies (those that have been used frequently in the past but are not installed on the new Logix control systems).

In the next section, we will introduce some of the primary network technologies we see today in Rockwell Automation-based networks.

Primary network technologies

Logix relies on the following three primary technologies for network communication:

• DeviceNet
• ControlNet
• EtherNet/IP

DeviceNet and ControlNet were developed by Rockwell Automation and are based on the Common Industrial Protocol (CIP). They are maintained by the Open DeviceNet Vendors Association (ODVA). EtherNet/IP (note the capitalized N and that IP, in this case, stands for Industrial Protocol) was developed by ODVA in 2001 and was adopted by Rockwell Automation.

Let's discuss each of these in more detail in the following subsections.

DeviceNet

DeviceNet is explicitly designed to communicate with and provide power to the lowest-level field devices. It can communicate with intelligent devices that support a DeviceNet module, such as sensors, Variable-frequency drive (VFDs), valves, motors, and distributed I/O blocks. DeviceNet is convenient as it provides both signal and power to the device. However, it's not designed to handle high volumes of network traffic. DeviceNet networks must be configured using Rockwell Automation's RSNetworx for DeviceNet software. DeviceNet allows selectable data rates (such as 125, 250, or 500 KBaud), which enables system integrators to potentially trade speed for communication-run distances. DeviceNet devices are not required and do not necessarily support all the data rate ranges, but many do.

In this book, we will primarily focus on EtherNet/IP networking, but links to further DeviceNet resources are available at the end of this chapter.

In the next section, we will introduce the Rockwell I/O backbone protocol, known as ControlNet.

ControlNet

ControlNet is a deterministic network technology that acts as the I/O communication backbone for a control system. Deterministic data collection guarantees that new data arrives within a predefined interval. It enables the interconnection of multiple DeviceNet networks and shares data with other controllers. It is capable of full network redundancy, so if a redundant ControlNet network is configured and one cable is broken, the network can continue operating normally.

Connections over ControlNet are configured as scheduled or unscheduled. The ControlNet Network Update Time (NUT) is the millisecond interval for collecting updated data in a ControlNet network. ControlNet is a highly repeatable deterministic communications method—albeit slow (locked at 5 Mbits/s), but deterministic nonetheless. Deterministic networks are ideal for processes that require continuous synchronized data and cannot tolerate any data update delays, such as motion control.

The ControlLogix backplane, ControlBus, is nearly identical to the ControlNet networking. It is based on an open standard for industrial network protocols known as Fieldbus and is based on the same Fieldbus IEC 61158 communication standard as Foundation Fieldbus, PROFIBUS, and Interbus. The Fieldbus architecture provides the skeleton that ControlNet is based on. However, due to fundamental implementation differences, it is not able to directly communicate with any other Fieldbus protocols. On top of the Fieldbus architecture, at the protocol application layer, ControlNet uses the CIP to provide its functionality. ControlNet can upload and download programs, perform I/O forcing and online editing, and communicate with the remote I/O racks.

ControlNet uses a Quad Shield RG6 coaxial cable as the networking media, which requires it to use network taps for each drop and terminating resistors at the end of each segment. In this book, we will primarily focus on EtherNet/IP networking, but more ControlNet resources are available in the appendix.

We will introduce EtherNet/IP in the next section.

EtherNet/IP

EtherNet/IP is the most widely used communications technology in the Integrated Architecture ecosystem today because of its speed, scalability, and ease of integration with enterprise-level network hardware. EtherNet/IP is the primary communication method used in the examples in this book. It combines the IEEE 802 standard Ethernet technology stack with the object-based CIP. Basing the communications on Ethernet allows ease of integration with the existing enterprise IT networks.

The CIP application-layer protocol also allows the Logix controllers to communicate control, safety, synchronization, motion, configuration, and diagnostic information with devices from hundreds of different vendors. CIP enables EtherNet/IP to upload a program, download a program, force I/O values, monitor code, perform online edits, and connect to the remote I/O racks.

In the legacy I/O systems, the Programmable Logic Controller (PLC) would poll (request data at a set interval) digital input modules for new data. The CIP protocol on EtherNet/IP digital input modules can perform the following tasks:

• Return data on Change of State (COS).
• Return data at a Request Packet Interval (RPI), scheduled in milliseconds.

One notable difference from the other methods of data collection is that EtherNet/IP is non-deterministic. Deterministic, in the context of communications, means the delays in delivering a packet of data across a network are known in advance and are not subject to change. Non-deterministic data collection does not guarantee that the new data will arrive within the RPI. IEEE 802 standard Ethernet is fundamentally non-deterministic as it is not scheduled within a set time window. However, the speed of EtherNet/IP with a modern full-duplex switch almost negates this fact.

EtherNet/IP uses User Datagram Protocol (UDP) to communicate the basic I/O and non-critical information on the network. UDP does not perform any error checking or handshaking mechanisms, so the delivery of information is not guaranteed and the data is susceptible to any network-related data loss. UDP is ideally suited for real-time information as it trades speed for guaranteed data delivery. In real-time systems, dropped packets are preferable to waiting for delayed packets. It is far better to get the next most-recent value than circling back for a packet that is already stale. Using UDP, EtherNet/IP is capable of collecting data by polling cyclic and change-of-state monitoring. EtherNet/IP makes use of UDP port number 2222.

Transmission Control Protocol (TCP) is used by EtherNet/IP for critical data, in tasks such as writing set points, parameters, and recipes and uploading and downloading programs. TCP has a built-in error checking and a three-way handshake mechanism that ensures that no packets are lost during the data transfer. TCP sacrifices data transfer speed for guaranteed delivery of information. EtherNet/IP makes use of TCP port number 44818.

Stratix is a line of industrial networking and security solutions from Rockwell Automation that has been engineered specifically for EtherNet/IP (and is based on the ubiquitous Cisco hardware platform). However, because EtherNet/IP is based on the IEEE 802.x standards, it is possible to use normal network switchgear with EtherNet/IP. Rockwell Automation recommends that you use robust industrial-grade networking equipment with the Logix controllers. Furthermore, Stratix switches seamlessly integrate with the Logix platform and can easily provide health and status information as the native Logix tags.

In Chapter 15, Building a Robot Bartender in Logix, we will go through the process of selecting an EtherNet/IP card (ENBT, EN2T, and so on) for our Robot Bartender project.

Now that we have covered the three primary network communications used in modern Rockwell solutions, we will cover some of the legacy network technologies in the next section.