A panel that passed its factory test has since been transported, lifted and cabled. Before the incomer is closed: inspect inside and check torque marks, measure and record insulation resistance, compare protection settings with the coordination study, prove earth continuity and phase rotation, and confirm tools are out and covers on.
An I/O list with only a tag and a signal type leaves the programmer to choose ranges, alarms and fail states. For each point, give a description and drawing reference, range and unit, alarm limits with delay and priority, the fail state, and a GMP-critical flag with calibration interval. Test scripts then come from it.
Good engineering practice is the design review, installation checks and commissioning that sit under qualification. WHO's HVAC guideline defines it, and EU GMP Annex 15 (2015) allows data from outside the manufacturer's own programs if justified and controlled. Commissioning sheets made with calibrated instruments and signed by the person who measured can then be relied on.
Hold time is the established period a material may be held under specified conditions and remain within specification. WHO's guidance says the study should use the production container, or a simulated one, and the same environmental conditions as the quarantine area. The staging room's temperature and humidity record is therefore part of the claim.
O&M manuals are often accepted unread. Open the manual at one system you know and check that it describes the equipment as built, lists settings by tag, gives each maintenance task an interval and a method, lists spare parts with part numbers, and has fault-finding that starts from the symptom.
Default drive parameters will run an AHU fan but may not restart it after a power dip. Set the motor nameplate data, run the motor identification, set speed limits and ramps, add skip bands where the unit resonates, then decide and test behaviour on power loss, signal loss and fire mode. File the parameter list.
Operators run the plant from BMS graphics, so errors there become operating errors. Check each screen for five things: every value has a unit and a tag, the setpoint sits beside the actual value, alarm colours follow one convention, stale values look different from live ones, and room names match the room codes.
A chilled water coil piped in parallel flow still cools, so the error often survives commissioning. It shows as a valve at 100%, return water colder than design and an off-coil temperature that misses setpoint. Mark inlet and outlet at the factory, check at FAT, and look again before the insulation goes on.
A room pressure loop tuned with every door shut hunts once production starts. Balance the airflows with the loop in manual, stroke the damper, tune from the reference room outward, then open and close doors and record the swing and the settling time. Those records are what the alarm delays should rest on.
Sensible heat in watts equals 0.335 × airflow in m³/h × temperature difference in kelvin, so 10,000 m³/h across 10 K carries 33.5 kW. The constant comes from air density and specific heat. Correct it for density at altitude, and remember it leaves out the latent load.
An ISO 9001 certificate shows that a supplier's quality management system was audited within a stated scope. It does not show that the equipment or service you bought meets your requirement. Supplier qualification assesses what you buy, against your URS, to a depth set by risk: questionnaire, audit or factory hold points.
A work order closed with ticks holds no data. Rewrite each 'check' as a named measurement with its acceptable range and the action to take outside it, and record the as-found value before any adjustment. Start with the assets the criticality assessment ranks highest. The as-found figures become the trend.