Water is essential to many data centre cooling systems, but an uncontrolled leak near servers, power distribution equipment or network infrastructure can quickly become a serious operational risk. Chilled-water pipework, coolant distribution units (CDUs), cooling manifolds, CRAH or CRAC units, condensate drains and roof or plant-room services can all become potential leak sources.

A cable-type water leak detection sensor for a data centre provides continuous monitoring along a defined risk zone. When water contacts the sensing cable, the system can activate a local alarm and transmit a signal to the building management system (BMS), data centre infrastructure management platform (DCIM) or remote alarm panel. This gives the operations team an opportunity to investigate and isolate the source before water reaches critical equipment.

Why Data Centres Need Early Water Leak Detection

Data centres combine high-value electronic equipment with increasingly complex cooling infrastructure. Traditional facilities may use chilled-water CRAH units and condensate drainage, while high-density AI and HPC deployments are accelerating the use of direct-to-chip liquid cooling, rear-door heat exchangers and rack-level CDUs.

ASHRAE guidance for liquid-cooled data centres emphasizes continuous leak detection and containment, including zone-based monitoring around racks and manifolds. A leak detection system should therefore be considered part of a coordinated risk-control strategy that also includes suitable pipe routing, drainage, secondary containment, isolation valves, alarm procedures and preventive maintenance.

Early detection can help operators:

• Reduce the probability of water reaching servers, UPS systems, PDUs or communication equipment
• Identify a cooling-system or condensate problem before it develops into a major incident
• Direct technicians to the affected room, zone or cable location more quickly
• Generate BMS or DCIM alarms for staffed and unmanned facilities
• Support a documented incident-response and preventive-maintenance programme

How Cable-Type Water Leak Detection Works

A water leak detection system normally consists of a controller, a lead-out cable, a water-sensing cable, an end connector and cable-mounting accessories. The sensing cable is routed along the floor or around equipment where leaking water is most likely to travel.

When water bridges the sensing elements, the controller changes to an alarm state. Depending on the selected model, the system may provide a relay contact, an audible or visual alarm, a display and an RS485 communication signal. These outputs can be used to notify operators or initiate an engineered response through the facility control system.

The sensing cable should be installed at the expected leak path or collection point rather than simply near the equipment. Floor slope, kerbs, drip trays, pipe routes, raised-floor penetrations and drainage paths should all be considered during design.

WINTERS LPDA1: Non-Location Water Leak Detection

The WINTERS LPDA1 Water Leak Detection Sensor is a non-location cable-type system. It confirms that water has been detected within the monitored cable zone, but it does not identify the exact distance to the leak.

Key LPDA1 characteristics include:

• 12-24 VDC power supply for the standard configuration
• Selectable normally open or normally closed relay output
• Local audible and visual alarm indication
• Power-on self-test function
• Customisable sensing-cable arrangements to suit the application
• Fixed housing with mounting holes
• ABS and polycarbonate enclosure
• CE certification and a five-year warranty stated in the product brochure

The LPDA1 is well suited to a small server room, individual mechanical room, compact data hall or clearly divided leak-detection zone. Its relay contact can be wired to a BMS input, alarm annunciator or other approved monitoring interface.

WINTERS LPDA2: Positioning Water Leak Detection

The WINTERS LPDA2 Water Leak Detection Sensor is a positioning-type system. In addition to confirming a leak, it indicates where along the sensing cable water has been detected. This can significantly reduce investigation time in a large data hall or an extended pipework route.

Key LPDA2 characteristics include:

• 9-36 VDC standard supply, with 24 VAC available as an option
• Relay output, local alarm and RS485 communication
• Integrated front display for leak-location information
• Positioning sensing cable available in multiple lengths, customisable up to 200 m according to the brochure
• 35 mm DIN-rail controller installation
• Four-core sensing cable using separate conductors for detection and transmission
• CE certification

The LPDA2 is generally the stronger choice for larger data centres, multiple rack rows, long service corridors or liquid-cooling installations where maintenance personnel need a more precise leak location.

LPDA1 vs LPDA2 for Data Centre Applications

Selection factor	WINTERS LPDA1	WINTERS LPDA2
Detection type	Non-location: identifies that a leak occurred in the monitored zone	Positioning: identifies that a leak occurred and indicates its location along the cable
Main outputs	Relay plus local alarm indication	Relay, local alarm, RS485 and front display
Typical data centre use	Small server rooms, individual equipment zones and compact mechanical rooms	Large data halls, long pipe routes, multiple rack rows and liquid-cooled deployments
Controller mounting	Fixed wall or panel mounting	35 mm DIN rail
Communication	Hardwired relay to BMS or alarm interface	Relay and RS485 integration
Cable coverage	Customisable; confirm the required length from the order code and product label	5-200 m options stated in the brochure

Engineering note: The LPDA1 brochure and instruction manual show different cable-length ranges. The final cable length and compatible accessories should therefore be confirmed against the selected order code, product label and latest manufacturer documentation before procurement.

