Date of Graduation


Document Type


Degree Type



Statler College of Engineering and Mineral Resources


Mechanical and Aerospace Engineering

Committee Chair

Marc Besch

Committee Co-Chair

Arvind Thiruvengadam

Committee Member

Arvind Thiruvengadam

Committee Member

V’yacheslav Akkerman

Committee Member

Saroj Pradhan


Accurate measurement of oxides of nitrogen (NOx) concentrations at near-zero levels using the constant volume sampling (CVS) system is a critical consideration in emissions quantification for engine and vehicle research and certification purposes. Various technological efforts and research are being implemented to ensure emissions from automobile and on-road transportation sources tend towards lower concentrations. This development has resulted in NOx concentration measurements to shift towards near-zero and thus, reach the detection limits of traditional measurement approaches and methods. There are several factors that might deter achieving measurement accuracy at this level, as their effect becomes significant at low NOx concentrations. A number of selected factors were investigated as part of this study, including, making modifications to the current CVS measurement set-up to assess and ascertain the capability of the CVS in accurately quantifying NOx at near-zero concentration levels. The effect of background variability of NOx and total hydrocarbons (THC) was investigated, with both species exhibiting variations in concentrations within a 10 minutes span across 3 days of measurements. Additionally, a dilution air filter was incorporated into the CVS dilution and engine intake air duct system to reduce and stabilize the NOx and THC concentrations which resulted in a 31% average NOx across both analyzers and a 5% average THC reduction. Furthermore, low NOx capable analyzers were also utilized to enhance the CVS measurement system. In general, an average drift error of -1.7% was observed for both analyzers used over a 3-day period of measurements. Finally, the CVS dilution tunnel was heated to ensure a stable temperature across the tunnel within a range of 49℃ to 59℃ and thereby prevent any possible condensation inside the tunnel that could result in a reduced NOx and THC concentration measurement. Virtual NOx injections were carried out to simulate an average loss of NOx in the CVS sampling set-up. The analysis of the selected influencing factors showcased a tendency to affect the accurate characterization and quantification of NOx concentration measurements at near-zero levels