Common CRAC Unit Faults & How to Fix Them (Australia)
Common CRAC Unit Faults: How to Identify and Prevent Them
Quick answer
The most common CRAC unit faults are airflow restrictions, refrigerant issues, sensor failures, compressor problems, condenser issues, humidifier faults, and condensate drainage problems. Most are preventable with regular servicing, proper environmental conditions, and timely response to alarms. Catching faults early is the difference between a planned repair and an unplanned outage.
This guide walks through each fault category: what causes it, how to recognise it, and what to do about it.
Why CRAC unit faults are different from regular HVAC faults
A CRAC unit is not a comfort appliance. When it fails, server rooms heat up fast (often within 5 to 10 minutes) and modern IT equipment does not tolerate thermal events. That changes the way faults need to be diagnosed and fixed:
-
Speed matters. Diagnosis time is part of the failure cost.
-
Root cause matters. Patching symptoms gets you back online, but understanding cause prevents the next outage.
-
Reporting matters. Repeat faults should drive design changes, not just repeat call-outs.
With that context in mind, here are the faults we see most often across Australian sites.
1. Airflow restrictions
Symptoms: rising room temperature, high coil pressure differentials, fans running flat-out, hot spots in the rack.
Common causes:
-
Blocked or loaded filters, by far the number one cause across every site we service.
-
Dirty evaporator coils, which restrict heat transfer and airflow simultaneously.
-
Failed or worn EC fans, leading to reduced airflow at full speed.
-
Damper or actuator failures in units with variable airflow.
-
Cable bundles, blanking panel gaps, or floor tile placement disrupting room airflow.
-
Belt-driven blower issues on older units (slipping, broken, or misaligned).
How to prevent:
Quarterly filter inspections, regular coil cleaning, and a structured airflow management programme (blanking panels, brush strips, hot/cold aisle discipline). On older units, consider upgrading to EC fans for both reliability and efficiency.
2. Refrigerant issues
Symptoms: poor cooling capacity, ice on suction lines, abnormal compressor pressures, low-pressure or high-pressure alarms.
Common causes:
-
Slow refrigerant leaks at flare joints, schrader valves, or coil welds.
-
Loss of charge following a recent service or component swap.
-
Wrong refrigerant added during prior maintenance.
-
Restricted expansion device (TXV or electronic expansion valve).
-
Non-condensables in the system (air or moisture).
How to prevent:
Annual leak testing, proper evacuation procedures during any refrigerant work, and using only licensed technicians for refrigerant handling. In Australia, refrigerant work requires an ARC licence and proper documentation, and is non-negotiable.
3. Sensor failures and miscalibration
Symptoms: room temperature appears fine but units do not run as expected, conflicting alarms, humidifier and dehumidifier fighting each other, erratic compressor cycling.
Common causes:
-
Drift in temperature or humidity sensors over time.
-
Failed thermistors or RTDs showing implausible values.
-
Disconnected or damaged sensor cables after rack moves.
-
Condensation or contamination affecting sensor accuracy.
How to prevent:
Calibrate sensors annually as part of preventative maintenance. Use independent monitoring sensors as a cross-check. If your CRAC unit and your environmental monitoring disagree, you need to investigate.
4. Compressor problems
Symptoms: compressor not starting, hard starts, high amperage draw, short cycling, unusually loud operation.
Common causes:
-
Failed start capacitors or contactors, often due to electrical wear or heat.
-
Liquid slugging from poor refrigerant management.
-
Worn bearings or scroll components at end of life.
-
Voltage issues, including supply problems, phase loss, or unbalanced phases.
-
Repeated short cycling wearing the unit out prematurely.
How to prevent:
Annual electrical inspection, thermal imaging of contactors and connections, refrigerant superheat verification, and addressing short-cycling root causes (oversized units, low-load conditions, control hunting).
5. Condenser issues
Symptoms: high-pressure alarms, compressor cutting out on hot days, reduced capacity in summer.
Common causes:
-
Dirty or blocked condenser fins, including leaves, dust, lint, insects, and bird nests, especially in regional Australia.
-
Failed condenser fans, where single-fan failures on multi-fan units often go unnoticed until summer.
-
Coastal corrosion of fins on units near the ocean.
-
Recirculation from poor condenser placement.
