The bumpy road to Ethernet-APL (I)

The Spanish version of this article is available here

The coming of Ethernet-APL technology.

This week, while I was doing some research about the status of Ethernet-APL technology, a couple of interesting thoughts came to my mind, and I wrote them down before being forgotten in the daily workflow.?

So here we go: Ethernet-APL stands for Advanced Physical Layer, and it is one of the most disruptive ideas that have appeared in the last 10 years of technology development on Instrumentation and Control (I&C), especially in the field of Process Automation.?

Ethernet-APL enables I&C users to connect their field devices to control systems using Ethernet while, at the same time, maintaining cabling compatibility with existing IEC 61158-2 H1 fieldbus-based networks i.e., Foundation Fieldbus (FF) and Profibus PA (PA).

In this way, Ethernet-APL provides the process industry with a solution that fully embraces the digitalization models proposed by the IIoT and Industry 4.0 concepts. The ability to reach the field with native Ethernet communications offers numerous advantages (this subject requires a specific article, so I will leave it for another day) and provides end-users with a variety of upgrade paths for the digitalization of their existing installations.?

Well, perhaps not all users, but we will come back to this point later.

Some historical background.

One of the drawbacks of IEC 61158-2 fieldbus networks was that they were only a part of a complete system. They were known as H1 networks, were designed for field use, and were supposed to work in combination with a high-speed backbone network known as H2, which was usually based on the RS-485 standard.

This characteristic of fieldbus networks implied the need for a physical media conversion and a protocol conversion to allow the fieldbus integration into the upper levels of the automation hierarchy. These conversions required the use of a linking device (a specialized gateway) that allowed data exchange between the IEC 61158-2 networks and the upper-level networks across two different physical layers while simultaneously delivering power to the field devices.

A complex additional requirement

One additional requirement from the Process Industry was the ability to work in hazardous areas or locations where there is the possibility of the existence of a potentially explosive atmosphere due to the presence of inflammable gases, mists, or dust in suspension.

This requirement could be fulfilled if the IEC61158-2 fieldbus standard allowed the system either to be installed with explosion-proof protection methods or preferably by certifying that the fieldbus system could be intrinsically safe.

Although the US market likes explosion-proof installations, most of the world prefers the intrinsic safety protection method.

Intrinsic safety is the most secure explosion protection method in existence, and as its name implies its inherently safe because it is designed to work using a level of energy that is lower than the required to cause the ignition of the surrounding atmosphere. The basis of intrinsic safety is energy limitation in the field and that is achieved using an energy limiting device, commonly known as a “barrier”. The natural point where this energy limiting device could be located was initially thought to be the linking device. This meant that the whole fieldbus segment would be intrinsically safe, but this requirement also meant that the available power in the fieldbus segment was dramatically reduced.

Several alternatives were presented over the years to address this issue, but the subject is too extensive to be covered by this article and will be revisited soon. For this article's purpose, it’s enough to know that there is a wide variety of installation methods that can avoid the need for an intrinsically safe linking device, although this option is still available from some suppliers.

The FF way

In Foundation Fieldbus, H2 networks were ditched at an early point of development and the linking device was designed to deliver power to the H1 fieldbus segments (which use MBP or Manchester Bus Powered encoding) and to connect them to the HSE (High-Speed Ethernet) networks, which would employ FF over Ethernet communications to enable all the functionality promised in its conception, such as Control in The Field, interoperability, multivariable devices, the user layer featuring function blocks, and determinism.?

But Foundation Fieldbus was a technology developed by end-users for end-users. The Fieldbus Foundation was in fact a non-profit organization that had created a revolutionary technology that none of the industry’s main Control System suppliers was willing to support. The long-time promise of HSE never came to fruition since the adoption of that technology would have made the Control System’s suppliers' proprietary real-time Ethernet networks unnecessary.

And the use of proprietary real-time network protocols was one of the key factors that these companies employed for differentiation and keeping their customer base engaged.

FF turns into a blind alley.

The Fieldbus Foundation seemed to be doing well until the early 2010s, when it presented interesting technical developments, such as the HSE-RIO (Remote Input/Output) and Foundation for ROM (Remote Operations Management) projects.

But suddenly it ran out of gas. There were several reasons for that: DCS suppliers put their focus on Smart marshalling or smart I/O system technologies instead of fieldbus interfaces, the booming Oil & Gas industry, which was one of the main supporters of the Fieldbus Foundation, saw the oil prices plummet due to overproduction, DCS suppliers were not happy being dependent on third party suppliers of fieldbus infrastructure components, they could make more money using their own proprietary I/O systems, etc.

The accumulative effect of all those issues was that, after 2015, FF was not a thing anymore.

The enemy of my enemy…

As a measure for survival, the Fieldbus Foundation made a joint manoeuvre with the HART Foundation, and both entities merged into the Fieldcomm Group, a new organization that is focused on the development of HART over IP communications and on information presentation technologies such as FDI. They are also members of the Ethernet-APL development team.

This unfortunate chain of events has left FF users without a clear update path to Ethernet-APL technology. It looks unlikely that the Fieldcomm Group may invest in further development of FF technology since the lack of acceptance of HSE has left FF in a state of a de facto orphanage.

The PA way

On the other side of the pond, Profibus PA users face a much more hopeful situation. Profibus PA was created to reach the field using Profibus technology. Therefore, its communication model was based on a Master and Slave model instead of the Publisher-Subscriber model used by FF, so there were no interoperability issues created by either the availability or the lack of control functions in the devices since all function blocks were executed in the controller.?

This conservative approach made Profibus PA simpler to install and configure than FF, and economically more convenient, as the number of devices per segment was typically larger in PA than in FF.

A short tale of two couplers

Profibus PA linking devices were usually called couplers, with each supplier creating its own naming variant. IMHO, the main critique that may be done to the Profibus specs is that although the requirements for a linking device between PA (the IEC 61158-2 side of a Profibus network or H1) and DP (the H2 or RS-485 based side of the network) were present in the specs, there was not a technical description or specification in the standard on how such a device should work.

This fact created a product grey area that was covered by the offerings of a pair of suppliers that developed their own vision of how a segment coupler (Profibus jargon for linking devices) should work. And this event, due to rapidly changing market requirements, quickly evolved into two different methods for data exchange between the field and the control room. They were in fact, two design philosophies that became vastly different as time went by.?

The Siemens approach was based on the device known as the DP/PA Link, which was a segment coupler that worked more like an RIO system for PA devices than a gateway but had a low cost that made it well suited for large PA networks with simple devices.?

That characteristic created a market opportunity for a solution that could work better with large networks of complex devices, where DP/PA links struggled, so the German company Pepperl+Fuchs was happy to fill that gap with its transparent segment coupler technology, which made its gateway solution work like a network proxy between DP and PA segments.

PostScript?

This article covers the first part of the story, the second part will become available next week, where the appearance of Industrial Ethernet probes to be a disruptive force.

Addendum

You will find additional information about segment coupler technology in this 2016 article. It is interesting to view the status quo from 5 years ago.

Mirko Torrez Contreras is a Process Automation consultant and trainer that became fascinated with industrial networks since he first found out that they existed. A lifetime geek, he particularly enjoys the curious blend of technologies employed in those kinds of networks. The people that know him personally have given up trying to cure him of his technology-induced OCD. Not that he cares much about that.

This article has been sponsored by Phoenix Contact. The opinions exposed in this article are strictly personal. All the information required for and employed in this article is of public knowledge.