Most humidity problems in battery energy storage are not humidity problems. They are surface temperature problems.
A battery cabinet can sit at 55 % relative humidity all week and stay dry. The same cabinet at 55 % can wet its busbars in twenty minutes if a cold front drops the enclosure skin temperature faster than the air inside can respond. Relative humidity did not change. The failure condition did.
This distinction decides how a storage project should specify environmental control, which equipment belongs inside the enclosure, and what the control system should actually measure.
Sealed Electrical Enclosures Do Not Behave Like Rooms
Building HVAC assumes a large air volume, continuous mixing, moderate surface temperature differences, and occupants who notice when something is wrong. A battery cabinet inverts all four.
The air volume is small, often a few hundred litres of free space once cells, racks, busbars and cabling are installed. That air has very little thermal or moisture inertia. A small amount of liquid water — a few grams — represents a large relative change in the enclosure's moisture content, while the same few grams in a warehouse are undetectable.
Surfaces inside the enclosure are not at one temperature. Cells generate heat during charge and discharge. Power electronics generate more. The enclosure skin exchanges heat with outdoor air, and on a clear night it also radiates to the sky, which can pull the skin several degrees below ambient air temperature. The result is a cabinet where the warmest and coldest surfaces can differ by 10 K or more at the same instant, with one shared body of air between them.
Condensation does not care about the average. It happens on the coldest surface.
The Failure Mode Is Condensation, Not "High Humidity"
Water condenses on any surface whose temperature is at or below the dew point of the air touching it. Dew point is an absolute measure — it describes how much water the air actually holds, not how close it is to saturation at its current temperature.
Three worked examples, using standard psychrometric relationships:
| Air condition | Dew point | Any surface below this temperature will wet |
|---|---|---|
| 25 °C, 60 % RH | ≈ 16.7 °C | 16.7 °C |
| 35 °C, 50 % RH | ≈ 23.0 °C | 23.0 °C |
| 30 °C, 80 % RH | ≈ 26.2 °C | 26.2 °C |
Read the middle row again. Air at 35 °C and 50 % RH sounds unremarkable. It will deposit liquid water on anything cooler than 23 °C. In an outdoor cabinet that reaches 35 °C on a summer afternoon and cools to 15 °C before dawn, every internal surface passes through the condensation threshold every single night.
That is the mechanism behind most field corrosion in outdoor electrical enclosures. It is not driven by extreme humidity. It is driven by the daily temperature swing acting on air that was perfectly acceptable when it entered.
Why Sealed Enclosures Still Get Wet
A common assumption is that a high ingress-protection rating solves this. It does not, and the reason is worth understanding.
IP ratings describe resistance to the ingress of solid objects and liquid water from outside — spray, jets, immersion. They describe nothing about water vapour, which is a gas and passes through gaps that stop liquid, and nothing about water that condenses out of air already inside the box.
Enclosures also breathe. When internal temperature rises, the air inside expands and some is expelled past gaskets, cable glands and pressure-equalisation elements. When it cools, the enclosure draws air back in — and the air it draws in is outdoor air at outdoor humidity. Each thermal cycle imports a small quantity of water vapour. When the enclosure next cools below the dew point of its internal air, some of that water is deposited as liquid on internal surfaces and does not leave.
Over hundreds of cycles this accumulates. An enclosure can therefore be simultaneously well sealed against rain and steadily accumulating internal moisture. Sealing slows the exchange; it does not reverse it. Removing accumulated moisture requires either a drying device inside the boundary or deliberate conditioned ventilation.
What Moisture Actually Damages
Naming the failure mode matters, because it determines how tightly the environment must be held.
Insulation resistance and tracking. A liquid film across an insulating surface between conductors at different potentials creates a leakage path. Where the film carries dissolved contamination — salt in coastal sites, dust, flux residue — the leakage current is higher and can carbonise the surface, leaving a permanent conductive track that persists after the surface dries.
