Before changing an inherited facility or design, a written gap assessment against the URS and target regulations gives management and inspectors the same picture: evidence from documents and spot measurements, findings graded by risk, and the fix, cost range and schedule impact of each, plus what can stay.
After more than twenty years with automation vendors and running an EU-GMP EPC contractor, Tony Nguyen saw problems start at the handovers between companies rather than inside any one scope. This note explains why Humiley is organized as one accountable team across design, project management, equipment and maintenance.
EN 1886 class L2 allows an AHU casing 0.44 l/s per m² of leakage at −400 Pa. On a witnessed FAT, casing leakage, fan airflow at design static pressure, filter bypass and the control panel are checked against the approved specification, and both engineers sign the sheet that goes into the equipment file.
300 TR is about 1,055 kW, some 55 kW more than a 1,000 kW offer, enough to use up a 5% design margin. Converting fixes the unit only. Capacity also depends on leaving chilled-water and condenser conditions, so ask every vendor for figures at your design conditions, in one table.
EU GMP Annex 15 (2015) treats three consecutive PPQ batches as a common minimum, with the number justified from risk; the FDA's 2011 guidance names no count and works in three stages. A rationale that serves both draws on commercial-scale variability, parameters near their proven limits, prior knowledge and a statistical target on the CQAs.
A new AHU started during construction loads its coil, drain pan and final filters with site dust that does not vacuum off wet fins. Clean the casing before power-on, seal duct openings, run on temporary pre-filters, fit final filters after the final clean and record it for IQ.
By ASHRAE's occupant table, light bench work releases about 140 W of latent heat per person, roughly 200 g of water an hour. Ten operators add about 2 kg an hour to a room at −60 °C dew point. Size the dehumidifier on peak headcount, including shift changeover, and write gowning assumptions into the design basis.
ISO 14644-3 allows installed filter leak scans by aerosol photometer or discrete particle counter. The photometer is quick but needs roughly 10 to 100 mg/m³ upstream; the counter works at far lower loads. For ULPA and filters you want kept free of oil, choose the counter method at design stage.
Electric reheat is paid for twice, as heat and as the cooling that removed it. A wrap-around heat pipe reheats with no energy input but adds pressure drop: 100 Pa at 5 m³/s and 60% fan efficiency is about 833 W, every running hour. The comparison is a per-unit calculation.
Duct leakage, cascade balance, controls tuning and recovery can only be measured once an AHU is connected on site. A FAT can still speed up those tests by recording clean filter and coil pressure drops, the casing leakage result, fan airflow at design static and the panel software version with its I/O list.
EU GMP Annex 1 (2022) gives WFI circulation above 70 °C as an example of minimizing the risk of microbial growth. Keeping a loop hot drives the design: ASME BPE dead legs of L/D ≤ 2, turbulent flow to the last user point, planned sanitization and sensors on the return. The CCS should justify all of it.
The sequence of operation decides what a cleanroom does on restart, on a trip and in a power cut. QA usually meets it at OQ, when every change costs a software revision and a retest. Walk it line by line with QA and maintenance before panels are built: start order, trips, alarm delays and fail positions.