When a DX compressor cycles off and the fan keeps running, the water film left on the coil re-evaporates into the supply air, so net dehumidification falls short of the rating at part load. Cleanrooms cannot cycle their fans, so they need modulating or staged capacity, or chilled water.
At 20 air changes an hour, a material airlock needs about 14 minutes for a 100:1 particle drop after the outer door closes. The interlock needs a hold-open timer and alarm for trolleys, and the countdown should start at door closure. The gowning airlock is laid out around people instead.
A 50 mm clash tolerance hides a duct running 40 mm into a beam. Pipes and ducts modelled at bare size leave no room for insulation, and trade pairs that were never run never appear. A zero-clash report should state its tolerance, the pairs tested and what each model leaves out.
Heating 18 °C, 60% RH air to 22 °C drops the reading to about 47% RH while the air keeps 7.7 g of water per kilogram. A room that meets RH only on reheat has no margin for one fault. Trend reheat output beside room RH: rising reheat means the coil is losing latent capacity.
Cartons and labels from an 80% RH warehouse release moisture in a 20% RH packing room. If board gives up 2% of its weight, a 500 kg delivery adds 10 kg of water. Stage packaging in a controlled area, unwrapped, for a time set by weighing sample cartons.
A shared exhaust duct spreads a solvent's hazardous zone, cakes process dust in humid air, carries potent-compound contamination through the building if joined upstream of the HEPA, and moves fume-hood face velocity with other dampers. Draw each stream with its hazard, treatment and fan first.
A failed acceptance criterion and a protocol step not executed as written are different OQ deviations. The first needs root cause, impact assessment, a fix and a repeat test; the second needs an assessment of whether the result still stands. Changing a criterion after a failure invites audit questions.
Air leaving a coil at 12 °C and 95% RH reads about 50% RH at 22 °C with the same 8.3 g/kg of water. A duct RH sensor after the coil works at the top of its scale. Control the coil on dew point or leaving-air temperature, and keep duct RH after reheat as a wet-filter alarm.
A 3 m/s supply jet near a pressure tap can shift the reading by up to about 5 Pa, down or up depending on its angle, which is a third to half of a 10–15 Pa cascade. Tap position, tubing without low loops, matched tube lengths and a still reference port decide whether a room pressure trend is trustworthy.
A 100 cm² door gap at 15 Pa passes about 110 m³/h, but only doors on the suite boundary cost fresh air. Internal door leakage returns through the neighbouring room's grille. For twenty doors with four on the boundary, the leakage to make up is 440 m³/h rather than 2,200 m³/h of wet tropical air.
Cooling tower leaving water equals wet-bulb plus approach, and each kelvin of condenser water costs compressor power. At a 2–3% per kelvin rule of thumb, one extra kelvin of approach on a chiller drawing 180 kW is about 4.5 kW, or roughly 36,000 kWh over 8,000 running hours.
On a two-wire Pt100 each ohm of lead resistance reads as about 2.6 °C, and 30 m of 0.5 mm² copper is about 2 Ω. Three- or four-wire connection or a head-mounted transmitter removes it. Only then does IEC 60751 class matter: ±0.40 °C for B and ±0.19 °C for A at 20 °C.