Industrial Ethernet alternatives
When choosing between an industrial network or other option, such as industrial Ethernet network or industrial wireless, sort out the cost and timing of legacy upgrades.
Yes, the train has left the station, the jury verdict is in. It is official; Industrial Ethernet in its various forms is beginning to dominate manufacturing networks globally as the solution platform of choice. It’s a communications network; it’s a fieldbus; it’s highly interoperable; it’s the superman of networks. So does that mean it is the right solution for your application? Not always. Keep reading to see when you might want to consider an alternative to the Industrial Ethernet juggernaut.
Not so long ago, industrial networks were mainly proprietary creatures. Control devices were largely hardwired. Information crawled along at relatively low speeds, and network topologies came in a variety of configurations. Network cabling media was largely copper and more often than not installed by an electrician rather than a network cabling firm. The desire to integrate real-time manufacturing information with business systems was very much present, but the state-of-the-art often fell short of the mark.
Many proprietary industrial communication networks evolved including DH1, DH, and DH+; Modbus RTU; TiWay; and many, many more. Each controller manufacturer also had at least one proprietary device control I/O network, sometimes more depending on the variety of makes and models it offered. To make matters worse, it was not uncommon to have multiple networks co-mingled on the same plant floor.
In response to this sea of proprietary platforms was a push for open standards from a variety of sources. The IEEE 802 project led to the establishment of Ethernet on the communication side of things. In the I/O world, Allen Bradley (now Rockwell Automation) developed a standard that eventually became Device Net and Control Net, while Siemens, along with other European manufacturers, developed a standard that eventually became PROFIBUS. The AS-Interface is another popular open technology originally developed by a consortium of sensor manufacturers that is still much in use for simple field I/O devices. Many other standards of all sorts exist today.
Now that these great open standards are commonly available, why would we want to go back to our old proprietary ways? One reason is that our past is still alive and well in many of our manufacturing facilities. Your choice of networking solutions comes down to how much of this legacy stuff you have in your existing automation infrastructure, and the effective remaining lifecycle of the components of that infrastructure. This latter part is particularly important.
Suppose, for example, you have to upgrade a portion of your facility currently running Allen Bradley 1771 series I/O on an RIO network platform.—a legacy platform to be sure, but one supported by a large community of control engineers and systems integrators. If the upgrade is small, and your parts are all still active, it will generally make the most sense to stay on this legacy platform for small projects.
Take the same set of circumstances except that you discover some of the components in this system are designated as Silver Series, a term Rockwell uses to indicate a part is near the end of its lifecycle. You now have to give this situation much more consideration, even if the project is small, and the upgrade costs are disproportionate to the amount of work at hand.
The same argument will go for any modification to a legacy system. You will always need to balance the extra cost and time of upgrading a legacy system to a modern platform, which could be quite significant, against the benefits gained from such an upgrade. Keep in mind that eventually you will have no choice but to upgrade. The point is to plan the migration rationally. In the short term, maintaining the legacy platform will often make the most sense.
On the fieldbus side, recent trends have been away from some of the open platforms such as Control Net, an open standard but with deployment mainly constrained to Rockwell Automation systems. In that case trending has been toward Industrial Ethernet solutions. Alternatively, the AS-Interface is growing in acceptance and installations, and plays nicely as a partner network with higher level fieldbus platforms including those utilizing Industrial Ethernet.
Why not wireless?
Are legacy platforms the only reason not to jump on the Industrial Ethernet bandwagon? Not really; there are many instances where a wireless solution might be your preferred route. Let’s start with a look at Wi-Fi, governed by the IEEE 802.11 standard. There are going to be some issues when used in an industrial environment, so let’s start there.
Most laptops and mobile devices are going to operate at the 2.4 GHz frequency band, which is susceptible to electronic noise generated by motors and other devices common in many manufacturing environments. This frequency is also shared with Bluetooth devices. There are often contention issues that could lead to network interruptions. Wi-Fi operating at 5 GHz offers much better immunity to electronic noise, and there also tends to be a lot fewer contention issues at that operating frequency. The trade-off will be that it will likely require twice as many access points for a given coverage area.
A good application for Wi-Fi in an industrial environment is for mobile data presentation and remote system interactions. Maintenance requirements are a great example of this. Maintenance staff might want to examine diagnostic data for a group of devices, while visually inspecting and manually actuating them, all from the physical device locations versus a tethered operator work station. A tablet application would be perfect for such circumstances.
Many analysts feel that there will be a substantial migration in the very near term from web browsers to apps, in particular for manufacturing system data presentation and systems user interactions. Mobile devices are not just poised to take advantage of this; they are largely driving this shift.
There are many wireless I/O options. Some utilize Wi-Fi, but in addition to the issues above, Wi-Fi access points operate in half-duplex mode and can only service one thing at a time. As such, use of this technology with great amounts of I/O on a large scale is considered unpractical by many. RF technology has long been in use and is a good option for long haul runs, mobile equipment, and tough environments where traditional cabling solutions struggle. New solutions are using 3G and 4G connections to move process information to and from the cloud. Expect this general connection type to grow and evolve in concert with cellular network technology.
And the winner is…
Actually, all of us are winners. Much has been written about the Internet of everything. Information is exploding on the plant floor, and its uses are growing exponentially. There will be security challenges to be certain, but the upside potential for efficiency improvement is enormous. Industrial Ethernet is being embraced as it helps facilitate the convergence of administrative, control, and device networks on one network infrastructure.
As these deployments continue to grow, this growth will be tempered by legacy issues like the ones we talked about. You will need to sort out the cost and timing of legacy upgrades, and until we get them all done, they will continue to factor in our work for years to come. Wireless technologies and cloud integration will continue to expand in the manufacturing arena. Migration from browsers to apps is right in front of us. Things are moving quickly, and for the better. Exciting new technologies are right on the horizon, increasing our capacity as controls professionals to provide innovative and value-packed solutions to our industries and clients. With a nod to the past, and a look to the future, it’s a great time to be in the controls business. Enjoy the ride.
- David McCarthy is president and chief executive officer of TriCore Inc. Edited by Mark T. Hoske, content manager, CFE Media, Control Engineering and Plant Engineering, firstname.lastname@example.org.
See more about