Industrial Moisture & Humidity Control

The first task is to write down the total mass of water that will enter the closed volume. The moisture content of the cargo itself, wooden pallets and dunnage, corrugated packaging, cardboard layer pads under the stretch film, and any non-absorbent parts that simply arrive wet are listed separately. Three readings with a resistance moisture meter are enough on timber; below 15% is low risk, above 20% calls for action on its own.

Separating hygroscopic from non-hygroscopic material is critical. Steel parts, castings, glass and ceramics release no water — but their packaging and pallets do. The most common field mistake is deciding “the cargo is dry, no desiccant needed” without counting the packaging. A dry product travelling thirty days on a wet pallet still produces condensation damage, even though the product itself never held any water.

The balance becomes a table: item, mass, moisture content, releasable water. That table is both the input to the desiccant calculation and the only objective document left when damage is disputed afterwards.

The same cargo in the same packaging behaves differently on different routes. What decides it is the lowest surface temperature expected along the way and the size of the day-night difference. On a line that crosses the equator and then heads north, the box first loads up with moisture and then condenses in the cold leg; that two-stage profile is far riskier than a route held at constant temperature.

Deck stowage and under-deck stowage are treated separately as well. A box on deck radiates to the night sky and can fall 4–6 K below air temperature; a box in the hold escapes that radiation but is more poorly ventilated. The target dew point is set at least 3 K below the lowest expected steel temperature.

The moment of loading belongs to the profile too. A box stuffed at night in a humid port leaves with a higher absolute humidity than one stuffed in the morning. The longer the doors stay open, the wider that gap grows; a two-hour stuffing operation can visibly shift the starting point of the desiccant calculation.

The nominal catalogue capacity is not used directly. The 28–35% measured at 25 °C and 90% relative humidity over 30 days typically falls to a band of 18–26% on a voyage spent at an average of 60–70%. It is this reduced value that enters the calculation.

Two sources are summed for a closed volume: the releasable water already waiting inside, and the water carried in by air exchange. The second is the daily air exchange (0.2–1.0 volumes) multiplied by the difference in absolute humidity between inside and outside, scaled over the voyage days. For barrier packaging the calculation runs on desiccant units to DIN 55474 — the quantity that binds at least 3 g of water at 23 °C and 20% relative humidity — but a container is not a barrier bag; it breathes, which is why empirical dosing tables are used there.

The resulting mass is divided by unit weight to give a count. Units of 1–1.25 kg work better than delivering the same total mass in a few large ones: surface area and distribution both increase, and the far corner of the box is covered. The safety margin is not cut below 15–20%; delays, transhipment and port waiting are ordinary risks that stretch the voyage.

In a fixed warehouse the target is written first: 45–60% for general industrial storage, below 50% for metal parts and electronics. There is a lower bound too; going under 35% causes cracking in wooden packaging, curl in paper and board, and static problems in electronics assembly. The target is a band, not a single number.

The building side often buys more than equipment does. To cut capillary rise from the ground, a 150–200 mm layer of washed, single-sized aggregate (16/32 mm) is laid under the slab with a vapour barrier above it; this layer breaks capillary continuity. Insulating the roof deck and the thermal bridges raises night surface temperature and closes the condensation window directly.

Dehumidification equipment is chosen last, because its capacity only means something once the building's leaks are known. Compensating for leaking doors and an uninsulated roof with a dehumidifier is reaching the same target permanently, at a higher energy bill.

The sensor goes where the risk is, not in the middle of the hall: beside the stack leaning against the coldest exterior wall, at product height. In addition, surface temperature is read with an infrared thermometer and compared with the dew point of the air. The decision comes from the difference between those two numbers; a relative humidity reading alone misleads.

What gets logged should be dew point, not relative humidity. Because relative humidity moves with temperature, logs of it are not comparable; dew point represents the mass of water in the air, so two readings taken at different temperatures can be compared directly.

Capacitive sensors drift 1–2 points a year. They are verified annually at three points with saturated salt solutions: LiCl 11.3%, MgCl₂ 32.8%, NaCl 75.3% at 25 °C. Data from an unverified hygrometer carries no evidential weight in a damage file.

Where barrier packaging is used, the desiccant only performs as calculated if the barrier is intact. Aluminium foil laminate has a water vapour transmission rate below 0.1 g/m²·day; a single pinhole or a badly welded seam multiplies that figure many times over and voids the calculation. Checking the seams before closing is more effective than adding units.

With bulk powders the problem is not condensation but caking, which starts once the critical relative humidity is exceeded. Every powder has its own value; for sucrose it sits around 85% at room temperature, and above it the surface dissolves and bridges form. In food and sugar industry silos the target is to stay at least 10 points below the product's critical humidity.

A frequent mistake here is assuming that an inert filler added against caking solves the moisture problem. Flow aids delay the interlocking of particles but do not bind water from the air; unless the humidity target is met, caking is only postponed.

Units are hung on the lashing rings of the side walls, at the level of the upper third of the cargo. They are not spread over the top of the load and not allowed to touch the roof panel: if a condensate drop lands directly on the membrane, that spot saturates locally and the rest of the unit stops working. At least 5 cm of free air is left around each unit.

Distribution matters as much as total mass. The same 12 kg gives a markedly lower equilibrium humidity when spread evenly along both side walls than when piled behind the doors. The door end is the most frequently opened and the most moisture-exposed zone; adding one extra unit there is a reasonable precaution.

The aluminium barrier bag is opened at the last moment; the target is under 30 minutes from opening to sealing the doors. The stuffing record notes the batch number, unit count and total mass, the time the barrier was opened, and the seal number.

As soon as the doors open, the data logger is read first and the cargo inspected second. Then a few randomly chosen units are weighed: the mass gain gives the capacity actually realised. Below 15% gain the dose was oversized; above 30% the units are near saturation and the count must go up next time.

The reacted desiccant contains Ca(OH)₂ and, in part, CaCO₃; it is solid and produces no free liquid. Being alkaline, it is handled with gloves and goggles and its dust is not inhaled. Where local rules allow, it can be put to use liming acid soils; the reasoning behind that use is set out on the agriculture and soil improvement page.

The last step is feeding the numbers back into the next shipment's calculation. After three or four recorded voyages you have your own dosing table per route and no longer need generic ranges. For sizing support with your route, cargo and packaging data, reach us through the contact page.