-
Glycol contamination or pump issues on water-cooled or glycol systems.
How to prevent:
Pre-summer condenser cleaning is the single highest-value maintenance task in Australia. Add coil-protect coatings on coastal sites, and verify all condenser fans are operating during every service visit. Our DX CRAC servicing programme covers all of these checks across Brisbane, Sydney, and Melbourne.
6. Humidifier and reheat faults
Symptoms: humidity drift outside target range, humidifier alarms, scale build-up visible at service intervals, reheat element failures.
Common causes:
-
Steam canister scale build-up on hard water sites.
-
Failed water inlet solenoids or conductivity sensors.
-
Reheat elements burning out after repeated thermal cycling.
-
Mineral deposits clogging drain lines.
How to prevent:
Replace steam canisters at recommended intervals (do not push them past their used-up state), use treated or RO water where appropriate, and inspect reheat elements during every service visit.
7. Condensate drainage issues
Symptoms: water on the floor, leak detection alarms, musty smells, intermittent humidity issues.
Common causes:
-
Blocked drain lines from biofilm, dust, or debris.
-
Failed condensate pumps or float switches.
-
Improperly sloped drain runs installed during commissioning.
-
Frozen drain pans in low-load conditions with low refrigerant.
How to prevent:
Clean drain pans and lines during every service visit, install drain line tracing tablets where appropriate, and test condensate pumps during preventative maintenance.
8. Controls and communication faults
Symptoms: BMS shows offline, alarms do not propagate, settings cannot be changed, firmware update failures.
Common causes:
-
BMS interface card failures (Modbus, BACnet, SNMP cards).
-
Network configuration issues after IT changes.
-
Firmware bugs (yes, even on critical infrastructure).
-
Battery-backed memory loss on older controllers.
How to prevent:
Maintain a current backup of controller configurations, document network settings, and stay current on firmware advisories from your manufacturer.
A practical fault-response checklist
When a CRAC alarm goes off, follow this sequence:
-
Acknowledge the alarm but do not silence it without understanding it.
-
Check room temperature trends. Is the room actually rising, or is it just an alarm condition?
-
Identify which unit is affected and whether redundant units are picking up the load.
-
Capture the alarm code and history before resetting.
-
Inspect the obvious, including filters, condenser, drain pan, and recent changes.
-
Call your CRAC service provider if root cause is not immediately clear.
-
Document everything for the post-incident review.
The single biggest mistake we see is reset-and-forget. An alarm that comes back is telling you something. Listen to it.
When does a fault mean it is time to replace the unit?
CRAC units are built to last 15 to 20 years, but past 12 to 15 years the economics shift. Consider replacement when:
-
Compressor failure is imminent or has occurred
-
Refrigerant is being phased out (R22, some R407C scenarios)
-
Spare parts availability is becoming limited
-
Energy efficiency is dramatically below modern equivalents
-
Multiple major repairs have occurred in a short window
A modern CRAC unit can be 30 to 40% more efficient than a 15-year-old unit, which often pays for the upgrade within a few years. You can browse our current CRAC unit range for replacement options.
Frequently asked questions
What is the most common CRAC unit fault?
Airflow restriction from blocked filters and dirty coils, by a significant margin. It is also the easiest to prevent.
Why does my CRAC unit keep tripping high-pressure alarms?
Almost always a condenser-side issue: dirty coil, failed fan, or recirculation. Check the condenser first.
My CRAC unit is making strange noises. What should I do?
Do not ignore it. Grinding usually means bearings, screeching often means belts, and rhythmic banging can mean refrigerant slugging. Get it inspected promptly.
Can a faulty CRAC unit damage my servers?
Yes, both directly (thermal events causing component failure) and indirectly (humidity excursions causing ESD or corrosion).
Do you do emergency CRAC repairs in Australia?
Yes. Indigi Power & Cooling offers responsive call-out and emergency support across Australia.
For more questions, see our UPS, battery and CRAC FAQ page.
Get ahead of CRAC unit faults
The cheapest fault is the one that never happens. A structured maintenance plan, calibrated sensors, and an experienced service partner will eliminate the vast majority of CRAC issues before they become outages.
Contact our team about a CRAC service plan, or learn more about our DX CRAC servicing across Brisbane, Sydney and Melbourne.
