Service
CRAC Environmental Monitoring & Sensor Systems | Australia
Continuous IoT sensor monitoring for CRAC units, server rooms and data centre environments. Real-time dashboards, threshold alerts and predictive maintenance.
CRAC Services installs and operates continuous environmental monitoring systems on CRAC units, server rooms and data centre mechanical plant. Our IoT sensor networks capture temperature, humidity, airflow, refrigerant health, power consumption and water leak data in real time, feeding a cloud platform that delivers live dashboards, threshold-based alerts, and predictive maintenance insights.
A CRAC unit failure in a data centre can cause thermal shutdown of IT equipment within minutes. Traditional reactive maintenance finds problems after they cause downtime. Continuous monitoring detects the early warning signs (compressor current drift, refrigerant pressure anomaly, supply air temperature deviation) hours or days before failure, giving your team time to act.
Our monitoring architecture combines industrial-grade sensors at the CRAC unit with edge gateways (LoRaWAN or 4G LTE-M connectivity), a cloud time-series platform for data storage and analysis, and output channels including web dashboards, SMS alerts, email notifications and BMS/DCIM integration via BACnet, Modbus or SNMP traps. The system is vendor-agnostic and monitors Vertiv Liebert, Schneider Uniflair, Stulz, Daikin Applied, Mitsubishi Heavy and Climaveneta CRAC equipment.
Scope
Our service
includes
- 01Supply and return air temperature monitoring (precision +/- 0.5C)
- 02Humidity monitoring (relative humidity and dew point at supply and return)
- 03Differential pressure monitoring (filter condition, room pressurisation)
- 04Airflow velocity monitoring (under-floor plenum, supply grille CFM)
- 05Refrigerant circuit monitoring (suction/discharge pressure, superheat, subcool)
- 06Compressor current monitoring (motor load, bearing health via vibration)
- 07Power consumption monitoring (per-unit kW, partial PUE calculation)
- 08Water leak detection (under raised floor, condensate overflow, CHW loop)
- 09Smoke and gas detection (refrigerant leak early warning)
- 10Cloud dashboard with real-time and historical trend views
- 11Threshold-based alert engine (SMS, email, SNMP trap)
- 12BMS/DCIM integration (BACnet/IP, Modbus TCP, SNMP)
- 13Monthly performance reports with efficiency trending
Recommended schedule
Preventative service intervals
01
Continuous (automated)
- Sensor data collection (60-second intervals)
- Threshold alert evaluation
- Dashboard refresh
- Data archival to time-series database
02
Monthly (review)
- Performance report generation
- Sensor calibration check
- Alert threshold review
- Anomaly trend analysis
03
Quarterly (maintenance)
- Physical sensor inspection and cleaning
- Edge gateway firmware update
- Sensor recalibration where required
- Dashboard and alert rule audit
04
Annually (comprehensive)
- Full sensor system health audit
- Cloud platform performance review
- Integration test with BMS/DCIM
- Monitoring strategy review against capacity changes
Equipment
Equipment types
we service
FAQs
Frequently asked questions
What sensors are installed on CRAC units?
Each CRAC unit is fitted with supply air temperature and humidity sensors, return air temperature sensors, differential pressure sensors across the air filter, and a power meter for kW draw. DX units additionally receive refrigerant high-side and low-side pressure sensors, and a CT (current transformer) sensor on the compressor motor. CHW and CDW units receive chilled/condenser water inlet and outlet temperature sensors and a flow sensor on the water loop. All sensors feed an edge gateway mounted near the unit that transmits data to the cloud platform via LoRaWAN or 4G LTE-M cellular connection.
How does the monitoring platform work?
Sensors on each CRAC unit collect data at 60-second intervals and transmit readings to an edge gateway via Modbus RTU or analog 4-20mA signals. The edge gateway aggregates data and sends it to a cloud platform over LoRaWAN (for sites with gateway infrastructure) or 4G LTE-M cellular (for standalone deployments). The cloud platform stores time-series data, evaluates threshold rules, generates alerts via SMS, email and SNMP traps, and serves web-based dashboards with real-time and historical views. Operators can configure custom alert thresholds, view per-unit performance trends, and export data for compliance reporting.
Can monitoring integrate with our existing BMS or DCIM?
Yes. The platform supports BACnet/IP, Modbus TCP, and SNMP output protocols for integration with all major BMS and DCIM systems including Schneider StruxureWare, Vertiv Trellis, Honeywell EBI, Tridium Niagara, and Siemens Desigo. Point lists and tag mapping documentation are provided as part of the integration handover. For sites without a BMS, the cloud dashboard operates as a standalone monitoring interface.
What alerts are configurable?
Every measured parameter has configurable high and low alert thresholds. Common examples include: supply air temperature deviating more than 2C from setpoint, return air (hot aisle) exceeding 35C, relative humidity falling outside the 40-60% range, compressor current exceeding 110% of full-load amps, refrigerant pressure ratio falling outside normal bounds, differential pressure across the filter indicating blockage, and any water leak detection event. Alerts are classified as warning or critical, with different notification channels for each level.
What is the return on investment for continuous CRAC monitoring?
A single unplanned CRAC failure in a data centre typically costs $10,000 to $50,000 or more in emergency callout fees, replacement parts, and IT equipment downtime. Continuous monitoring detects the early warning signs of compressor degradation, refrigerant loss, and airflow restriction days or weeks before failure occurs. Our customers typically see monitoring pay for itself within the first prevented outage. Beyond fault prevention, monitoring data supports energy optimisation (PUE reduction), capacity planning, and compliance documentation for insurance and audit requirements.
How quickly can a monitoring system be deployed?
A standard CRAC monitoring deployment takes 2 to 4 weeks from site survey to live dashboard. Week 1 covers the site survey and sensor specification. Week 2 covers hardware procurement and staging. Week 3 covers sensor installation (typically 2 to 4 hours per CRAC unit, performed during a scheduled maintenance window). Week 4 covers cloud platform configuration, alert rule setup, BMS integration testing, and operator training. For urgent deployments, we can compress the timeline to 7 to 10 days with priority hardware sourcing.
Locations