Design of a 1-Butanol Scale Dynamic Olfactometer for Ambient Odor Measurements J. E. Sorel, R. O. Gauntt, J. M. Sweeten, D. L. Reddell, A. R. McFarland* MEMBER MEMBER ASAE ASAE ABSTRACT A portable system has been developed for quantifying odors in atmospheric air. Panelists compare the intensity of ambient odors with the intensity of discrete levels of 1-butanol provided by the olfactometer. Range of delivered 1-butanol concentrations is 0 to 80 ppm in air at a flow rate of 15 L/min. Laboratory tests have been performed to ascertain precision, panelist response, variability between two olfactometers, and effect of delivery method (ascending, descending and random odor presentation). Nineteen panelists made determinations of comparative odor intensities from two identically constructed olfactometers, where one unit provided a set concentration of 1-butanol and the second provided variable concentrations to the panelists. At a given set concentration the panelist would be exposed to a sufficient number of intensities from the variable-output system such that a decision could be made on the match of intensities. Totally, 855 data points of matched intensities were acquired in the study. For the entire data set the mean value of the ratio of measured concentration to set concentration was 0.984. The geometric standard deviation of the concentration ratios was 1.44. INTRODUCTION Odor complaints are the most frequent problems for air pollution control agencies; for example, during 1981 the Texas Air Control Board received 1,467 complaints of which 51% were related to odors. Fourteen percent of these odor complaints were related to agricultural sources (Bradford, 1982). There are two techniques which agencies presently use Article was submitted for publication in November, 1982; reviewed and approved for publication by the Structures and Environment Div. of ASAE in March, 1983. Presented as ASAE Paper No. 82-3085. Mention of trade names is for identification only and does not imply either an endorsement or a preference over other products not mentioned. The authors are: JAMES E. SOREL, Graduate Research Assistant, Industrial Engineering Dept.; RANDALL O. GAUNTT, Research Associate, Civil Engineering Dept.; JOHN M. SWEETEN, Associate Professor, Agricultural Engineering Dept.; DONALD L. REDDELL, Professor, Agricultural Engineering Dept.; and ANDREW R. McFARLAND, Professor, Agricultural and Civil Engineering Depts.; Texas A&M University, College Station. Acknowledgment: Financial support for this study was provided by the Texas Air Control Board (TACB) through a grant to the Texas Agricultural Experiment Station—this assistance is gratefully acknowledged. The advice, suggestions and help of several individuals have been of considerable value in the development and testing of the system: Messrs. C. Bradford and R. Tannis of TACB; Messrs. R. DeOtte, Jr. and C. Ortiz of the Texas A&M Air Quality Laboratory; and Messrs. C. Collingsworth and G. Barnett of the Texas Agricultural Experiment Station. The authors also wish to express appreciation to the 19 panelists who donated their time to this study. *Individual to whom correspondence should be addressed. to determine whether odor levels are excessive. First is a purely subjective method in which observers state whether or not an odor is offensive. The second category is one which is based upon a combination of both analytical and subjective assessments. This technique is exemplified by the currently-used dilution-to-threshold method and the more recent 1-butanol reference method. In the former method (ASTM, 1978), a sample of odorous air is diluted with odor-free air and a determination is made of the dilution ratio at which 50% of the test subjects cannot detect the odor. At this threshold the number of odor units, Z, is calculated from: where V d = volume of dilution air mixed with a given sample to achieve the threshold level V 5 = volume of odorous air mixed with dilution air to achieve threshold level This particular test can be easily conducted in the field through use of a Scentometer (Barnebey-Cheney, Columbus, OH). Here, odor-free air, which is obtained by passing ambient air through activated charcoal, is mixed with ambient air at preset ratios in order to obtain a range of dilution ratios. The State of Colorado has an ambient odor standard based on the dilution-to- threshold technique. An odor violation can be considered to exist if Z ^ 8 in a residential or commercial area or if Z ^ 16 in other land-use areas. Animal feed lots are specifically exempted from odor regulation (Colorado Air Quality Control Commission, 1971). Because of the simplicity of the measurement procedure and data reduction, there is a certain appeal to use of the dilution-to-threshold approach; however, the methodology does have serious limitations: 1. During odor measurements, a difference in results is obtained if the observers attempt to detect the threshold of the particular odor of interest rather than simply detecting the presence of an odor without regard to the character. Typically, to identify the specific odor requires a concentration 1.5 to 10 times as large as the non-specific determination (Hellman and Small, 1974). 2. The change in perceived odor intensity is not a linear function of concentration and, therefore, not a linear function of the dilution ratio (Katz and Talber, 1930). If odor intensity is the parameter of significance, then perhaps it should be measured directly. 3. There is a range of threshold values for the population of observers, so care must be taken to eliminate odor panelists who may have olfactory defects (ASTM, 1978). 4. The manner of presentation of the dilution ratios (i.e. ascending, descending or random) can influence the 1983—TRANSACTIONS of the ASAE © 1983 American Society of Agricultural Engineers 0001-2351/83/2604-1201$02.00 1201