Measuring Humidity in Sealed Glass Encasements

A technique has been devised for measuring the relative humidity levels in the protective helium/water vapor atmosphere in which the Declaration of Independence, the United States Constitution, and the Bill of Rights are encased behind glass panels on display at the National Archives in Washington, DC. The technique is noninvasive: it does not involve penetrating the encasements (thereby risking contamination or damage to the priceless documents) to acquire samples of the atmosphere. The technique could also be applied to similar glass encasements used to protect and display important documents and other precious objects in museums. The basic principle of the technique is straightforward: An encasement is maintained at its normal display or operating temperature (e.g., room temperature) while a portion of its glass front panel is chilled (see Figure 1) until condensed water droplets become visible on the inside of the panel. The relative humidity of the enclosed atmosphere can then be determined as a known function of the dew point, the temperature below which the droplets condense. Notwithstanding the straightforwardness of the basic principle, careful attention to detail is necessary to enable accurate determination of the dew point. In the initial application, the affected portion of the glass panel was cooled by contact with an aluminum plate that was cooled by a thermoelectric module, the exhaust heat of which was dissipated by a heat sink cooled by a fan. A thermocouple was used to measure the interior temperature of the aluminum plate, and six other thermocouples were used to measure the temperatures at six locations on the cooled outer surface of the glass panel (see Figure 2). Thermal grease was applied to the aluminum plate and the thermocouples to ensure close thermal contact. Power was supplied to the thermoelectric module in small increments, based on previous laboratory tests. A small flashlight and a magnifying glass were used to look for water droplets condensing on the inner surface of the glass. The temperature readings of the thermocouples were taken during cool-down and upon observing condensation. In determining the dew point, it was necessary to make a correction for the differences between the temperatures measured on the chilled outer surface of the glass and the temperature of the inner surface, where the condensation took place. The correction was derived from a laboratory test on a measurement setup that was nearly identical, except that the dew location on the inner surface was also instrumented with a thermocouple. The test showed that the temperature at the dew location on the inner surface of the glass panel was 0.9 C above the temperature determined from the measurements on the chilled outer surface of the panel.