Date of Graduation


Document Type


Degree Type



Statler College of Engineering and Mineral Resources


Industrial and Managements Systems Engineering

Committee Chair

Steven Guffey


This study determined the deviations between Pitot traverses performed by traditional hand-held methods and a traverse device designed to hold the Pitot tube perfectly in position, effectively eliminating errors due to yaw, pitch, and insertion point deviations. The hand-held measurements were performed twice by nine different test subjects, while each position was measured with the traverse device five times. Two ten-point traverse measurements were taken at various distances downstream and upstream from a 90° elbow. Two different-sized ducts were used in the study, a 3.875-inch and a 6.875-inch. Airflows determined from these measurements were compared to the presumed "gold-standard" values for each duct, which is the airflow at the most downstream location measured using the traverse device (QRef ). The velocity in each duct was set to approximately 4500 FPM.;Results of the hand-held measurements (QHand) were surprisingly close to the corresponding values of QRef. When the averages of the two perpendicular traverses were taken, deviations from QRef were within +/- 2.5% when taken at the most downstream location. Values of QRef had a coefficient of variation of 0.59% for the 3.875-inch duct and 0.34% for the 6.875-inch duct. Values of QHand were within +/- 5% of the mean value of QRef at locations closer to an elbow or a plain duct entry with a mean deviation of 2%. The exception was 1 value of QHand taken at 4D downstream of the elbow that was 8.75%. At the same locations, measurements taken with the traverse device deviated from the mean value of QRef by as much as 5.8%, suggesting that disturbances to airflows due to the elbow and duct inlet were responsible for greater deviations than holding the Pitot tube by hand. This study demonstrates that manually traversing ducts contributes only modest errors in estimating duct airflows, assuming the subject deviations from QRef values estimates adds roughly 0.8% error and 1.4% to the coefficient of variation to Pitot traverse measurements.