A dry room is not simply a workplace with low humidity

In lithium-ion battery manufacturing, a dry room is a production environment where airborne moisture is reduced to an extremely low level to handle moisture-sensitive electrodes and electrolytes. A battery manufacturing facility at Oak Ridge National Laboratory described in U.S. Department of Energy material operated a dry room for cell assembly and electrolyte filling at approximately 20°C, a dew point of -56°C or below, and relative humidity of 0.1% or below. This does not mean other battery processes must use exactly the same figures, but it is clearly an environment far outside the operating range of ordinary indoor humidity sensors.

The purpose of a dry room is primarily product quality and process stability. Moisture enters through raw materials, worker entry, door opening, process equipment, and outdoor-air leakage. Rather than monitoring only one point at the dehumidifier outlet, locations with differing moisture loads—such as work areas, electrolyte filling points, material entry points, airlocks, and ventilation return ducts—must be monitored separately. Even if the average meets the criterion, locally wet areas can form.

Why dew point matters more than relative humidity

Relative humidity is the ratio of moisture in the air to the amount air can hold at saturation at its current temperature. When temperature changes, relative humidity changes even with the same amount of water vapor. Dew point, by contrast, is the temperature at which condensation begins when air is cooled, making it more direct for interpreting the absolute moisture condition of dry air. Under extremely dry conditions, relative-humidity values are compressed toward 0%, making small differences difficult to distinguish reliably. Dry-room control specifications are therefore normally managed not only by relative humidity but also by dew point or frost point.

When reviewing sensor specifications, it is not enough to check only the lower display limit. Measurement range, accuracy in the relevant range, repeatability, response time, and long-term drift must all be considered. For example, even if a range is stated as extending to -70°C, if accuracy near the actual control point is ±2°C, alarm margin and process tolerance must be designed more broadly than that. A sensor displaying one more digit does not mean it is more accurate.

Criteria for selecting moisture instruments

First, secure a measurement lower limit sufficiently below the expected normal dew point. The range should include not only the normal operating target but also commissioning, dehumidifier performance testing, and restart conditions after a process shutdown. Second, check recovery time. Even if an extremely low-humidity sensor quickly detects a humidity spike caused by worker entry or door opening, it can distort the decision to resume production if it continues to show a high value for several hours as the space dries again. Both rising and falling responses must be reviewed.

Third, assess chemical resistance. Battery processes may contain solvent vapors, electrolyte constituents, and cleaning agents. Manufacturer documentation should confirm that exposure of sensor elements, filters, seals, and cable glands to these materials does not cause sensitivity loss, swelling, or embrittlement. Automatic calibration or sensor-purge functions may help long-term stability, but they do not mean contamination can be ignored.

Fourth, include installation locations in the specification. Wall-mounted sensors that measure a representative indoor value and process-control sensors in dehumidifier ducts differ in air velocity, pressure, temperature, and accessibility. In ducts with fast airflow or pressure changes, allowable air velocity and pressure, as well as insertion depth, must be reviewed. Indoor sensors should avoid workers’ breathing, door jet streams, and locations immediately beside local exhaust, and should be distributed across multiple points so spatial variation can be identified.

Gas sensors have separate humidity limits

The fact that a dry-room dew-point sensor operates at extremely low humidity does not mean every gas sensor in the same space will operate normally. Electrochemical gas sensors are affected by the moisture balance of their internal electrolyte and have product-specific allowable humidity ranges. The Honeywell iSeries safe-use instructions state continuous-operation humidity ranges of 5~95%RH non-condensing for some CO, H2S, and O2 sensors, and 15~90%RH non-condensing for SO2 sensors. The combination of a -56°C dew point and 20°C may correspond to relative humidity much lower than this.

Therefore, a sensor must not be selected solely by the name of the gas to be detected. Normal and abnormal temperature and humidity, cross-sensitivity, oxygen concentration, expected gas concentration, response time, recovery time, and the possibility of sensor poisoning must be reviewed together. Directly exposing a sensor in a dry room when its manufacturer does not guarantee continuous operation at extremely low humidity can cause zero drift, sensitivity loss, and slow response. Even if test gas produces a normal response, that does not prove performance after long-term exposure to extremely low humidity.

There are three options: use a detector verified for direct exposure to extremely low humidity; place the sensor in a suitable environment outside the hazardous area and transfer a sample; or establish a conditional calibration and correction method with the manufacturer. In every case, decisions should be based not on the catalog’s general measurement range but on the environmental performance and certification scope of the completed detector.

Pretreatment design for sampling detection

Remote sampling protects sensors and makes maintenance easier, but adding piping immediately introduces new measurement errors. Sample-line length and internal diameter, and pump flow rate, determine transport delay. Leaks draw in outside air, dilute concentrations, and alter the dry room’s low-dew-point condition. Filters block particles and droplets but can adsorb the target gas or delay response. Tubing material must also be selected based on solvent-vapor adsorption and chemical compatibility.

Preventing condensation is important, but it is not simply a matter of applying heat mechanically. The wall temperature must not fall below the sample’s dew point at any point along the line, and the effect of pressure drop on the behavior of moisture and the target gas must be reviewed. Conversely, if a humidifier or membrane conditioner is used to adjust humidity conditions for a gas sensor, target constituents can be lost or react. Pretreatment equipment must be validated for each target gas, including recovery rate and total response time.

Sampling points should be arranged according to leak scenarios. First identify potential release sources, such as electrolyte fillers, solvent-transfer connections, storage and waste-liquid containers, and local-exhaust capture points, and account for vapor density and actual airflow. Applying only a rule of floor or ceiling placement can miss movement paths created by dehumidification circulation air and local exhaust. Pump failure, reduced flow, filter blockage, and line breakage must be monitored as separate fault conditions.

