Feb 21, 2022

Migrating a distributed control system: Modernizing industrial controllers and I/O boards

  • Article
  • distributed control system
  • DCS
  • industrial controllers
  • I/O boards

In a previous blog article, we discussed the challenges and technologies to use when carrying out a major upgrade of human-machine interfaces (HMIs) for a distributed control system (DCS). In this last of three blog articles, we’re topping off our series with the second phase of migrating a DCS: upgrading industrial controllers and input/output (I/O) boards.

  1. A question of performance

    If the technological infrastructure (networking, virtualization servers, HMI consoles) has been properly upgraded for some time, we can assume that the team members operating it have adapted to their new work environment and are going about their daily business smoothly. This is the ideal time to focus on a trickier component, namely upgrading the industrial controllers and I/Os. It’s worth making the investment required for such an upgrade when certain critical conditions that undermine plant performance are present, such as:

    • Repeated stability issues or failures with obsolete electronic components
    • Difficulty sourcing critical spare parts
    • Manufacturers announcing the end of component life cycles
    • New required features

    Given the significant amounts of money and potential impacts associated with such projects, we at BBA recommend carrying out a rigorous budgetary assessment, ideally performed by a neutral and objective agent. During this assessment, many hypotheses need to be validated and properly aligned with the organization’s business objectives:

    Is this the right time to establish or apply the organization’s control system standard?

    • Should we carry out a complete migration of the installation or proceed by sector?
    • Should we replace all existing hardware or only certain types of industrial controllers or I/Os?
    • Have we evaluated all options from our current supplier or other solutions on the market?
    • Can we recover the existing control programs, or do we have to start from scratch?

    The challenges

    The challenges involved in migrating a DCS can be divided into three distinct areas: the required hardware, programming and planned downtime. In general, hardware is the first feature that comes up. From the start, the procurement process requires an assessment of costs for electronic modules, cabinets and their mounting, as well as wiring and installation by a certified contractor.

    A simple one-for-one substitution may not apply in all cases. Some electrical and electronic characteristics of the original I/O boards may not be found in the new generation of components, like:

    • I/O operating voltage
    • NPN vs. PNP transistor I/O output stage
    • Maximum allowable output amperage
    • Maximum component operating temperature

    Using conversion kits can be a good idea, but it also has its share of drawbacks, depending on the case:

    • Presence of built-in fuse vs. external fuse
    • Mechanical constraints, depending on cabinet location

    Moreover, if the installation is significantly large, you should even consider the costs of disposing wiring, electronics and used cabinets to obtain a fair budgetary assessment. All in all, there are many technical details involved in selecting replacement components. While investing in the detailed engineering of electrical and electronic components may seem high, it’s usually amortized when the work is carried out on site and commissioned.

    Software, a key component

    Software is more difficult to gauge in advance, but it shouldn’t be underestimated. In some cases, current control strategies cannot be converted because the programming languages differ so much.

    If available migration options let you keep or convert current control strategies, the project team should take this into consideration. However, based on our field experience, if the control strategies cannot be converted to the desired new technology, this can increase projects costs by 35% to 50%. Human beings could completely rewrite new control programs using the previous version or base them on new functional descriptions. However, this inevitably leads to process errors, which will need to be solved during testing or commissioning, thereby inflating the bill once again.

    Plans and drawings should be updated to reflect various aspects, like migration and construction strategies and future maintenance of the existing system. The engineering should therefore include the implementation and migration strategies, thereby minimizing downtime or even taking advantage of planned downtime to implement the new control system. Additionally, documentation must be prepared to support maintenance teams during the transition from the previous system to the new one. This documentation must also include existing documentation in the programming code. Failure to address this issue may lead to confusion between the old and new systems, especially for electricians and instrumentalists who need to rely on the correct information.

    Despite all the effort put into the conversion, there may still be some disparities between the two systems after the migration. As a result, operational teams will need to become familiar with these new control system behaviours. For example, using remote Ethernet I/O will cause some latency during start-up, given the time required to initialize network components. The same applies to the initial I/O values, which may differ depending on the technology used. In some cases, PID algorithms have a different response from their predecessors, which can affect process control, if not calibrated accordingly.


    Successful upgrading relies on good planning and flawlessly executing the implementation strategy at all project stages. Give our BBA experts a call for any support you may need.

This content is for general information purposes only. All rights reserved ©BBA

Latest publications
See all
think it further
Contact us