Innovations in I/O systems: Rockwell updates the Flex I/O Ex (Snapshots from the near future: Part 3)

Highlights of the Hannover Messe

The bumpy road towards the digital process plant

The current trend towards digitization that is taking over the Industrial Automation world has been somewhat slower in the conservative Process Automation market. The digital fieldbus protocols that seemed to be destined to conquer the process plants (Foundation Fieldbus and PROFIBUS PA), although have managed to became commonplace in certain niche markets (Oil & Gas in the Middle East and South East Asia for FF; Chemical, Food and Beverage, Biofuels and Mining for PROFIBUS PA, which managed to gather a broader range of applications) failed to come close replace the traditional 4 to 20 mA devices, which in the last decade also included HART support in nearly all cases.

After an initial push from Emerson Process Management, other DCS vendors followed the trend offering Foundation Fieldbus support. All of them included options to integrate the standard into their systems but they remained open to other communication protocols as well.

But Foundation Fieldbus adoption was crippled by the fact that no major vendor supported the higher level HSE (High-Speed Ethernet) protocol. This decision shrunk the technology’s feature set sending it into slow oblivion.

Additionally, DCS suppliers didn’t like to depend on third party suppliers for Fieldbus infrastructure, which became a great business for some companies.

The Smart I/O concept

Eventually, Emerson, the main pusher for Foundation Fieldbus technology became the first one to launch the concept of Smart I/O or Intelligent Marshalling or Soft Marshaling, you can call it what you like.

The idea was to use Remote I/O stations connected to the controller via a redundant Ethernet connection (either copper or fiber) and populate this Remote I/Os with modular slots that could be “characterized” as a different kind of IO (DI, DO, AI, AO, Temperature, Pulses, whatever) by inserting to corresponding module. In this way, the resulting field cabinets could be standardized and, due to economies of scale, the total cost of the IO subsystem decreased sharply.

There were a few caveats, of course, DI/DO signal density was low, and you still had to have several types of modules for expansions or replacements, but the idea spread quickly.

In a couple of years, Invensys, Honeywell, and Yokogawa developed their own Smart IO subsystems. Yokogawa chose a hybrid slot model that had optional slot type signal conditioning modules, Honeywell and Invensys went for the software based approach: each IO card featured a fixed number of IO points, and you could make them work as DI, DO, AI, AO, etc. by configuring them via software from the control room.

The demise of IS interfaces for Zone 2/Class I Div 2 applications

Most importantly, in the vast O&G industry, the typical Hazardous Area applications are not classified as Class I Div 1 or Zones 0/1 but are mainly Class I Div2 or Zone 2 kind of business. Smart IO systems are usually rated as Ex nA IIC T4 Gc, Ex nA nL IIC T4 Gc(obsolete), Ex nA [ic] IIC T4 Gc (replacement of the nL method), (ATEX or IECEx) or Class I, Division 2, Groups A, B, C, D, T4 (UL, FM or CSA).

This means that in most cases, these systems do not need intrinsically safe subsystems, i. e. IS barriers or galvanic isolators, thus pushing the price even lower.

Add built-in HART support in the I/O modules and you could get 80% of the Fieldbus functionality for less money. And for DCS suppliers that meant getting rid of third party suppliers (or use them as OEMs, whatever option was cheaper). If intrinsic safety was necessary (20% of the time) then you could use intrinsically safe galvanic isolators from a third-party supplier, or like in Emerson’s example, create an intrinsically safe module that could fit in the corresponding slot. Case closed.

The “good enough” approach

This technology is an example of a “good enough” solution: is not as fancy as Fieldbus, but for most cases is mostly adequate. It enables you to use HART functionality in a simple way (no more need of expensive multiplexers, no messier RS485 cables or serial servers).

The latest adopters of the Smart IO technology approach, Rockwell Automation®, and ABB followed suite this year presenting their solutions at the Hannover Messe.

The 1719 Ex I/O system

Rockwell Automation introduced the 1719 Ex I/O system, which features Analog I/O modules that can work as inputs or outputs and are HART-capable, DI modules with NAMUR input support and DO modules for most typical solenoids. The head station communicates with the controller via redundant EtherNet/IP™ connections and the power supply can be configured to offer N+1 redundancy. Three sizes of backplanes are offered and one extension backplane can be added, so you can create the most suitable configuration for your needs. The Smart I/O concept is applied only for the analog signals, but this arrangement has the advantage of allowing multichannel DI cards and application specific DO cards. In slot based Smart I/O systems there is always the issue of low density DI and DO arrangements since each slot is single channel.

