Rotork and Parameter, market-leading global providers of mission-critical infrastructure monitoring systems and flow control and instrumentation solutions, have aligned their respective technologies to improve leak detection, isolation and mitigation in liquid-cooled data centres.

The collaboration is focused on combining Parameter’s real‑time leak detection and monitoring technologies with Rotork’s compact, fast and reliable electric actuators for valve control. Together, these systems detect and isolate fluid leaks in real time, reducing the risk of equipment damage and unplanned downtime. This is essential as artificial intelligence compute demands push infrastructure to unprecedented power densities and heat loads.

Global data centre electricity consumption is on track to more than double by 2030, driven largely by AI workloads. AI racks now commonly exceed 100 kilowatts per rack, far beyond the limits of traditional air cooling.

“Liquid cooling is a non‑negotiable for AI infrastructure, but it also introduces new operational hazards,” said Mike Blazes, CEO at Parameter. “Integrating high‑sensitivity leak detection with fast, automated fluid isolation can mean the difference between a contained incident and a cascading failure.”

When a leak is detected, Rotork’s fast‑acting, lightweight and compact, UL‑certified Hanbay electric valve actuators enable fluid isolation at rack, row or system level, supporting long‑term reliability in liquid‑cooled data‑centre environments. Combined with Parameter’s sensing and control platforms, the solution supports early detection, rapid response and centralized oversight. These capabilities are increasingly demanded by hyperscalers, colocation operators and enterprises.

“Protecting uptime, assets and safety is a critical priority in liquid‑cooled data‑centre environments,” said Tony Vangasse, Head of Strategy, CPI, Rotork. “Our collaboration with Parameter brings together real‑time leak detection and fast‑acting electric valve actuation to support rapid isolation when issues arise, helping reduce risk and maintain reliable operation.”

Industry analysts estimate that 40% of AI data centres globally will adopt liquid cooling this year, heightening risk. Even minor coolant leaks can disrupt operations or damage equipment, costing millions of dollars per rack. U.S. data centres consumed about 176 terawatt‑hours of electricity in 2023. This is forecast to surge to between 325 and 580 terawatt‑hours by 2028.

By addressing leak detection and fluid isolation as a unified system rather than isolated components, Parameter and Rotork aim to help data centre operators scale liquid cooling with greater confidence as AI infrastructure continues to expand.

Find out more about Rotork's Data Centre solutions