Electrochemical migration on printed circuit boards. BMS boards, contactor control boards and communication cards carry closely spaced conductors. Under bias and in the presence of a moisture film and ionic contamination, metal can migrate between adjacent conductors and form dendrites, which produce intermittent faults before producing hard shorts.
Corrosion at terminations. Busbar joints, lugs, and screw terminals rely on stable metal-to-metal contact. Corrosion products at the interface raise contact resistance, which raises local heating, which accelerates further degradation. This is a self-reinforcing loop, and it is a common root cause behind connections that were torqued correctly at commissioning and are hot two years later.
Connector fretting. Repeated micro-motion from thermal cycling and vibration, combined with an oxidising environment, degrades plated connector contacts over time.
Sensor and monitoring drift. Humidity sensors, gas sensors and insulation monitoring devices are themselves affected by condensation. A wet insulation-monitoring device may report a fault that does not exist, or mask one that does. Nuisance trips from insulation monitoring after a cold night are a recognised symptom of an enclosure that is periodically passing below its internal dew point.
None of these failures announce themselves at the moment the water appears. They surface months later as unexplained trips, rising joint temperatures, or a BMS board that fails and is replaced without anyone identifying why.
Cabinet, Room and Container Are Three Different Design Problems
The phrase "battery humidity control" covers three physically different boundaries. Treating them as one problem is the most frequent specification error.
| Outdoor or indoor cabinet | Dedicated battery room | Containerised system | |
|---|---|---|---|
| Free air volume | Small, often under 1 m³ | Building-scale | Intermediate, but densely packed |
| Dominant moisture source | Breathing through the boundary | Infiltration, occupancy, door openings | Breathing plus door openings during service |
| Temperature swing | Large; skin follows outdoor conditions closely | Moderate; building mass damps swings | Large on the shell, moderate in the aisle |
| Equipment format | Compact unit mounted inside the enclosure | Room-scale dehumidifier, ducted or freestanding | Compact or ducted unit integrated into the thermal design |
| Service access | Infrequent, often requires shutdown | Routine | Scheduled site visits |
| Repeat deployment | Usually a standard product across many sites | Usually project-specific | Usually a repeated standard design |
A room can tolerate a freestanding dehumidifier that someone empties or drains. A sealed cabinet cannot: it needs a device that fits the internal volume, drains or discharges without breaking the enclosure boundary in an uncontrolled way, and survives the enclosure's full temperature range without intervention.
This is where refrigerant-cycle equipment often runs out of envelope. Compressor-based dehumidifiers in this catalogue are rated for a 5–38 °C operating range; an outdoor cabinet in a continental climate spends part of the year below that. Enclosure-scale devices intended for this duty are built differently — the semiconductor (thermoelectric) units in the energy-storage range are specified down to −30 °C for exactly this reason.
Ventilation and Dehumidification Are Not Alternatives
These two functions are routinely conflated in storage specifications, and they answer different questions.
Ventilation is a safety function. In lithium-ion installations its purpose is to dilute flammable off-gas so that a concentration capable of deflagration does not accumulate. NFPA 855 frames this quantitatively: continuous mechanical ventilation at not less than 1 CFM/ft² (5.1 L/s·m²) of floor area, or intermittent ventilation initiated at 25 % of the lower flammable limit. Explosion control is then addressed through deflagration venting to NFPA 68 or explosion prevention to NFPA 69, unless UL 9540A test data supports the conclusion that the hazard is not credible for that specific product and configuration.
Dehumidification is a reliability function. Its purpose is to keep internal surfaces above the dew point of the surrounding air so that the failure modes described above do not initiate.
The two can work against each other. Ventilation brings in outdoor air. If outdoor air is humid, ventilation raises the internal dew point and increases condensation risk on cold surfaces — the ventilation system is doing its safety job correctly while making the reliability problem worse. A design that treats them as one system, or assumes that "we have ventilation, so humidity is handled", has not resolved this.
