Combined control and safety debate

Ged Farnaby -- Control Engineering, 10/1/2005

Discussion about the acceptability of combining safety and process control functions in a common system has heated up recently. Much of the public debate, however, isn't fueled by end-users, but by suppliers of traditional, multi-modular redundant systems who have a vested interest in their technologies. Some vendors have targeted the end-user community with fear tactics insinuating that your plant will not be "as safe" when combining control and safety onto a single, robust platform.

However, through the use of modern, high-integrity processing techniques, firewalls, and active diagnostics, functional separation of control and safety in a combined environment is ensured and compliant with the international safety standards.

Some of the more common arguments against combining the technologies are addressed here:

"Process controllers can't be used for safety."This statement isn't referring to a controller and I/O modules designed for safety and certified by an internationally accredited certifying body such as TÜV, but rather the use of a basic process control system (BPCS) in a safety application.

"If you're not using a triple-modular redundant system, you're increasing your risk."Having a logic solver that is a triple---quadruple, or quintuple-modular redundant system doesn't mean that your safety function meets the necessary risk reduction requirements. In fact, if you audited installed double-or triple-modular redundant systems, you may find many of the sensors and final element components don't meet the SIL (safety integrity level) requirements for the SIF (safety instrumented functions). This is alarming, as most system failures occur with field devices, rather than the logic solver. Redundancy is for availability, not reliability; all safety systems have some level of redundancy. Triple module redundant systems use redundancy to reduce the probability of a dangerous failure occurring. With newer, accepted technologies, dangerous failure modes can be designed out and near 100% diagnostic coverage provided for system integrity.

"Combining control and safety just isn't good practice."Having dual engineering tools, operator interfaces, additional system components, and doubling lifecycle costs for training and spare parts isn't good practice either. A major point is usually missing from those arguments bashing combined control and safety—with most new system offerings, you don't have to combine control and safety if you don't want to. Most systems that offer the capability to combine process control and safety in the same system, or even the same controller, also have the ability to be autonomous.

Why combine the two worlds of safety and process control? Because end-users now have a chance to reduce costs while maintaining required levels of safety integrity. There also are savings recognized in the project engineering implementation and testing phases. For example, moving I/O points within the same system is a lot less expensive and engineering-intensive than moving them to an entirely different system, when documentation and design issues are considered.

Additional savings can be realized during commissioning, when the entire solution can be pre-tested in a controlled environment, instead of in two isolated entities brought together on-site for the first time. This kind of pre-testing enhances an owner's knowledge base of their chosen technology, thereby increasing the integrity of the overall solution.

Then there are the life-cycle costs. Purchasing, spare parts handling, and training are just a few advantages that end-users recognize as a reason to look at a robust and reliable combined system.