Modernizing a decades-old coal plant network is one thing. Doing it while consolidating 27 PLCs into a fully redundant PROFINET system, with zero unplanned downtime, is another.
That challenge is still underway at one of Alabama’s largest power plants where Revere is leading a multi-year effort to modernize the facility’s fuel handling system without interrupting operations. With much of the work already complete and more scheduled in the months ahead, the project has already provided valuable lessons in managing such a complex, large-scale upgrade.
Consolidating 27 PLCs Without Shutting Down
Revere had previously partnered with this plant on control panels and a municipal wastewater upgrade, but this engagement is on a different scale entirely. The fuel handling system, essential for keeping coal moving and megawatts flowing, was supported by a patchwork of RX3i PLCs, Emerson Cimplicity SCADA, and Hirschmann switches that had gone largely untouched for years.
Past attempts to add redundancy, such as beginning a Media Redundancy Protocol (MRP) ring, had never been completed. What the plant needed was an upgrade to provide reliability, modernization, and a network architecture ready to support decades of service.
From an engineering standpoint, this job was massive:
- Consolidate 27 PLCs into one redundant PLC rack (plus backup)
- Migrate to a fully PROFINET-enabled network
- Ensure that any failure in one device wouldn’t cascade through the system
From an operational standpoint, it was even more daunting. The plant powers a wide swath of Alabama’s grid, and an extended outage would mean thousands of homes and businesses losing electricity. With the network’s age and complexity, even well-laid plans carried significant risk. The IT department initially preferred Cisco switches, but those couldn’t support PROFINET — a non-negotiable requirement for the plant’s future control architecture.
In addition, the work had to be completed while keeping the plant online and coordinating every move with operators in a facility that runs nearly 24/7 — all without introducing new failure points during the cutover process.
Three Years of Preparation
To address the complexity of the upgrade, the project began with a FEED study and multiple site audits, stretching back nearly three years.
Preparation included:
- Testing miles of fiber optic cable and Cat6 runs without disturbing active connections
- Designing and placing new switch cabinets
- Pre-programming redundant PLCs, PROFINET scanners, and HMI mappings before the first cutover
- Risk-ranking every PLC in the plant
Lena Barnett, Revere’s lead engineer on the project who would remain at the helm from planning through commissioning, visited the plant at least five times before the first planned outage and still found surprises during installation.
“We ran into hidden fiber runs and patch panels no one had documented,” Lena recalls. “You think you’ve got everything mapped, and then during commissioning you find connections you didn’t even know existed.”
“The stacker was another surprise,” she adds. “We assumed it was hardwired into the network, but it turned out to be connected by radio antenna, with intermittent signals. That made integration trickier than expected.”
In addition to mapping the existing network, designing switch cabinets, and testing fiber and Cat6 runs, the team invested heavily in pre-programming all PLCs, PROFINET scanners, and HMI mappings before the first cutover. This pre-work was especially important because the plant’s Emerson PROFINET environment was less familiar to the team than the Allen-Bradley or Siemens platforms they worked with regularly. By working through configurations and testing devices in advance, they reduced the risk of delays or unexpected issues once cutovers began.
A Phased, Risk-Managed Approach
Early in the planning process, the idea of replacing all 27 PLCs in a single 8-hour outage was discussed. While appealing for its simplicity on paper, both Revere and the client recognized that the risk was too great. Any unexpected issue during such a concentrated cutover could have forced extended downtime — something the plant could not afford.“If we had stuck to the eight-hour swap plan, it would have been a huge regret,” recalls Barnett.
Instead, the team developed a phased plan. Each PLC and associated system was evaluated for its operational criticality, the complexity of integration, and the potential impact of downtime. This collaborative assessment between Revere’s engineers, the project manager, and plant operations staff resulted in a risk ranking that determined the sequence of upgrades.
Risk Categories and Examples:
| Risk Level | Example Systems | Explanation |
| Low Risk | Stacker (isolated), conveyor 6 chute (dedicated PLC) | Could be taken offline without affecting coal flow or plant output. |
| Medium Risk | Mod fuels building, trestle, washdown system (foam box, wash PLCs) | Short outages possible; affects support processes, not core production. |
| High Risk | Tripper systems (tripper 4, tripper cars), coal loading systems | Core to plant operation; downtime directly impacts power generation. |
By starting with low-risk systems, the team could refine their process, verify communication and redundancy in a live environment, while demonstrating the upgrade’s reliability to plant staff. This approach built trust on both sides and ensured that by the time high-risk systems were addressed, the methodology was proven and confidence was high.
Every swap had to be timed with plant operations. On some days, Barnett’s team had only a 47-minute window to remove, replace, test, and confirm communication for a PLC before returning the system to service.
That pre-programming effort paid off during live swaps. With most device configurations validated ahead of time, installation windows could focus on the physical changeover and final checks, rather than troubleshooting new setups under pressure.
Extreme care was taken with fiber; one wrong disconnection could ripple across the network. Phasing also meant operators were running some systems on old HMIs and others on the new ones, requiring constant communication across shifts to prevent confusion.
“On some days, the operators could only give us a 45-minute or so window to shut a system down, replace the PLC, and test it before bringing it back online,” says Barnett. “That forced us to be extremely precise — every step had to be ready before that window opened.”
Lessons Learned: The Value of Planning, Trust, and Redundancy
This project so far has brought to light several key principles for executing large-scale control system upgrades in live industrial environments:
- Thorough FEED studies are essential. Multiple site audits and detailed documentation uncovered hidden infrastructure — such as undocumented fiber runs and patch panels — before they could cause unplanned outages.
- Vertical integration improves outcomes. Having the same engineer lead from planning through commissioning ensured accountability, continuity, and deep system knowledge.
- Risk-based phasing works. Sequencing upgrades from low- to high-criticality systems built trust with plant staff and minimized operational risk during cutovers.
From a technical standpoint, the Media Redundancy Protocol (MRP) ring was the single highest hurdle. “I needed to make sure unplugging one cable wouldn’t take down a bunch of scanners,” Lena explains. “Pre-programming all switches, scanners, and the redundant PLC panel before the outage helped, but the sheer number of connected devices made setup complex.”
Other challenges included integrating remote equipment like the stacker, which communicated via radio antenna and produced intermittent signals, and adapting to Emerson PROFINET, which was less familiar to the team than Allen-Bradley or Siemens platforms.
For Lena Barnett, the project has reinforced the importance of persistence: “You just have to keep going — there’s always a way to solve things, even if it doesn’t seem like it at the moment.”
What’s Next?
By mid-project, half of the PLC racks have already been replaced and the plant’s first redundant PROFINET loops are in service. The upgraded sections now operate with full redundancy, meaning a single device failure no longer cascades across the system.
The next outage phase, will target the most critical equipment: the tripper cars that load coal into the plant. The trencher will also be upgraded to PROFINET during this window. These systems carry the highest risk because downtime directly impacts fuel delivery and power generation, but the phased approach has built both confidence and proven methods to manage the complexity.
Looking ahead, additional modernization may follow. Equipment such as the trencher and other fuel handling subsystems are candidates for future PROFINET upgrades once the current high-risk phase is complete. With the most difficult groundwork now behind them, Revere and the client are positioned to carry this reliability-focused architecture across the remainder of the facility.
As this project is still in progress, we’ll share an update as the final phase is completed. In the meantime, explore our other case studies to see how Revere is helping clients modernize complex systems without compromising operations.