Instrument Uncertainty Effect on Calculation of Absolute Humidity Using Dewpoint, Wet-Bulb, and Relative Humidity Sensors

As part of the U.S. Department of Energy's Advanced Desiccant Technology Program, NREL is characterizing the state-of-the-art in desiccant dehumidifiers, the key component of desiccant cooling systems. The experimental data will provide industry and end users with independent performance evaluation and help researchers assess the energy savings potential of the technology. Accurate determinationof humidity ratio is critical to this work and an understanding of the capabilities of the available instrumentation is central to its proper application. This paper compares the minimum theoretical random error in humidity ratio calculation for three common measurement methods to give a sense of the relative maximum accuracy possible for each method assuming systematic errors can be madenegligible. A series of experiments conducted also illustrate the capabilities of relative humidity sensors as compared to dewpoint sensors in measuring the grain depression of desiccant dehumidifiers. These tests support the results of the uncertainty analysis. At generally available instrument accuracies, uncertainty in calculated humidity ratio for dewpoint sensors is determined to beconstant at approximately 2%. Wet-bulb sensors range between 2% and 6% above 10 g/kg (4%-15% below), and relative humidity sensors vary between 4% above 90% rh and 15% at 20% rh. Below 20% rh, uncertainty for rh sensors increases dramatically. Highest currently attainable accuracies bring dewpoint instruments down to 1% uncertainty, wet-bulb to a range of 1%-3% above 10 g/kg (1.5%-8% below), andrh sensors between 1% and 5%.