Control System Replacement to Improve Production

In migrating to a new automation system or upgrading to a newer version, there’s something very reminiscent of the process of keeping personal computers current. After all, the questions to ask are similar: When do you stop pouring more money into the old-and-getting-older system? When do you gut everything and—budget allowing, of course—replace that with the latest and greatest?

Such a rip-and-replace event occurred with the Epoxy, Phenolic and Coating Resins Division (EPCD) of Columbus, Ohio-based Momentive Specialty Chemicals Inc. (www.momentive.com). Momentive produces a broad range of thermoset technologies and specialty products for global wood and industrial markets.

The company had about 20 projects worldwide requiring attention. All involved the replacement of an original equipment manufacturer’s software and hardware. “Because our systems were so old, we couldn’t get hardware replacements. We also couldn’t find people who could speak the languages of the systems,” says Brooke Robertson, EPCD regional control specialist located in Deer Park, Texas.

Getting past those hindrances started in September 2007, when Robertson joined Momentive. She developed and presented to management a visually based, easy-to-grasp, problem-definition-and-solution process. Next, she began assessing what hardware and software the company had. Two months later, they started looking for the solution that, for most projects, would start at the marshaling boards and move backward into production areas. Then they hired a system integrator. That’s when the engineering work of migration began.

Taking the project global
Once engineering began, Momentive management raised questions that led Robertson and her colleagues to go beyond their one facility project and take a site-by-site look at the problem. “It was then that we thought we should take the solution globally. In 2010, that’s when we made our decision [to do so] in our division.”

But Robertson still faced a big hurdle: finding a technology that could communicate with the company’s disparate systems. Following collaboration with their selected vendor, Emerson Process Management (www.emersonprocess.com), whose control system Momentive would be migrating to, the company chose to install several other technologies to ensure seamless communication throughout the company’s systems.

“We are using their most current-generation control system. And some sites are using the latest SIS [safety instrumented system],” she explains. Some of the other technologies selected include Emerson’s SX series of controllers, the single-channel Characterization Module (CHARM) I/O, and DeltaV Alarm Help. “We’re also going to be some wireless with a small asset management system installation.”  In Europe, Momentive will use the Emerson SIS I/O and control module.

Momentive’s goal remains to migrate technology sensibly to take advantage of new technology improvements, Robertson states. The company now works with Emerson, with which it has global framework agreements, to create site-by-site five-year refresh programs to mitigate any further large one-time expenses. “With this agreement in place, when upgrades are necessary, it doesn’t require nearly the amount of investment we faced initially,” she says. “But that significant investment is already paying off through improved control. If we get just part of the savings we predict, that improvement will be very significant.”

Migration lessons
Through completion of the first two projects, Momentive has learned some profitable lessons. One, it doesn’t necessarily need a system integrator’s help. Two, the process could take nine to 12 months, not years. Three, it’s important to bring potential vendors to the site to demonstrate their technologies.

“The operators were really excited about the migration,” Robertson says, “but a lot of the engineers really were thinking of going in a different direction than we had proposed. Holding an open discussion without vendors present enabled us to settle the do-we-have-to-change-the-culture question.

Two other lessons Robertson noted include:

• “It’s really important that the project team stays as consistent as possible with the vendor and company.”, and

• “Everyone must be clear on the scope, so there are no false expectations.”

The definition-and-solution process as well as the new distributed control system really impressed Paul Barletta, EPCD’s Norco, La.-based vice president of manufacturing. He specifically notes the new level of consistencies and best-practice sharing “we are getting across the globe” as a result of the migration.

But success travels in peaks and valleys. The worst part of the process was its length, Barletta says. “The best part was bringing consistency within our own multiple sites that had multiple legacy systems in place.”

Recipe-driven change
At Rancho Cordova, Calif.-based AMPAC Fine Chemicals (www.ampacfinechemicals.com), the company works with pharmaceutical companies to research, develop and manufacture active pharmaceutical ingredients found in many prescription medications. At any given time, AFC typically manages up to 15 development projects.

