By Ethan Pretsch, WatchDog Robotics
In pulp & paper, the cost of a fire often comes from the delay between ignition and effective suppression.
If they think about it at all, production managers & process engineers rarely think about fire risk in abstract terms where it actually occurs: across the dry-end, at conveyor drives, dryer sections, bearing housings, press rolls, duct work, and other areas where heat, moving equipment, lubricants, and accumulated fiber exist side by side. And then, of course, there are raw materials, finished goods and dust or paper accumulation.
The concern is not usually a fully developed building fire at the outset. It is the smaller ignition that starts near the process and grows before an effective suppression response begins. That interval matters. It affects equipment exposure, machine uptime, restart complexity, and the overall cost of the event. For that reason, fire protection belongs in the same operating conversation as OEE, maintenance planning, asset preservation, and production continuity.
Fire risk in mills is distributed across dozens of process-related ignition points. Overtemperature events on dryer cylinders, overheated bearings at conveyor transfers, lube oil contacting hot surfaces, and fiber buildup in enclosed areas all create credible ignition scenarios. These are not theoretical conditions. They are part of the operating environment in aging, high-throughput facilities.
The dryer section deserves particular attention. With steam-heated cylinders operating roughly between 100°C and 180°C, it creates a constant thermal environment in which combustible material and ignition sources remain in close proximity. Once accumulated fiber meets a hot surface, or an oil-wetted felt catches a spark, fire moves quickly. But the risk extends across the entire dry-end, including the calender stack, reel, and winder areas where fiber dust accumulation, friction heat, and high-speed paper handling create additional ignition opportunities.
That is one reason industry risk assessments consistently rank Yankee dryers and continuous digesters among the largest sources of business interruption exposure in the mill, behind only the recovery boiler. The ranking reflects repair reality. Once major damage occurs, recovery timelines are often measured in weeks, not shifts.
Gaps in the Defensive Line
Mills already have fire protection. The problem is not the absence of protection. The problem is that the main protection layers often operate at different stages of the event.
Point protection systems (micro-misters) can be effective for narrow, well-defined hazards. Ceiling-level sprinkler infrastructure remains essential for larger fire control and life safety. Firefighting teams are effective, but response times vary. But these layers do not address the same response window.
A standard wet-pipe sprinkler system depends on enough heat building at the ceiling to activate a fusible element. Activation commonly takes two to five minutes from initial ignition depending upon fuel source, ventilation, etc. In a machine room, that is significant time because fire growth is not linear. In those minutes, the amount of combustible material involved can increase rapidly, adjacent surfaces heat up, and flame spread moves exponentially beyond the original source. A localized ignition in a dryer hood or conveyor drive enclosure may begin as a manageable event. A few minutes later, it may involve structural elements, nearby conduit, and a much wider fiber-laden area.
For production managers, this is not a separate issue from the broader downtime challenge. It is a concentrated version of it. Industry benchmarks, including data from the International Society of Automation, suggest that most plants lose 5% or more of production capacity to unplanned downtime. For pulp and paper operations, the daily cost of unplanned downtime is commonly estimated between $200,000 and $600,000, depending on grade, machine width, and market conditions. A single fire event that leads to inspection, cleanup, equipment assessment, and restart can erase hard-won operating gains very quickly.
When suppression solves one problem and creates another
None of this diminishes the importance of sprinklers. Their role is fundamental. But in machine-room applications, the consequences of ceiling-level discharge deserve direct attention.
Steam-heated dryer drums operate under substantial thermal load. When overhead water discharge cools those surfaces rapidly, differential thermal stress can develop across the drum shell and journal areas. Older cast iron components are especially vulnerable to thermal shock. Even where there is no immediate fracture, abrupt temperature cycling can accelerate fatigue in welds, housings, and related components. That can lead to inspection delays, non-destructive testing, and in some cases replacement of equipment with long lead times.
Control systems introduce another layer of vulnerability. Motor control centers and PLC-based drive architecture represent both major capital value and operational dependency. Water ingress into MCC enclosures, whether from direct discharge, overspray, or runoff, can cause insulation breakdown, corrosion, and board-level failure. Even after visible water is removed, trapped moisture can create latent electrical problems that extend downtime well beyond the original flame damage.
In a dryer section, a ceiling-level discharge may release between 35 and 150 gallons per minute per activated head, with multiple heads often operating in a developing event. A targeted autonomous suppression layer is designed very differently. It applies water directly at the confirmed fire location and stops once suppression is achieved. In practical terms, that can be the difference between a localized cleanup and a lengthy equipment recovery.
ARFSS and the Missing Middle Layer
ARFSS, or Autonomous Robotic Fire Suppression System, is gaining attention for a practical reason. It fills the response gap between narrowly focused point protection and conventional sprinkler activation. In process terms, its value is intervening early enough to keep a localized ignition from developing into a broader machine-room event.
These systems combine triple-infrared flame detection with thermal video analytics to identify a developing fire, calculate its 3-D position, direct a precision nozzle, and initiate suppression without waiting for human intervention. The process of detection to extinguishing the fire is typically 12-15 seconds--much faster than a conventional sprinkler activates. For a production manager or process engineer, that difference is not academic. It changes the whole scale of the incident.
ARFSS adoption in the U.S. has been slower than Europe, where several industries and jurisdictions have formally recognized or require autonomous suppression capability in higher-risk applications. In the U.S., NFPA codes have been slower to acknowledge the technology within the existing regulatory framework, meaning domestic deployment is driven not by mandate but by economics. For production managers & process engineers protecting dryer sections and control infrastructure, that distinction rarely matters. When the daily cost of unplanned downtime is measurable and suppression delay produces collateral damage that outlasts the fire itself, the economic case is straightforward.
A Better Layered Approach
For process engineers, the framework is straightforward: identify the ignition scenarios most likely to threaten dryer availability, controls integrity, and machine uptime, then compare that exposure to the response speed and discharge profile of a targeted middle-layer suppression system.
This is not an argument against point protection. It is not an argument against sprinklers. It is an argument against leaving the most expensive part of the event undefended. Point systems handle specific equipment hazards. Sprinklers handle larger building-level events. ARFSS is designed for the interval in between--when a fire is growing, the mill is increasingly exposed, and the cost of delay compounds by the minute. That is the interval that decides whether a mill deals with a localized cleanup or a prolonged outage.
In the dry-end of a pulp and paper mill, the most expensive fire is the one that stops production, damages surrounding systems, and turns a short disruption into a long recovery. A layered protection strategy acknowledges that reality and responds to it accordingly.
Ethan Pretsch is the President of WatchDog Robotics, where they're rethinking fire suppression, and bringing long-overdue innovation to the industry. In his free time, he's in the mountains, travelling, or sharing meals with friends.