This is the long-awaited replacement for the aging FLEX Ex™ I/O system, and like its predecessor combines I/O cards with intrinsically safe galvanic isolation.

The I/O modules can be plugged in and out under power and, if mounted on adequate housings, can be placed in Zone 2 / Class I Div. 2 environments.

When compared with the old FLEX Ex I/O, the main difference is that the older system could be mounted in Zone 1 or Class I Div. 1, with the help of a Control Net adapter that enabled connections via fiber optics or coaxial cable. The power supply was cumbersome: it was contained in an Ex d housing (Explosion Proof) and featured IS power supply channels. A careful design was needed to correctly configure the I/O modules power supply connections.

The tricky business of Zone 1 /Class I Div 1 RIO system’s installation

The new 1719 Ex I/O system can’t be mounted in Class I Div. 1 or Zone 1 environments directly. There are a couple of ways to achieve this goal:

You could use a field cabinet equipped with a purge and pressurization system, if you are willing to spend the additional costs and complexity.

Or you could use Ex-e/Ex-d or Explosion Proof cabinets, but their cost is considerable.

The main issue is maintenance. Extreme care must be taken to service a RIO system mounted in Zone 1 or Class I Div. 1. I always remember the cryptic warning included in the Siemens ET-200 ISP RIO system manual:

“Death or severe personal injury may result if the following precautions are not taken. Hazardous location zone 1 and zone 2: You are allowed to open the enclosure of the ET 200iSP briefly for the permitted maintenance work.”

This kind of warning discouraged a lot of ET200ISP users from mounting the cabinets in Zone 1 because there is no specification of the exact meaning of the term “briefly”. The same precautions were valid for the old Flex I/O Ex.

“Have we met before?”

The 1719 Ex I/O is great, but if you think you had seen this before the answer is: yes.

The 1719 Ex I/O system started its complicated life in Germany as a CEAG AG product, this company went into multiple reorganizations and acquisitions finally becoming part of the Cooper Crouse-Hinds division of Eaton. An offspring of CEAG AG was the Intrinsic Safety Business Unit created around two smaller companies called Digitable and Apparatebau Hundsbach GmbH. This offspring developed the LB/FB Remote Input Output system in the early 2000’s which was employed by Siemens first and ABB afterwards before each company developed their own solutions.

The LB/FB RIO systems previous incarnation had housings that were the same color of Perry the Platypus (if you haven’t watched the “Phineas and Ferb” cartoons, you are really missing something) and now in its latest reincarnation has become Rockwell’s leap into the Smart I/O trend.

To be honest, to compare the 1719 Ex I/O system with the original CEAG LB/FB RIO system is unfair.

It would be like comparing a 1967 Porsche 911 with a 2017 version: same basic shape, completely re-engineered inside. In other words, a good design can withstand the passage of time, especially if it is constantly upgraded and improved. In fact, that’s what we all have in mind when we talk about German engineering.

So many options…

The point where I think Rockwell fell somewhat short is not a technical one but rather a marketing issue: Rockwell has four series of RIO systems: the 1715 Redundant I/O, the 1719 Ex I/O, the 1734 POINT I/O and the 1794 FLEX I/O. If the company could use the new system for non-hazardous area applications, by offering non-intrinsically safe I/O cards, they could simplify their product line, lower their costs, and simplify customer’s selection process. That was the original idea of the Smart I/O concept.

Maybe a good performance of the 1719 Ex I/O system would pave the way for that approach. Time will tell…

Thanks to Analía Kelhoffer and Luciano Uribarri from Rockell Automation Argentina for their help in proof checking this post.

Mirko Torrez Contreras is a Process Automation consultant and trainer with a keen interest on the evolution of Process Autioamtion I/O systems. He and the AUTEX SA team are looking forward to receive you at the upcoming PROFINET trainings at the PITC/PICC (Profibus International Training Center / Profibus International Competence Center.

Please check the training schedules at www.profibus.com.ar

You can find the first article of the series here, the second one here and the third one here.

This series will continue next week.