Solvent and electrolyte processes require separate hazard assessments

Depending on their formulation, electrode-coating processes may use solvents such as NMP, while cell-assembly areas fill and seal organic-carbonate electrolytes. Even in normally enclosed equipment, vapors can be released during filling, disconnection, cleaning, sample collection, waste-liquid handling, leaks, and maintenance. The detector gas to select must be determined using the latest SDS for the substances actually used, process temperature, quantity used, ventilation, and leak scenarios.

One must not assume that a single VOC sensor quantifies every solvent and electrolyte hazard in the same way. Photoionization detectors have substance-specific ionization energies and correction factors; catalytic-combustion sensors are affected by oxygen conditions and catalyst poisoning; and infrared sensors depend on the absorption characteristics of the target molecule. Total VOC trend monitoring, assessment of exposure to a specific substance, and lower-explosive-limit monitoring have different purposes and units, so the displayed value of one instrument must not be interpreted as another.

Dry rooms and explosion-hazardous areas must not be confused

The fact that a space has a low dew point does not by itself make it an explosion-hazardous area. A dry room is an environment defined by moisture-management criteria. A hazardous area, by contrast, is classified based on the likelihood and duration of flammable gas, vapor, mist, or combustible dust being present at an ignitable concentration. OSHA 29 CFR 1910.307 also requires each room, compartment, or area to be assessed individually and classified according to the nature of the materials present and the likelihood of ignitable concentrations occurring.

The converse is also true. Installing an explosion-protected certified sensor does not automatically ensure dry-room measurement performance at extremely low humidity. Explosion-protection marking means it has been assessed to avoid becoming an ignition source for a particular gas group, temperature class, protection method, and area. Low-dew-point accuracy, solvent resistance, sample adsorption, and calibration interval are separate performance items. Ultimately, sensors must be assessed on two axes: environmental performance and suitability for the hazardous area.

Hazardous-area classification should result from a process hazard assessment reflecting ventilation and enclosure level, normal releases and abnormal leaks, material flammability, and release frequency and duration. The inside of an electrolyte filler and its connection areas, a separate solvent-storage room, and waste-liquid handling points may have different classifications; categorizing or not categorizing the entire dry room in one sweep has weak justification. A qualified explosion-protection specialist must document the classification in accordance with the laws and standards of the country concerned.

The relationship between ESD measures and explosion-protection design

Extremely low humidity can make surface-charge dissipation more difficult and increase the potential for static-charge accumulation. Charges can arise from workers’ clothing, movement of film and powder, plastic tubing, transfer containers, and filter replacement. Static electricity can cause quality problems in sensitive electronic components and can also be an ignition source when flammable mixtures are present.

Basic measures include bonding and grounding conductive equipment; electrostatic-control floors, shoes, and clothing; ionization equipment; appropriate transfer speeds; minimizing nonconductive materials; and regularly checking grounding continuity. OSHA technical material describes bonding as a connection that equalizes potential between conductive objects, and grounding as a connection that dissipates accumulated charge to earth. However, grounding alone cannot remove charge from every insulator, so an ESD assessment that includes actual process materials and work activities is needed.

Sensors, pumps, flowmeters, junction boxes, and communication equipment inside a hazardous area must use protection methods appropriate to that area and its materials. Intrinsically safe circuits must be considered as a system including barriers and wiring, and explosion-protected cable glands and seals must not be replaced with arbitrary parts. Even when a sample is sent to an analyzer outside the hazardous area, the path through which vapor or flame may travel through piping, safe handling of exhaust gas, and the installation location of the pump itself must be reviewed.

Installation and maintenance plan

At a minimum, the sensor-selection table should record the following items.

  • Measurement purpose and target substance; alarm or control function

  • Normal and abnormal dew point, temperature, pressure, airflow, and chemical exposure

  • Measurement range, accuracy, repeatability, response and recovery time, and cross-sensitivity

  • Direct installation or sampling method and total delay time

  • Hazardous-area classification, gas and dust group, temperature class, and protection method

  • Calibration method, test gas, inspection interval, spares, and replacement criteria

  • Fault signals, treatment of missing data, alarm interlocks, and safety-shutdown conditions

During commissioning, map spatial variation by comparing with a reference instrument, and verify responses to actual disturbances such as door opening, worker entry, material entry, and dehumidifier changeover. Function-test the gas-detection system from the sensor through the sample line, alarm, and interlock using test gas at a known concentration. For a sampling system, measure whether the calculated transport time matches the actual alarm time.

Maintenance does not end with calendar-based calibration. Manage abnormal flattening of dew-point trends, discrepancies between adjacent sensors, reduced pump flow, filter differential pressure, line leaks, and failure to return to zero as condition indicators. After a solvent leak, high-humidity exposure, extended shutdown, or sensor replacement, perform additional inspection rather than waiting for the regular interval. Calibration gas, tubing, and regulators must also be checked for suitability under dry-room conditions.

Final selection begins with two questions

The first question is, “Does this sensor maintain the required accuracy and responsiveness at the target dew point and in the actual chemical environment?” The second is, “If the installation point is a hazardous area, has this completed system been certified and designed as suitable for that area and its materials?” Satisfying only one of these questions is not sufficient.

The core of battery dry-room sensor design is not installing the greatest number of sensors. It is defining moisture-ingress and chemical-release scenarios separately, selecting technology suited to the purpose of each measurement, and validating the entire measurement chain from sampling through alarms and maintenance. With this structure in place, low-dew-point quality control and fire-and-explosion prevention can be connected in a single operating system without confusing them.

References

Cover photo

Original photo on Pexels