Because of a lack of in-house resources to support the company’s legacy Modicon system, in addition to a longstanding relationship with Rockwell Automation (www.rockwellautomation.com), AFC worked with Banks Integration Group (www.banksintegration.com) of Vacaville, Calif.— a Rockwell integrator—to retrofit its older production system. Banks engineer Neil Yates recommended a move to Rockwell Automation’s PlantPAx Logix Batch & Sequence Manager (LBSM).
According to Yates, LBSM is a configurable solution designed for small systems with simple requirements, which allows end users to keep recipe structures and recipes in controllers. A single unit can hold up to 32 recipes, he adds.

“You manage the recipes through one engineering station with LBSM. The operator must use the recipe that is already loaded,” Yates explains. Using modeled normal process parameters, the technology’s built-in modular phases—e.g., temperature, pressure, agitation, raw material addition, etc.—are based on the ISA-88 batch-control standard.

When the project started in December 2009, AFC was using Modicon 984 programmable logic controllers (PLCs). “We totally gutted that and started over,” Yates recalls. AFC’s operations are mostly manual, but had some hard-coded PLCs. Banks removed those codes to provide more flexibility. Commissioning and site-acceptance testing occurred in November 2010.

One benefit AFC’s seen from the new automation system is reproducibility. “Reproducibility is wanting to run batch processes the same every time,” Yates says. Another benefit, though perhaps unintended, is proof of the concept of how AFC could continue to automate. “Management is slowly coming on board with the idea of increased automation, thinking this is the way in the future. I think this is the route they’ll take eventually,” Yates observes.

Beyond increased reproducibility and proving the benefits of automation is the company’s new production flexibility. “The new technology gave them a tool to make their plant easier to operate and push more product. It also forces personnel to think about things up front, rather than thinking about things on the fly.” 

Pumping more crude
The Glas Dowr is a floating production storage and offloading vessel. It has been in operation since August 1997 by Bluewater Energy Services B.V. of Hoofddorp, The Netherlands. Of the ship’s many functions, some include receipt of fluids from subsea wells; control of those wells; processing incoming fluids into crude, water and gas; storage of stabilized crude; and offloading crude into tandem-moored shuttle tankers.

During conversion of the Glas Dowr to prepare the ship for operations on the Kitan Field, some 311 miles north of Australia in the Timor Sea, it was discovered that the ship’s oil-related automation was obsolete. The DCS’s existing cross-wiring and field cables were in bad shape and there was a reasonable risk that the wiring might not work with any new system. Therefore, Bluewater needed a replacement system that could offer at least a 10-year lifetime, says Paul Sep, Bluewater’s section head of the electrical instrumentation department.

Initial contacts with potential vendors and the process of developing new system specifications took about three to four months. It then took another year to select a vendor and about another year to complete the automation project, according to Sep.
Bluewater chose Invensys Operations Management (iom.invensys.com) to upgrade its existing systems and implement a new integrated control and safety system (ICSS). The ICSS includes: Foxboro DCS, Trident Safety System, Tricon Safety System, Foxboro instrumentation (including level, pressure, flow and temperature sensors), and a Wonderware historian.

The new system faced its share of I/O issues since a lot of the existing I/O had to be replaced. Existing I/O also had to be made to fit the new system. “With existing I/O, you can look into the hardware and software, document it and then bring it into the new system,” says Sep, who was the migration’s project lead engineer for control and electrical systems. “But with the new I/O, using in-house engineers, we had to define new I/O and software functionality.”

That new I/O became the project’s biggest challenge, Sep recalls. “We were basically doing as much preparation as possible for software and hardware.” But he knew some information and wiring would come late. “We decided to do some cross-wiring on site, rather than in the factory.”

The new technology has improved uptime and helped meet a critical oil-related milestone. “Typically, uptime in the field was around 95 percent. Now it’s more than 99 percent. That means about 1,600 more barrels per day of crude,” Sep says.

>> Control Room Redesign: Find out how to incorporate human factors and user-centered design. Visit http://bit.ly/awwp044