Clean Room
QA inspection checklist

Envelope & Finishes

1. 1.

   Wall and ceiling panels flush — no gaps
   What to look forWall and ceiling panels (typically PIR-cored, hygienic-faced) flush at joints. No visible gaps. Ceiling installed with manufacturer T-grid or solid-flush system. Penetrations for lights, sprinklers, sensors sealed.
   Acceptable whenPer ISO 14644-4 / EU GMP Annex 1: panel surfaces smooth, non-shedding, easy-clean; joints flush within ±1 mm; gaps ≤ 0.5 mm sealed.
   StandardISO 14644-4:2022 — Cleanrooms and associated controlled environments. EU GMP Annex 1 (pharma). VDI 2083.
   Common defectsPanel joints stepped — particle trapsGaps at penetrations — filtration lossPanel face damaged — fibres / particles shedT-grid not coved — micro-particle trapSealant cracked or wrong type
2. 2.

   All joints sealed — no visible gaps or cracks
   What to look forAll wall-wall, wall-ceiling, wall-floor joints sealed with non-shedding sealant per spec. Coving used at floor and ceiling. Continuous bead, no gaps. Sealant compatible with cleaning agents.
   Acceptable whenPer ISO 14644-4 / EU GMP Annex 1: all joints continuously sealed with cleanroom-grade sealant; coving radius ≥ 25 mm at floor and ceiling. Sealant resistant to specified cleaning agents.
   StandardISO 14644-4:2022. EU GMP Annex 1. ASHRAE Cleanroom Design (HVAC reference).
   Common defectsJoint not sealed — particle entrySealant cracked over time — replaceWrong sealant — degraded by IPA / hydrogen peroxideNo coving — square corner traps particlesSealant smeared — surface fingerprint hard to clean
3. 3.

   Floor finish smooth, seamless, and coved at edges
   What to look forFloor finish (PVC welded, epoxy / polyurethane, terrazzo) seamless across area. Welded joints (PVC) flush; epoxy seamless without joints. Coved up walls 100 mm minimum. Surface flat and slip-resistant per use.
   Acceptable whenPer ISO 14644-4 / EU GMP Annex 1: floor finish smooth, non-shedding, coved at walls (radius ≥ 25 mm, height ≥ 100 mm). Welds full-depth, ground flush. Slip USRV ≥ 35 wet (BS 7976).
   StandardISO 14644-4:2022. EU GMP Annex 1. BS 7976-2 (slip USRV).
   Common defectsWelded joint not flush — abrasion creates particlesNo coving at walls — particle trapFloor lifted at jointsWrong adhesive — failure at chemicalsSlip USRV < 35 — H&S fail
4. 4.

   Pass-through and air locks installed correctly
   What to look forMaterial airlocks (pass-through chambers) and personnel airlocks per design. Interlocked doors — only one open at a time. Pressure cascade maintained when doors operated. UV / wipe-down inside as design.
   Acceptable whenPer ISO 14644-4 / EU GMP Annex 1: airlocks with door interlock; pressure cascade maintained ≥ 5 Pa to lower-grade space; cleaning regime per SOP.
   StandardISO 14644-4:2022. EU GMP Annex 1.
   Common defectsDoor interlock not working — both doors open simultaneously, contaminationPressure cascade reverses on door operationNo UV / wipe-down inside as designedPass-through too small for largest itemSealing gasket damaged

HVAC & Filtration

1. 5.

   HEPA filters installed and integrity tested
   What to look forHEPA / ULPA filters installed in terminal housings (filter ceiling). Filter face DOP / PAO tested for integrity — scan with photometer / particle counter. Bypass leakage at filter housing checked. Filter classification (H13, H14) per design.
   Acceptable whenPer ISO 14644-3:2019 (testing): integrity test with PAO aerosol; pinhole leak ≤ 0.01% (H13) / 0.001% (H14). Bypass at frame ≤ 0.0001 of upstream challenge. Filter classification per EN 1822.
   StandardISO 14644-3:2019 — Test methods. ISO 14644-1:2015 — Air cleanliness classification. EN 1822 — HEPA/ULPA classification.
   Common defectsPinhole found in media — filter not replacedBypass at filter frame — fitting faultWrong filter class (H13 vs H14)Test pressure differential not recordedNo PAO equipment / not calibrated
2. 6.

   Air change rate meets classification

   Record: ACH
   What to look forAir supply volume measured at supply terminals. Air change rate calculated from flow ÷ room volume. Compared with design and ISO 14644 classification target. Recovery time tested per ISO 14644-3.
   Acceptable whenPer ISO 14644-1 / -3: ACH typical — ISO 8 (Class 100,000): 5-48 ACH; ISO 7 (Class 10,000): 60-90; ISO 6 (Class 1,000): 150-240; ISO 5 (Class 100): unidirectional flow ≥ 0.36 m/s. Recovery time per ISO 14644-3.
   StandardISO 14644-1:2015. ISO 14644-3:2019. EU GMP Annex 1.
   Common defectsACH below classification target — particle count failsTotal flow short of AHU duty — leakageRecovery time too long — ACH or filter inadequateTest made at off-occupancy — unrealisticNo design intent ACH target
3. 7.

   Pressure cascade correct between zones

   Record: Pa
   What to look forDifferential pressure between cleanroom and adjacent zones measured (digital manometer at door / port). Higher classification rooms positive vs lower (or negative for containment). Cascade per design. Doors close fully.
   Acceptable whenPer ISO 14644-4 / EU GMP Annex 1: pressure differential ≥ 10-15 Pa between adjacent grades (positive for clean, negative for containment). Pressure maintained at all doors closed; recovery on door cycle.
   StandardISO 14644-4:2022. EU GMP Annex 1. ASHRAE Cleanroom Design.
   Common defectsDifferential below 10 Pa — particles leak from dirty sideReverse cascade — contaminationDoor does not close — pressure lostManometer not calibratedNo record of pressure
4. 8.

   Particle count within classification

   Record: Particles/m³
   What to look forAirborne particle count measured per ISO 14644-1 with calibrated particle counter — number of measurement points and sample volumes per ISO. Counts at design particle size (typically 0.5 µm and 5.0 µm). At-rest and operational measurements per spec.
   Acceptable whenPer ISO 14644-1:2015: particle counts at sample points within Class limits — e.g. ISO 5 (Class 100): ≤ 3,520 particles/m³ at 0.5 µm; ISO 8: ≤ 3,520,000. Sample volume per ISO formula. UCL per Appendix B.
   StandardISO 14644-1:2015. ISO 14644-3:2019.
   Common defectsParticle count exceeds class — system / build failureSample volume too small — high uncertaintyCounter not calibratedOperational counts measured at-rest conditionSingle point sampled — coverage gap

Run this inspection on a phone

The same 8 checks, with the guidance a tap away, photos against each fail, measurements recorded, an NCR raised automatically and a branded PDF at the end. Signed on site. 14 days free.