A clash found on Tuesday and reported on Friday costs three days. Early in a career the pull is to hold the problem until you have the fix. At Humiley, Integrity means saying it the same day, because the honest date we give clients only helps if it arrives while there is time to act.
A room held at 45% RH ±2% is really at 47% when its sensor has drifted 2% low, and the BMS still shows 45%. Tight bands need sensor accuracy and reference accuracy tighter than the band, calibration near the setpoint, evidence-based intervals and as-found records.
A 5 kW oven releasing its heat into a room needs about 1,500 m³/h of supply air at a 10 K difference, which can exceed what the grade needs. Before asking for HVAC prices, state heat to room in kW per item, which items run together, room exhaust and moisture on each room data sheet.
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.
Airlock door interlocks protect a 10–15 Pa step between grades; a fire needs the doors to open. Specify both together: interlock logic, status lamps, fire-alarm release, fail-safe locks, a logged BMS alarm on every override, and a release tested at FAT, at OQ and periodically.
Plot outdoor, mixed, coil-leaving, after-fan, after-reheat and room states on one psychrometric chart. A reviewer can see a mixed-air point off its line, a coil dew point without margin for the room latent load, over-drying paid for twice, and reheat oversized because fan heat was ignored.
QC labs receive samples from graded production rooms, handle open cultures and run fume hoods that shift the room balance. Each is an air-direction decision that belongs in the Contamination Control Strategy. Put the labs and the sample route on the same cascade drawing before the lab HVAC is sized.
Nearly all electrical power used in a cleanroom becomes room heat. Autoclave door faces, depyrogenation tunnel skin losses, hot loops in the ceiling void, panels installed in the room and FFU motors are rarely on a load template. Ask the process team for connected loads and surface temperatures early.
ICH Q10 expects changes to be evaluated after implementation. Closing a change the day the part is installed skips that step. Define the effectiveness check when the change is approved: the data that shows the objective was met, a period long enough to include the original conditions, and the result that closes it.
A ±5% airflow tolerance cannot be tested until the specification says where it is measured, with which instrument and at which filter condition. Fan inlet, duct traverse and terminal hood readings can differ by more than the tolerance. Write the method in the same sentence as the number.
A night-time 40% THDi reading on a lightly loaded EC fan wall can be 4% TDD at the point of common coupling. IEEE 519-2022 limits TDD, relative to maximum demand current, and the limit steps with ISC/IL at the PCC. Decide on active filtering from a study, not one feeder reading.
At 0.45 m/s, each square metre of Grade A unidirectional airflow moves about 1,620 m³/h. A 6 m² zone needs 9,720 m³/h; the whole 30 m² room would need 48,600 m³/h. Drawing the Grade A boundary tightly around exposed product is an energy decision.