More strings to the bow

Major changes seem on the cards in the area of field communications, with end users now adopting new wireless, electronic marshalling and fieldbus technologies - as highlighted at Emerson Process Management’s recent European press conference.

Most significantly, perhaps, the event in Berlin showed how companies in the steel and oil & gas industries are challenging the widely held opinion that wireless should never be used in control applications.

In the US, Northstar Bluescope Steel has replaced traditional wired instruments with Emerson’s Smart Wireless technology, based on the IEC 62591 (WirelessHART) standard, to improve furnace control at a steel mill in Delta, Ohio.

The previous hard-wired monitoring network, with hundreds of wiring junctions, had suffered frequent measurement failures due to flying slag, vibration, moisture and temperatures as high as 1,650°C in the mill environment. This was replaced by a network comprising 32 Rosemount wireless temperature transmitters; 28 used for control and four for monitoring.

Temperature control
The transmitters send data to a wireless gateway, which interfaces with the mill’s transformer-regulation and burner-control system. The data is used to control the temperature of cooling panels and water-cooled burners on the mill’s electric arc furnace.

The wireless network is critical to the operation: overheating cooling panels can lead to major furnace damage, with a blown-out panel costing as much as €14,500 to repair, as well as lost production.

The improved temperature control via wireless has allowed the mill to process up to one extra batch - worth as much as €145,000 - per day, reported Rob Kearney, maintenance supervisor at the facility.

“Between nine and 12 measurements per week would fail due to high temperatures or physical damage to sensors, cable or conduit,” said Kearney. “And when a measurement point fails, the furnace must be shut down.”

The furnace’s cooling panels, he added, are now operating consistently at a safe temperature, and there is less maintenance required around the hot furnace shell, where ambient temperatures can be 60°C.

Northstar has said that it now intends to implement a similar wireless system for a second electric arc furnace.

Several major oil companies are using wireless for steam-injection control on more than 2,300 remote wells, Emerson officials said - without identifying the companies involved.

The wells in question must be instrumented quickly to maximise field performance. However, high ambient temperatures, sandstorms, a lack of local power supply and remote locations made traditional technologies unreliable, hard to maintain and often impractical.

Emerson said its wireless technology overcame these challenges quickly and with fewer potential points of failure than wired solutions. The improved control, it added, has also enabled the companies to avoid costly and inefficient over-steaming.

While industry conservatism, plus concerns over security, signal reliability and battery life, still seem likely to slow the progress of wireless in control applications, electronic marshalling promises much more rapid adoption. The technology is being targeted at large automation projects, which often experience costly changes and delays.

Wires carrying signals and data from the field are traditionally screwed to a terminal strip in a marshalling cabinet and are then cross-wired to I/O points on a controller card.

A typical I/O card has eight or 16 channels that are grouped together for the same type of signal, but the field wiring on the marshalling strip has a mix of analogue and digital signals - making it difficult to get the right signals to the right cards. Any mid-project decisions to switch, for example, from a digital to analogue measurement presents a real headache for systems engineers, who sometimes have to completely redesign the marshalling and reroute all the signals.

Likewise, constant additions on the project often mean that an extra controller is required, which can, again, require splitting all the I/Os, rerouting and remarshalling them, and then redesigning the cabling.

To address these issues, Emerson introduced electronic marshalling with Charms in conjunction with the latest version of its DeltaV automation system. Charms I/O cards allow easy connection of field wiring on the terminal strip, while a built-in analogue/digital converting capability allows signals to go directly from the I/O card to up to four different controllers.

This design eliminates the need for extensive rewiring when changing signal type or to another controller and - by providing some extra Charms slots - reduces the need for spares, which are typically specified at levels of around 20% above base requirements.

In Berlin, conference presenters from engineering contractor Fluor and chemicals manufacturer Solvay described how electronic marshalling with Charms could reduce project cost, engineering effort and schedule.

Last year, Fluor carried out a detailed evaluation of how electronic marshalling would have impacted a recent project based on a conventional approach. The study identified potential savings of over $2 million - and four-months earlier delivery - among other benefits (see above panel) on the project to develop new onshore oil, gas and condensate reserves.

“The majority of costs were to do with having someone continually rewiring the marshalling cabinets, which from a client’s perspective has no value,” said Vince Grindlay of Fluor Supply Chain Solutions.