The relationship between the two, and how the required ventilation rate is established, is covered in detail in battery room ventilation and humidity control, which separates the safety case from the environmental case.
Control by Dew Point, Not by Humidistat
If condensation is the failure mode, the control variable should describe condensation risk. Relative humidity does not.
A humidistat set to 60 % RH permits a dew point of 16.7 °C at 25 °C air temperature and 26.2 °C at 35 °C air temperature. The same setpoint therefore permits wildly different absolute moisture contents depending on the enclosure's temperature, which is precisely the variable that changes across a day and a season.
A dew point strategy instead asks: what is the lowest surface temperature this enclosure will reach, and what internal dew point keeps a margin below it?
The practical implementation requires three inputs:
- Internal air dew point, measured directly or computed from a co-located temperature and humidity sensor.
- The coldest relevant surface temperature. In a cabinet this is usually the skin on the shaded or wind-exposed face, or the coldest point of a cooling circuit if one is present. It is measured, not assumed.
- A margin. Sensor tolerance, spatial variation and control lag mean the setpoint must sit below the coldest surface by more than the measurement uncertainty. The margin is a project decision, agreed with the system owner.
Control then holds internal dew point below the coldest surface temperature minus the margin. When the enclosure gets colder, the target tightens automatically — which is the behaviour that a fixed RH setpoint cannot produce.
Sensor placement determines whether any of this works. A sensor in the warm upper zone of a cabinet reports conditions that no cold surface experiences. Placement should reflect where condensation is expected, not where the cable run was convenient.
The physics behind surface-temperature-driven condensation, including how to estimate the coldest surface without instrumenting every panel, is developed further in condensation risk in battery enclosures.
Designing for Repeat Deployment
Most storage products are not one-off installations. A cabinet design is built once and deployed across dozens or hundreds of sites, in climates the original engineering review may not have covered.
This changes the environmental specification in three ways.
The envelope must cover the worst site, not the reference site. A design validated in a temperate climate and deployed in a coastal tropical location faces a higher outdoor dew point, a smaller day-night temperature swing but a higher absolute moisture load, and salt contamination that lowers the leakage current threshold at which tracking begins.
The environmental device must survive without site visits. Drainage that requires manual emptying is not viable across a distributed fleet. Discharge arrangements, filter intervals and failure behaviour need to match the actual service model. What the equipment does when it fails matters as much as what it does when it works: an enclosure device that fails silently leaves the fleet unprotected until the first corrosion-driven fault.
Interfaces must be defined once. Power supply, control signal, alarm contact or communication protocol, mounting pattern and clearances should be fixed at design stage. Retrofitting an environmental device into a deployed fleet is substantially more expensive than integrating it before the enclosure drawing is released.
Projects that expect to repeat should treat environmental control as part of the product definition rather than a site-level accessory.
Information to Prepare for an Equipment Review
A useful review starts from the enclosure and the site, not from a dehumidifier model number. The following inputs allow an equipment route to be evaluated:
- Boundary: cabinet, dedicated room, or containerised system; internal free air volume; construction and insulation.
- Location: indoor or outdoor; geographic region; site elevation; coastal or industrial contamination exposure.
- Temperature range: the full expected internal range including winter minimum, not only the design operating range.
- Outdoor design conditions: summer and winter dry-bulb and coincident humidity or dew point for the site.
- Internal heat sources: cell heat generation profile, power conversion equipment, auxiliary loads, and whether active thermal management is present.
- Air exchange: required ventilation rate and its basis; whether ventilation is continuous or triggered; enclosure leakage characteristics.
- The controlled parameter: whether the project intends to control relative humidity, dew point, or surface condition, and what the coldest relevant surface is.
- Interfaces: available power supply, control and alarm signals, communication protocol, and the monitoring system the device must report into.
- Physical constraints: available mounting envelope, clearances, drainage or discharge route, and service access.
- Deployment model: single project or repeated standard product; expected service interval; expected design life.
