Trends such as a changing workforce, supply-chain issues and product individualization are driving manufacturers to reduce costs, improve efficiency and be more flexible in production. Smart manufacturing provides a path to address these challenges, and sensor-level protocols provide a simple and cost-effective way to gather device data and communicate it to control, process and enterprise levels and/or the cloud.
Get your subscription to Control Design's print magazine, free to qualified individuals in North America.
The case for industrial networking and protocols
Industrial networking has rapidly gained acceptance and market share as a way to connect factory-floor devices and to integrate input/output (I/O). The major industrial Ethernet protocols—EtherNet/IP, Profinet, EtherCAT—were introduced between 2001 and 2003. Industrial Ethernet accounts for about two-thirds of all industrial networking, according to a 2022 survey from HMS.
In a controls-architecture approach, each sensor and actuator—I/O point—is individually wired back to the controller. This is an acceptable solution for smaller machines with a few I/O, but it presents significant challenges for larger machines with many I/O and long distances, including:
• long cables runs
• large wire bundles
• complicated and confusing panels
• wiring subject to electromagnetic interference (EMI)
• longer installation time and costlier wiring
• difficult-to-troubleshoot devices, wiring mistakes and/or short circuits.
The emergence of Ethernet-based industrial protocols addressed these challenges, but not all devices and I/O could or should be connected to Ethernet. Reasons include:
• high cost of chipset relative to device cost
• data capability in excess of what is required by most sensors and actuators
• individual Ethernet IP addresses required for each device
• IT resources may be required to implement Ethernet solutions, including networking of sensors and I/O.
Why use device-level protocols?
To deal with these issues, device-level networks came into existence. The benefits of these include:
• low-cost connections, even for inexpensive devices
• appropriate data capacity for most sensors and actuators
• ease of wiring, troubleshooting and adding/removing devices
• off-the-shelf cables
• digital signals with higher EMI immunity
• industry standard, open protocols.
IO-Link is a good example of a device-level protocol. It is an open protocol supported by more than 400 members with thousands of devices, and more than 27 million installed nodes. Original equipment manufacturers (OEMs) can mix devices from multiple suppliers in their applications. Connections are plug-and-play using industry-standard three- or four-conductor cables. The devices provide more data than on-off or analog sensors, with three types of information: process, device/parameter and event/diagnostic data.
A balanced solution
Many users find that the most effective solution is a network architecture that uses an industrial Ethernet protocol for control- and supervisory-level communications, a device-level protocol to connect sensors and actuators and a bridge or master in between. The device-level data is gathered and communicated over the device protocol to the master, which consolidates the information and communicates it to control and supervisory systems using industrial Ethernet. This approach combines the best of both worlds:
• Ethernet-based protocols to communicate large amounts of data at high speeds in real time, fast enough for motion control in some cases
• device-level protocols for simple, low-cost connection of devices with fast transmission of smaller amounts of data.
Growth and innovation in industrial Ethernet and device protocols have enabled users to create controls architectures that are fast, flexible and cost-effective.