Where to Install Water Leak Detection Cable in a Data Centre

1. Around CDUs and cooling manifolds

Route the sensing cable around the base of coolant distribution units, pump packages, heat exchangers, manifolds, valves and flexible hose connections. These locations contain multiple joints and serviceable components, making them priority monitoring zones.

2. Beneath liquid-cooled racks

For direct-to-chip cooling or rear-door heat exchanger systems, install cable beneath rack connections and along the likely flow path from quick-disconnect couplings. Large halls can be divided into zones, or monitored with a positioning system so the affected rack row can be identified quickly.

3. Around CRAH, CRAC and condensate systems

Install sensing cable around chilled-water CRAH units, DX CRAC units, humidifiers, drain pans and condensate pipework. Pay particular attention to blocked drains, failed pumps, loose joints and low points where water may collect.

4. Under raised access floors

Where raised floors contain cooling pipes or condensate drains, position the cable below the highest-risk services and along natural drainage paths. Keep the cable accessible for inspection and avoid obstructing airflow or creating a trip or maintenance hazard.

5. Along pipework, valves and penetrations

Monitor flange connections, isolation valves, flexible joints, branch connections, pipe penetrations and sections located above or adjacent to IT equipment. Pipework should still be designed to minimise exposure to critical equipment; leak detection is an additional protection layer, not a substitute for sound mechanical design.

6. In UPS, battery, electrical and network rooms

Where water-bearing services, condensate drains or neighbouring wet areas create an exposure, a dedicated leak-detection zone can provide early warning. Sensor placement should be coordinated with electrical clearances, drainage and the facility's emergency-response plan.

Integrating the Alarm with BMS or DCIM

A complete data centre leak-detection solution should define what happens after the sensor alarms. A typical sequence is:

1. The sensing cable detects water.
2. The WINTERS controller activates its local alarm and output.
3. The BMS, DCIM or alarm panel identifies the affected zone or LPDA2 cable position.
4. The operations team receives an alarm and investigates the area.
5. Water is contained, the source is isolated and affected equipment is assessed.
6. The cable is cleaned, dried, inspected and function-tested before the alarm is reset.

A relay output may also be used in an engineered cause-and-effect sequence for pump shutdown or valve isolation. Automatic isolation should only be implemented after a controls and operational risk assessment, because an incorrect shutdown could interrupt essential cooling. Fail-safe logic, alarm priorities, power redundancy and manual override requirements should be documented during design.

Installation Practices That Reduce False Alarms

The WINTERS LPDA1 manual recommends careful cable installation and periodic maintenance. Important practices include:

• Use suitable plastic mounting clips, cable mounts or approved cable ties
• Do not glue the sensing cable, as this may damage it or prevent water contact
• Avoid metal ties, metal filings, grease, lubricants and other contaminants
• Keep the cable away from sharp edges, strong magnetic fields, high-voltage equipment and extreme heat or cold sources
• Do not route the cable directly toward an air-conditioning outlet where condensation may cause nuisance alarms
• Keep the cable surface clean and accessible for inspection
• Identify each cable zone clearly on drawings, at the controller and in the BMS or DCIM

The manufacturer recommends servicing the LPDA1 periodically, at least every six months. After an alarm, the area should be inspected promptly, standing water removed and the sensing cable checked before returning the system to service.

Commissioning Checklist

• Verify the controller supply voltage and polarity
• Confirm cable continuity, end connector and controller self-test
• Apply a controlled amount of water at representative points along each zone
• Confirm the local alarm, relay state and RS485 indication where applicable
• Verify the correct BMS or DCIM alarm description, location and priority
• Test any approved valve or pump interlock sequence
• Record the cable route, zone name and test results in the commissioning documents
• Train operations personnel in alarm response, cleaning and reset procedures

Important Fluid-Compatibility Consideration

The WINTERS literature describes the LPDA1 and LPDA2 as water leak detection sensors. Data centre technology-cooling systems may use treated water or water-glycol mixtures. Detection performance can vary with fluid conductivity and composition, while the LPDA1 manual advises against corrosive, reactive, oily and chloride-solvent exposure. Compatibility and alarm response should therefore be verified with the manufacturer using the actual coolant before the system is specified.

Conclusion

A well-designed data centre water leak detection system can identify a small leak before it develops into a major incident. The WINTERS LPDA1 offers straightforward zone-based detection and relay integration, while the LPDA2 adds leak positioning, RS485 communication and a display for larger or more complex facilities.

The most effective solution combines the correct sensor model with risk-based cable routing, BMS or DCIM integration, alarm-response procedures, periodic maintenance and documented testing. When these elements are coordinated, water leak detection becomes an active part of data centre resilience rather than a standalone alarm device.