Control magazine has long recognized the potential impact of Ethernet-Advanced Physical Layer (APL) on the automation sector. Of late, with a cover story by editor-in-chief Keith Larson on the technology itself in June of last year and then in a multi-part series by columnist John Rezabek starting in September.
Keith, John and I were all members and survivors of the “fieldbus wars” 25 years ago, and learned a few lessons associated with introducing new technologies into facilities. One of those lessons is that the new technology needs to work in brownfield facilities with minimal change at both ends of the home run cable—field device and IO card/control system, not just on the cable itself. With few greenfield facilities built around the world in the past couple decades, this was one of the main reasons fieldbus was less successful than it could have been (hindsight is always 20:20).
Recent surveys by HMS Industrial Networks show that 2019 was the first year of decline for new fieldbus nodes, declining by 5% compared to 6% growth in 2018. Fieldbuses in the aggregate now account for 35% of the global market. The same studies confirm that industrial Ethernet surpassed traditional fieldbuses for the first time in 2018 and now has more than 60% of the market. Ethernet-APL, with its ability to reuse the existing installed shielded twisted pair infrastructure (like fieldbus did), has the potential to accelerate the trend to Ethernet to the field.
Meanwhile, the Telecommunications Industry Association TR42 committee responsible for “telecommunications cabling systems” is developing a new TIA-568.5 standard for the link transmission parameters (cable and connectors) for IEEE 802.3cg (APL standard). Many manufacturers have already developed prototypes based on the work being done by that committee. These and a range of APL products were to be shown at ACHEMA in June of this year, but that event has been postponed to April next year, so I expect the members of the Ethernet-APL Consortium will do a virtual event this year. In addition, chipsets will apparently be available in quantity before June, and suppliers will want to make us aware of why all of these new products should be part of your next project.
This “new and shiny” approach certainly piques our interest as automation professionals since most of us are, I suspect, early adopters, at least when it comes to home use of the latest geek toys.
At work, however, we're all from Missouri—the "show me" state—wanting someone else to use the technology for a while and work out all the bugs, thus reducing the risk in my facility. In the work environment, we also need to develop an economic reason or business case for how using this technology will improve operations.
Most facilities already don’t put the data they already have to work, so how is having access to even more data at a higher rate going to make things better? Until we make use of the information provided by these technologies to positively impact the bottom line, they're a drain—requiring additional training, more support tools and introducing added complexity.
Until we can turn that data into information—and identify the killer app Ethernet-APL can do that existing technology can’t—its adoption will be restrained. And by killer app I don't mean doing something that can already be done faster or in a different way (WirelessHART thumbs paralleling a wired HART network come to mind), but something totally novel that couldn't be done before.
Work by OPAF and NAMUR provide the broad framework for how Ethernet-APL could be used to better integrate production facilities in a single IP-based environment. Other industries are also using single-pair Ethernet (SPE, the non-intrinsically safe version of Ethernet-APL), which will help advance Ethernet-APL’s use and adoption.
I certainly believe that Ethernet-APL will have a positive impact on our industry. But we need to identify those killer apps to ensure it reaches its “absolute potential limit,” and doesn’t fulfill “another possible lemon” destiny.