Clean AHU cooling coils from the leaving face with low-pressure water or steam and an approved, fully rinsed detergent, after checking the drain. Take the frequency from the coil's pressure drop trend against a commissioning baseline, and record the method, detergent batch, before and after readings and photos of both faces.
Primary-secondary pumping is simple to control, but reverse flow in the decoupler sends warmer water to the coil that sets room dew point. Variable-primary saves pump energy but has to hold each chiller's minimum flow while chillers stage. For a GMP plant, choose the scheme your controls contractor can commission and prove.
EPC moves cost and schedule risk to one contractor at a fixed price; compliance risk stays with the owner under any contract form. If the tender never named who writes and executes the DQ, IQ, OQ and PQ protocols, the lump sum did not include them. Owners who cannot define that yet are often better served by EPCM.
An excursion record that ends with "returned to range" describes the logger trace and leaves the product question open. Assess duration and peak from the load logger, stability data for that product, affected batches, root cause, CAPA and a signed QA decision. In Viet Nam's climate, a 15–25 °C route also needs active cooling.
Annex 1 (2022) expects an aseptic process simulation to imitate routine processing with worst-case conditions. An APS run right after a filter change and rebalance tests a different process. Schedule maintenance away from APS runs and record the HVAC state for each run like the interventions.
EU GDP expects initial temperature mapping of a storage area before use, under representative conditions. Mapped empty, a hot spot is a design finding. Mapped full, it is a deviation with product exposed. Agree what representative means early, plan the seasonal repeat, and remap after significant changes.
A single differential gauge across several AHU filter stages cannot say which stage loaded, so teams change everything or guess. A gauge per stage, marked with the commissioning reading and change limit and kept on the calibration list, allows each stage to be changed on condition. BMS transmitters add a trend.
An FFU array can average 0.45 m/s and still have a corner at 0.30 m/s. Measure each unit on a grid at FAT with fan speed settings recorded, then repeat the grid across the installed array at OQ. A flat map is healthy; a low edge points at speed, plenum, blanking or lights.
EN 1886 classes AHU casing strength by deflection at test pressure: D1 up to 4 mm/m, D2 up to 10 mm/m. On a high-static GMP supply unit, a flexing panel works its gaskets loose and leakage rises. Ask which class was tested, on what model box, and whether your design pressure fits.
A spring isolator with its transport bolt still fitted transmits everything. Remove transport locks, select isolators for the lowest drive speed, keep flexible connections slack, remove rigid bridges such as conduit or drain lines, use inertia bases where needed, and file the balance report and a vibration baseline.
An AHU casing is described by five EN 1886 classes: strength D1–D3, leakage L1–L3, filter bypass up to F9, thermal transmittance T1–T5 and thermal bridging TB1–TB5. A witnessed FAT can test leakage on the real unit; transmittance and bridging come from a laboratory model box, so the model box report belongs with the offer.
A depyrogenation tunnel runs from the wash room into a Grade B filling room, across one or two grade steps, and releases heat next to the Grade A zone. Its own airflow balance and heat release per zone belong in the design basis, and the rooms are balanced with the tunnel at temperature.