1. Purpose
Define a controlled process for planned maintenance on a CRAC or CRAH unit while maintaining environmental stability, protecting critical loads and verifying that the cooling system is returned to an approved operating state after maintenance.
2. Scope
Applies to planned maintenance activities involving computer-room air conditioning or computer-room air handling equipment supporting data halls, technical rooms or other critical infrastructure areas.
This example does not replace manufacturer instructions, specialist HVAC procedures, refrigerant-handling requirements, electrical safety rules, water treatment requirements, local regulations or site-specific operational procedures.
3. Preconditions
- ☐ Approved maintenance window confirmed
- ☐ Formal change approval completed
- ☐ Affected cooling zone and supported room identified
- ☐ Current room temperature and humidity reviewed
- ☐ Redundant cooling capacity verified available where applicable
- ☐ No active cooling, leak or environmental alarm affecting the work area
- ☐ Required tools, parts, filters and consumables available before work starts
- ☐ HVAC vendor or qualified personnel confirmed onsite or available
- ☐ Monitoring and escalation channels active
- ☐ Rollback and stop conditions reviewed before execution
4. Responsibilities
| Role | Responsibility |
|---|---|
| Operations Lead | Coordinate maintenance activity, confirm readiness and approve progression between phases. |
| HVAC Vendor / Specialist | Perform technical maintenance according to manufacturer and approved site requirements. |
| Facility Operations | Monitor room conditions, alarms, airflow paths and operational impact. |
| Change Manager | Ensure approvals, notifications, records and closeout documentation are completed. |
| Electrical Responsible Person | Confirm electrical isolation or safety control requirements where applicable. |
5. Execution Procedure
| Step | Room/Tag | Action | Verification | Sign-1 | Sign-2 |
|---|---|---|---|---|---|
| 1 | OPS-BRIDGE | Conduct pre-job briefing with operations and maintenance personnel. | Roles, communication path, stop authority and rollback approach confirmed. | ||
| 2 | CHANGE RECORD | Confirm approved maintenance window, affected unit and scope of work. | Work scope matches approved change and site maintenance plan. | ||
| 3 | BMS / DCIM | Review current temperature, humidity and cooling-related alarms for the affected zone. | Environmental conditions are stable and within site-defined operating limits. | ||
| 4 | CLG-ZONE | Verify available cooling redundancy or alternative capacity before removing unit from service. | Operations Lead confirms sufficient cooling capacity for the maintenance window. | ||
| 5 | CRAC/CRAH-003 | Identify the exact cooling unit scheduled for maintenance. | Asset tag and location confirmed against approved work scope. | ||
| 6 | DATAHALL / CLG-ZONE | Check airflow paths, containment boundaries and access route before work begins. | No blocked airflow path, displaced containment or access restriction observed. | ||
| 7 | CRAC/CRAH-003 | Place affected unit into approved maintenance state according to site and manufacturer procedure. | Unit state confirmed and no unexpected alarm or environmental impact observed. | ||
| 8 | BMS / DCIM | Monitor room temperature and humidity after unit is removed from normal service. | Environmental trend remains stable during the defined observation period. | ||
| 9 | CRAC/CRAH-003 | Perform planned maintenance task according to approved technical procedure. | Maintenance completed by qualified personnel with no unresolved abnormal condition. | ||
| 10 | PIPE / DRAIN / FILTER | Inspect visible drain, filter, airflow and accessible connection areas after maintenance. | No visible leak, blockage, loose panel, abnormal vibration or obstruction observed. | ||
| 11 | CRAC/CRAH-003 | Return cooling unit to approved operating mode according to site and manufacturer procedure. | Unit starts or returns to service normally with expected status indication. | ||
| 12 | BMS / DCIM | Review post-maintenance alarms, environmental trends and unit status. | No unresolved alarm, abnormal temperature trend or humidity concern remains. | ||
| 13 | DATAHALL / CLG-ZONE | Confirm airflow paths, containment and work area are restored after maintenance. | Work area clean, access restored and no material left in airflow paths. | ||
| 14 | CMMS / CHANGE RECORD | Update maintenance record, observations, replaced parts and exceptions. | CMMS or approved record updated and change ready for closeout. |
6. Risk Considerations
Planned cooling maintenance can reduce local cooling capacity and thermal margin. The operational risk depends on current IT load, environmental conditions, cooling redundancy, room airflow management and the duration of the maintenance activity.
- Reduced cooling redundancy during maintenance
- Temperature rise in the affected cooling zone
- Humidity deviation or condensation risk if conditions are not controlled
- Blocked airflow caused by tools, panels, filters or temporary work materials
- Water leak, condensate drain issue or visible pipework concern
- Unexpected alarm or failure to return the unit to normal operation
- Human error caused by unclear scope, wrong unit identification or poor communication
Operational rule: Stop the activity if environmental conditions trend outside site-defined limits, if redundant cooling becomes uncertain, or if the system response differs from the approved plan.
7. Rollback Procedure
Rollback should restore environmental stability and return the cooling system to a known approved state. The exact rollback steps must follow manufacturer instructions and site-specific operating procedures.
- Stop maintenance progression
- Return the affected unit to service if technically safe and approved
- Increase monitoring of the affected cooling zone
- Escalate to HVAC specialist and operations leadership
- Deploy approved compensating cooling measures if available and authorized
- Notify stakeholders if environmental risk or maintenance duration changes
- Record abnormal findings and assign corrective action ownership
If the affected room cannot be maintained within approved operating limits, the condition should be managed as an abnormal or degraded operating state according to site governance.
8. Post Verification
- ☐ Cooling unit returned to approved operating state
- ☐ No unresolved cooling, leak or environmental alarm remains
- ☐ Temperature and humidity trends stable after maintenance
- ☐ Airflow paths and containment restored
- ☐ Work area clean and access paths clear
- ☐ Maintenance record updated in CMMS or approved system
- ☐ Replaced parts, filters or findings recorded where applicable
- ☐ Exceptions escalated and corrective actions assigned
- ☐ Stakeholders informed of completion status
9. Operational Notes
Cooling maintenance should be treated as an operational risk activity, not only as a mechanical service task. Removing a cooling unit from normal service can affect thermal margin, airflow distribution and room resilience.
Real maintenance plans should be aligned with facility design, manufacturer requirements, IT equipment environmental requirements, maintenance history, monitoring strategy and applicable guidance such as ASHRAE TC 9.9 thermal guidance for data processing environments.
A strong MOP should define the approved sequence, required tools and parts, communication path, stop conditions and return-to-service verification before work begins. This reduces delay, confusion and human-error risk during the maintenance window.
Key principles: approved maintenance window, cooling redundancy awareness, environmental monitoring, stop conditions, rollback readiness and verified return to service.
Template provided by © Tom Jensen | tomjensen.no