- Applicable requirements: the product safety, electrical and fire codes that apply in the destination market, and any customer-specific standard.
Where a value is not yet known, recording it as unknown is more useful than supplying an assumed figure — the unknowns usually determine which questions the review has to answer first.
Discussing a Storage Project
Yakeclimate designs and manufactures industrial dehumidification equipment for complex climate applications, with a focus on agriculture and energy projects.
We co-develop application-specific dehumidification equipment around the operating conditions, interfaces, and integration requirements of the wider project or system.
For a storage enclosure discussion, the inputs listed above are the starting point. Explore energy-storage dehumidifiers, review the energy and electrical environments route, or contact Yakeclimate to review the operating conditions for your enclosure.
FAQ
Frequently Asked Questions
What relative humidity should a battery cabinet be kept at?
Relative humidity alone is not a sufficient specification for an enclosure with significant internal temperature variation. The condition that causes damage is a surface at or below the dew point of the adjacent air. Two enclosures at the same relative humidity can have very different condensation risk depending on their surface temperatures. Where a project does specify relative humidity — often because an existing standard or customer requirement is written that way — the dew point implied by that setpoint at the enclosure's lowest expected temperature should be checked against the coldest surface.
Does a high IP rating prevent condensation inside the enclosure?
No. Ingress protection ratings describe resistance to solid objects and liquid water entering from outside. They do not describe water vapour transport, and they do not address water that condenses out of air already inside the enclosure. A well-sealed enclosure still exchanges air with its surroundings as it heats and cools, and each cycle can import moisture that remains after condensation. Sealing reduces the rate of exchange; removing accumulated moisture requires a drying device inside the boundary or deliberately conditioned ventilation.
If the enclosure has ventilation, is humidity already handled?
Not necessarily, and the two functions can conflict. Ventilation in a lithium-ion installation is primarily a safety measure to dilute flammable off-gas, sized against requirements such as those in NFPA 855. It moves outdoor air into the enclosure. When outdoor air is humid, that raises the internal dew point and increases condensation risk on cold internal surfaces. The safety function and the reliability function should be specified separately and then reconciled.
Can a standard room dehumidifier be installed inside a battery cabinet?
Usually not. Room equipment is sized for air volumes several orders of magnitude larger, assumes an accessible drain or a tank that someone empties, and is typically rated for an operating range that does not extend to outdoor winter temperatures. Enclosure-scale devices are sized to the free air volume, arranged so that condensate is managed without uncontrolled breaches of the enclosure boundary, and specified for the enclosure's full temperature range.
Which is more appropriate for an enclosure, a refrigerant unit or a semiconductor unit?
It depends primarily on moisture load and temperature range. Refrigerant-cycle equipment offers higher capacity but requires internal temperatures within the compressor's operating envelope — in this catalogue, 5–40 °C for the energy-storage refrigerant units. Semiconductor (thermoelectric) equipment offers lower capacity in a compact format and a wider operating range, with units in this range specified from −30 °C. Enclosures that stay warm and carry a meaningful moisture load point toward the refrigerant route; small sealed enclosures exposed to outdoor winter conditions point toward the thermoelectric route.
What data does an equipment review actually need?
At minimum: the enclosure boundary and free air volume, indoor or outdoor location with site design conditions, the full internal temperature range including winter minimum, internal heat sources, the required ventilation arrangement, the parameter the project intends to control and the coldest relevant surface, available power and control interfaces, the physical mounting and drainage envelope, and whether the design will be repeated across multiple sites.
References
- NFPA 855, Standard for the Installation of Stationary Energy Storage Systems
- NFPA 68, Standard on Explosion Protection by Deflagration Venting
- NFPA 69, Standard on Explosion Prevention Systems
- UL 9540A, Test Method for Evaluating Thermal Runaway Fire Propagation in Battery Energy Storage Systems
- IEC 60529, Degrees of protection provided by enclosures (IP Code)
