Mission-Critical Flow Control for Data Centres
As AI and High-Performance Computing (HPC) continue to accelerate digital growth, data centres face mounting pressure to optimise energy consumption, cooling capacity and sustainability. The shift toward higher-density racks and performance-intensive IT equipment is reshaping facility design, with operators striving to reduce Power Usage Effectiveness (PUE) through smarter, more efficient thermal management strategies.
A key enabler of this evolution is the precise, reliable control of liquid flow. From building-level chilled water systems to server-level coolant distribution, automated flow control now underpins the resilience, efficiency and scalability of modern data centres.
The acceleration of liquid cooling
Traditional air-based cooling systems – built around CRAC/CRAH units and airflow management – are reaching their thermal limits.
Today’s CPUs and GPUs generate dramatically higher thermal design power (TDP), driving a widespread transition toward liquid cooling solutions that remove heat faster, more efficiently and with a significantly lower energy footprint.
Most data centres now operate two liquid-cooling circuits:
- Primary circuit: chilled and condenser water connecting server rooms to cooling assets such as centrifugal chillers and rooftop air-cooled units.
- Secondary circuit: circulating water-glycol or refrigerant through Cooling Distribution Units (CDUs) and server racks.
Both circuits rely on precise, rapid fluid modulation to accommodate transient IT loads, minimise energy consumption and maintain system stability. This is where modern electric actuators and networked control architectures are becoming indispensable.
Automating the primary cooling circuit
The primary circuit mirrors mission-critical water systems found in hospitals and industrial plants, where uptime, control accuracy and thermal reliability are non-negotiable. Facilities can contain thousands of valves - yet only a small percentage are automated.
Motorised valves outperform manual alternatives by enabling remote operation, centralised control and real-time optimisation based on full system data. Electric actuators, such as Rotork’s NOAH range, provide precise positional feedback to the Building Management System (BMS) and integrate seamlessly via fieldbus protocols such as Modbus or Profibus. This improves flow accuracy, system responsiveness and operational safety while reducing manual intervention.
For manual valves that remain essential, solutions like chainwheels allow rooftop or high-level valves to be safely operated from ground level—avoiding costly and risky work at height.
Secondary circuits and direct-to-chip cooling
At rack level, cooling must match rapidly fluctuating compute loads. CDUs, underfloor piping networks and server manifolds require highly responsive flow control to maintain optimal chip temperatures without overcooling.
The rise of Direct-to-Chip (DTC) cooling adds further complexity. With up to 42 servers per rack, even a small coolant leak can disrupt operations. Compact, fast-acting electric actuators, such as Rotork Hanbay, paired with leak detection sensors allow facilities to isolate individual servers within seconds. This prevents full-rack shutdowns, protects IT assets and maintains uptime.
Smarter networks, safer facilities
Electric actuators increasingly serve as intelligent nodes within advanced automation architectures. By integrating actuators, sensors and metering devices on shared digital networks, operators gain real-time insight, predictive maintenance capabilities and simplified BMS integration. This modernised approach also extends to ancillary systems, including fire suppression and backup power, where consistent control philosophies enhance safety and response times.
Future-ready cooling through intelligent flow control
As data centres evolve to meet growing thermal demands, automated liquid flow control has become a foundational requirement for reliability, efficiency and sustainability. Intelligent electric actuation – supported by open network architectures, safety-first design and precision engineering – is enabling operators to optimise PUE, future-proof cooling strategies and maintain uninterrupted performance in an increasingly compute-intensive world.
