Industrial Ethernet integrates machinery, robotics

How a robust industrial Ethernet network can include machines and robots, while providing high performance, uptime, and enterprise-wide connectivity. Link to table of Ethernet protocols, more advice.

By Mike Miclot September 1, 2011

Manufacturers are increasingly using industrial Ethernet networks to connect human machine interfaces (HMIs), controllers, distributed I/O, instruments, and sensors to each other within a machine or robot, or to connect a customer’s communication system and link back to the home office from the customer’s site.

The ubiquity and inherent advantages of Ethernet over other networking protocols will eventually make many other kinds of networks uncompetitive by comparison. As a result, Ethernet is rapidly becoming the network of choice for machine builder OEMs and their customers. Demand from OEMs and their customers for increased connectivity and information flow is driving the growth of industrial Ethernet, which is forecast to achieve growth of about 10% per year over the next five years. For most industrial OEMs, it makes sense to evaluate Ethernet.

While the case for Ethernet on the plant floor is strong, there is no denying that some resistance remains regarding Ethernet connectivity with machines and robotics. In this article, we will examine some of the challenges and solutions and provide guidelines for more informed decision making and implementation.

Why choose Ethernet?

In the commercial and office environment, computer and device serial ports are being replaced by Ethernet ports with IP addresses. Much the same trend is evident in the industrial arena as PCs, controllers, I/O, motor drives, instruments, sensors, and other automation components increasingly include one or more Ethernet ports.

As depicted in the TCP/IP and OSI model illustration, IP addresses reside at Layer 3 of the OSI model. Though Ethernet is officially defined by Layers 1 and 2 of the OSI model, in reality most people use the term Ethernet to encompass Layers 1 through 4, thus including IP as well as TCP/UDP, and that’s the terminology used here. The beauty of the OSI model is that it enables the top layers 5 through 7, the space where all the protocols and network management components of a network reside, to be reused when upgrading from a serial network to Ethernet. As a result, it’s possible to quickly migrate from a serial-based protocol, such as Modbus, to an Ethernet-based protocol, like Modbus TCP, with minimal effort and, most importantly, with little or no additional end-user support or training.

Specifically, if Modbus was being used for networking over a serial link, most of the communication code and program would continue to be valid in the new Modbus TCP Ethernet environment. The main difference would be the capability to transfer data at much higher speeds. Other benefits would include wider availability of newer automation components with an Ethernet port as opposed to a Modbus port, along with increased ease of integrating these new components into the overall control and communications network.

The basis for this easy integration is the widely used and accepted standards at all layers of the OSI model. These standards support and enable integration of components from multiple suppliers in a common environment, effectively making it possible to plug and play with little or no effort required to create custom software communications code. Even the simplest of today’s production lines include a wide range of different machines and robots, most often supplied by a number of OEMs. Adding to the complexity is the fact that most machines and robots contain a number of automation components, such as HMIs, controllers, motor drives, instruments, and sensors. If not designed and managed properly, increased integration of real-time control and business networks can increase security risks and vulnerabilities. But this risk can be managed through proper policies, procedures, and network design so that the resulting system not only provides a fast reliable network, but also meets security requirements and ensures that the integrity of the system is maintained at all times.

Connections boost efficiency

The capability of OEMs to integrate information from the automation network throughout the enterprise—and potentially to the OEM’s supplier and customers—provides measurable benefits to all concerned.

  • Real-time inventory management becomes possible for the manufacturing facility, its supply chain, and customers, as part and material usage information is made available.
  • Recipes and production orders can be automatically downloaded and changed as required.
  • Production information becomes available throughout the enterprise, often provided by automation components equipped with Ethernet ports and web server capability.
  • Troubleshooting becomes easier as a wealth of status information can be provided not only to manufacturers, but to the machine OEM as well. This allows remote support of machines and robots, which can be continuously monitored by internal and external experts.

In addition, many OEM customers and suppliers are pursuing smarter systems by implementing plant floor information systems and microprocessor-based technologies to capture and automate key business processes.

Networks using standard Ethernet and common protocols also make it possible for each machine in an assembly line to communicate with the next unit in the line regarding its status and utilization. If each piece of equipment in one or more assembly operations knows the status of its neighbors, it becomes possible to synchronize machine inputs and outputs, thereby allowing for the maximum use of each machine and optimizing throughput across the facility.

Thus, if one machine is experiencing some form of difficulty or backlog, the result no longer has to be an unscheduled full shutdown and start-up with the associated expenses and lost production. Instead, the upstream machine’s output can either be transferred to a parallel line, or the machine’s output can be reduced until the backlog is cleared. The key to deriving these and other operational benefits from internal and external connections is high network reliability—and this must begin with the infrastructure components used to build the Ethernet foundation.

Robust industrial Ethernet

While office Ethernet shares the same standards-based structure as industrial Ethernet, the demands on the network equipment are very different. Since a system is only as good as the physical infrastructure on which it is built, it is imperative that network connectivity components used in industrial settings—cables, connectors, switches, and other hardware—be specifically designed to withstand harsh environments.

Commercial Ethernet products are designed for use in office environments, which tend to be relatively clean and quiet, with critical network equipment sequestered in the data center and telecommunications rooms. On the plant floor, however, network equipment may be exposed to a range of environmental hazards, including high temperatures, high moisture and humidity, dirt, dust, oil and chemical contaminants, machine vibration, “dirty” power quality, voltage spikes, and electromagnetic interference and radio frequency interference (EMI/RFI). Only ruggedized, industrial-grade components are designed and built to withstand the deleterious effects and damage that these threats can cause.

Automation and Ethernet components designed for industrial use are readily available, but their cost is somewhat higher than that of their commercial counterparts. However, considering the huge financial impact of an unscheduled outage (lost production, rework, loss of safety, and possible machinery damage), it becomes clear that the incrementally higher cost to purchase top quality, industrial-grade components is well worth it. It’s an investment that will pay off over years of service, in terms of high reliability, maximum uptime, and superior performance—and that’s why industrial Ethernet is fast becoming the standard in modern manufacturing plants.

– Mike Miclot is vice president, marketing, Belden’s Industrial Solutions Division. Belden Inc. designs and manufactures signal transmission solutions for enterprise, industrial, and wireless networking markets, and for specialty markets. Edited by Mark T. Hoske, CFE Media, Control Engineering, 

Industrial Ethernet: Application Optimization – webcast
Friday, October 7, 2011 at 2 p.m. ET/1 p.m. CT/11 a.m. PT (… and archived after that for viewing.)

See additional links below: More advice and table of Ethernet protocols used in industrial applications.