21 CFR 211.42 and 211.46 and EU GMP Chapter 3 ask for adequate ventilation, air control and filtration, but give no air change rate or pressure differential, and no filter class beyond HEPA for aseptic areas. Those come from Annex 1, FDA's 2004 aseptic guidance, ISO 14644 and the designer's justification, recorded in the design basis.
Filter loading can swing AHU resistance from 250 Pa clean to 450 Pa at change-out while airflow must stay constant. Selecting near 350 Pa and checking design airflow at 450 Pa with the drive at full speed avoids both a fan stuck right of peak and an airflow shortfall. The energy difference is modest.
ISO Class 1 allows 10 particles of 0.1 µm and larger per cubic metre; ISO 5 allows 100,000 at that size. At that level the installed leak scan must prove frames and grid, not just the U15 filter, single-digit counts need large sample volumes, and airborne molecular contamination needs its own sampling plan.
A replacement AHU supplying 5% more air, with a flow-controlled return and door gaps leaking roughly as ΔP^0.65, takes a door from 10 Pa to about 19 Pa. Fixed dampers and shared AHUs soften that, but requalification scope stays a change control decision: IQ always, OQ on the cascade, reasons recorded for any stage skipped.
A 50 m² Grade B room needs 10 classification locations under ISO 14644-1:2015, and EU GMP Annex 1 requires them at rest against ISO 5 and in operation against ISO 7, with the smoke study in both states. Whether each run sits in OQ or PQ is for the URS and risk assessment to justify.
Ducted terminal HEPA and FFU ceilings can both reach Grade B or C. A ducted system sets airflow at one AHU fan; an FFU ceiling has many settings to hold, but a single failure stays local. Energy depends on fan efficiency and duct losses, so calculate it for the layout.
Installed HEPA leak tests fail in the report more often than in the room. Check upstream concentration, challenge uniformity, probe size and scan rate, overlapping strokes, the frame seal, instrument calibration and filter serials matched to drawing positions before you sign.
Electrostatically charged filter media lose efficiency as the charge drains in service, while pressure drop barely changes. EN 779:2012 added a discharged minimum for F7 to F9; ISO 16890 rates ePM as the average of initial and discharged efficiency. Buy pre-filters on that rating.
EU GMP Annex 1 expects the final stage of an airlock to be, at rest, the same grade as the cleanroom it leads into. Into Grade B, that means 3,520 particles of 0.5 µm per m³, terminal HEPA supply, its own pressure step and a place in environmental monitoring, not a transfer grille.
5,500 CFM at 250 Pa is about 9,345 m³/h, 6.6% short of a 10,000 m³/h schedule at 400 Pa, and a 20 air-change room gets about 18.7. The static pressure gap shows only on the fan curve, and speeding the fan up for 7% more air costs about 23% more fan power.
ISO 16890 rates filters as ePM10, ePM2.5 or ePM1, and broadly a filter must remove at least 50% of that particle fraction to earn the group. In a GMP air handling unit the pre-filter grade sets how fast the terminal HEPA loads, and so how often it is replaced and leak-tested.
Cleanroom inspectors ask for the same HVAC evidence: qualification report, pressure trends and alarm log, HEPA leak tests, classification and recovery, smoke studies, as-builts and change controls. Index them per room, with document number, revision and location, so the back room answers in minutes.