Distributed control systems (DCSs) were first introduced in the late ’70s and early ’80s. They used the most current technology available at the time, but computer technology has evolved much faster than the manufacturing equipment being controlled. In most cases, the lifespan of the manufacturing line is at least twice as long as the system controlling it.
As control systems begin to fail in an operational plant, there are decisions that need to be made about the path for upgrades. Should a complete replacement be made? Is there a valid and cost-effective path for replacing parts? Or are the economics of the plant such that it might be best to just buy parts for maintenance from the reconditioned aftermarket and try to extend the installed system until the end of the plant’s lifecycle?
During an early part of my career, while working at Eastman Kodak, I was chairman of an internal committee to test and evaluate DCSs. The goal was to select a single preferred system to use as the standard for the entire company. The big contenders were Honeywell TDC, Fisher Provox, Bailey Net 90, Westinghouse DPF, Foxboro IA and Taylor MOD 300. Each had advantages and disadvantages, and although we tried hard, we were never able to select a single system for every application. Controlling a batch chemical plant was different than controlling a web-based machine. And neither had quite the same requirements as an electrical utility plant. Different plants had different lifecycles, different financial justifications, and different starting points and ultimate goals.
It is true that at some point in time, every DCS will need an upgrade to help leverage more reliability and higher productivity by using newer technologies. But the budget for capex and the cost of these upgrades often limits the options of what can be done. There is a thriving market for used and reconditioned parts to help maintain systems that are beyond the life supported by the manufacturer. Many of the companies that manufactured the earlier DCSs are no longer even in business to support their own products. In some cases, the companies have been acquired by other manufacturers and a migration path is provided. Where this is not the case, third-party suppliers have filled the void.
Most DCS manufacturers have a defined lifecycle for their products. It is generally accepted that Honeywell offers the best set of upgrade and migration options, and there is a large installed base of Honeywell systems that have been in continuous use for more than 30 years. Honeywell also provides a defined migration path for users that do not require complete removal and replacement of the old equipment. ABB, as a second example, breaks the lifecycle for its DCS products into four discrete phases that it calls active, classic, limited and obsolete. During the first three phases of the lifecycle, there are parts available for most essential components. Once declared obsolete, this is no longer the case.
Many plants in the U.S. and, even more typically, overseas are using DCSs now declared obsolete by the original manufacturer. As a result, there is a thriving industry supporting these obsolete systems. Those in support of obsolete systems buy used boards and equipment and recondition components to help meet the needs of these users. Classic Automation is a company that supports obsolete systems, and is our local supplier in Rochester, N.Y. I spoke with Fritz Ruebeck, the company’s founder and CEO, to see what the current trends are in DCS migration from the perspective of the reconditioned parts supplier. Fritz provided us with some insights on current user behavior.
Over time, the first part of a DCS to be upgraded is typically one of the highest-end parts of the system: manufacturing execution system (MES) interfaces, historians and batch managers. Supporting the obsolete computers, operating systems, floppy discs and other interface devices is nearly impossible. Following close behind the high-end computers is the upgrade of operator interfaces, again typically built on obsolete third-party computers, monitors and peripherals. 1980s-style computers are very hard to support at any level.
Remaining in the systems are controllers and I/O subsystems. Many of the controllers have proven to be very robust and reliable. And the same is true with the I/O. These are the devices most commonly supported by third parties like Classic Automation. Though newer boards or components might be used in the refurbishment, the controllers and I/O are sold as direct replacements with original model and part numbers assigned to them; they can be plugged into the original chassis and are plug and play.
Classic Automation has the tools to extract the data from the original controllers and load it on the new replacements for end users. I/O modules and the I/O wiring blocks are typically the most robust part of the original system. They’re often the most expensive to replace as well because of the field wiring they require. Refurbished parts for almost all classic DCSs are typically available. If a plant wants to do an expansion, it is even possible to use matching or compatible equipment to make this happen.
When a manufacturing plant has a longer anticipated lifecycle and a business can be made to upgrade the entire DCS, a well-planned and well-executed migration is important. Clearly, with all the enhancements that use the Industrial Internet of Things (IIoT), artificial intelligence (AI) and other features, there is much to recommend a current state of technology in the control system. This will improve plant reliability as well as provide a more adaptable production platform. Flexibility is a key feature that helps companies make the most of business opportunities. In these cases, it is wise for the plant to engage an outside consulting firm to help quantify the requirements document and help to evaluate the options available that best fit the needs of the plant.
Optimation and many other integrators registered with the Control System Integrators Association (CSIA) employ the experts that can lead and coordinate these efforts. A full evaluation should be made because the system selected could well need to operate the plant for the next 20-30 years. Replacing obsolete components sooner than that should not be necessary if the right choices are made. As the present trend is to establish more manufacturing in the U.S. and to maintain and grow the plants that are in operation, it might be time to review your manufacturing plant and consider all the options available.
Bill Pollock is president and CEO of Optimation Technology Inc., a certified member of the Control System Integrators Association (CSIA). For more information about Optimation, visit its profile on the Industrial Automation Exchange.