Conflict
Pointing to the conflict within a traditional project approach, Grindlay said Fluor had two clients: a construction client driving for information and then for the hardware; and the internal engineering client trying to finish the design. Construction, he noted, wants the information early, before the design work has been completed, “so you end up having to do rework in the wrong place, at the construction site”.

Conventionally, projects start with an instrument database from P&IDs (piping & instrumentation diagrams), which are then developed throughout the project. At the start of the project, the information required by Emerson, which had 80% of the automation scope, was not fully defined. Preliminary data for system engineering was, therefore, used to meet the schedule milestones.

With electronic marshalling, by contrast, Emerson did not need finalised data until much later in the project. Only projected system I/O information was required before cabinets could be shipped to site, based on calculated system size: adding or changing of Charms I/Os carried no significant cost or schedule impact.

“There is no need to have the hardware in front of me to be able to do integrated testing,” said Grindlay. “I can have a Charms box, standard carrier, a handful of Charms and go and do my integrated testing. This gives flexibility to support construction earlier. I can send all the hardware to site to support construction and installation, and concentrate on the critical areas of finishing the design and on the software.”

As is typically the case, Fluor’s project required the issuing of Modpacks (modification packs) - these define an often extensive block of changes as the client’s requirements become better defined. For its study, Fluor found that Charms would have eliminated the need for Modpack 1. Modpack 2, said Grindlay, did come along, but was executed much more efficiently with a huge reduction in documentation issues.

“Using Charms isn’t going to change the way EPCs receive information,” said the Fluor engineer. “I will still have to get mechanical package data later on because these are long-lead items. However, the flexibility means the impact is less.”

Grindlay went on to forecast that the adoption of Charms would be quicker than other new technologies, such as wireless. “The client recognises the issue of continually rewiring marshalling cabinets. When you talk to them about Modpacks they have all felt the pain and recognise that there is a technology that can solve it.”

Indeed, Fluor is understood to already be applying Charms I/O to support its front-end engineering and design work for Santos Ltd’s Gladstone LNG upstream project in Queensland, Australia - the world’s largest project to convert coal-seam gas to LNG.

Chemicals major Solvay is also adopting Charms technology at its site in Tavaux, France - the chemical group’s largest plant, with an annual production of 1,200 kilotonnes per annum (ktpa), including PVC 280ktpa and chlorine 360ktpa.

Emerson and Yokogawa are the main DCS suppliers for the Tavaux site, where process automation encompasses 15 control rooms and 45,000 I/Os. Adoption of Charms there followed a trial project for the modernisation of a plastics laboratory, involving 200 I/O points.

Test results
“We conducted this beta test in March 2010 to determine if Charms I/O was compatible with the plant’s existing field devices, including many from non-Emerson suppliers,” said Franck Jouault, systems department manager at Solvay Tavaux.

Jouault said the trial confirmed the ease of design and installation of the hardware, as well as a significant reduction in the overall engineering effort. Charms I/O, he added, allowed Solvay to save space and reduce the number of cabinets in the technical room. Other benefits, he said, included “greatly simplified” maintenance and reduced downtime, while Solvay personnel were readily able to understand the philosophy.

Solvay has, therefore, given the green light for Charms on a polyol plant modernisation project at Tavaux, which is due for startup in March. The project, which involves DeltaV and 1,300 I/Os, including 800 Charms and five Charms cabinets, will be more critical, with a safety instrumented system as part of the solution, said Jouault.

With other new projects on the books, Solvay could potentially have 4,000 I/Os with Charms installed at the site by the end of 2012, Jouault concluded.
The emphasis on wireless and electronic marshalling suggested that more conventional I/O technologies such as fieldbus could lose out in terms of attracting investment in future technology developments.

However, here Emerson officials were able to point to on-going developments in the area of control in the field (CIF) with Foundation Fieldbus. Work at the University of Strathclyde, Scotland, involving Shell and Shin Etsu, has shown that CIF can deliver a 30% improvement in control performance with very fast, fast and medium-speed process dynamics. CIF also provided up to three-times higher control loop availability than a DCS.

According to Shell, for simple and cascading loops, CIF can “reduce process controller loading and reduce network traffic, enabling more loops per segment, as well as very fast loop response”.

For its part, Shin Etsu reported: “Foundation Fieldbus CIF with inherent backup capability prevented two incorrect plant shutdowns that would have resulted from communication interruptions.”