MTech(Res) : Design of an Ex-Situ Gas Sampling Layout with Various Probes and Their Performance Assessment for Accurate Emission Measurements

Chemical kinetic mechanism development and validation require quantitative species data from canonical flames. Burner-stabilized pre -mixed flames provide a stationary post-flame region directly comparable to one-dimensional kinetic simulations. Ex-situ gas sampling delivers absolute CO and NO mole fractions from this region. The extraction step introduces two coupled biases. Reactions can continue inside the probe, and the probe can perturb the flame upstream of the inlet. This thesis develops a coordinated ex-situ sampling framework for atmospheric McKenna burner-stabilized CH₄/air flames. The framework combines a sampling and dilution layout, five sampling probes, and a staged correction methodology. The measurements span fuel-lean to fuel-rich flames at three heights above the burner. The sampling and dilution layout is one of the principal contributions of the present work. The complete layout was designed and assembled in this work to adapt a portable gas analyzer (PG-350) for laboratory flame diagnostics. The PG-350 is built for stationary-source emissions monitoring rather than laboratory flame research. The design repurposes this industrial emissions instrument for high-precision species measurement in flames. The adapted instrument serves fundamental academic flame research. The layout delivers a conditioned, metered, and range-compatible sample to the analyzer. A vacuum pump extracts the sample using a sampling probe and drives it through the line. Two water-cooled drain separators, designed and fabricated in this work, remove the post-flame moisture. An Alicat mass flow meter measures the conditioned sample flow rate. A controlled N₂ dilution arrangement extends the CO measurement range beyond the analyzer’s full scale for the rich flames. Five sampling probes represent four reaction-quenching strategies. The aerodynamic-quench (AQ) probe was designed using a one-dimensional gas-dynamics model developed in this work. A brute-force screening selected its geometry against the pump and analyser constraints. The water-cooled (WC), pressure-quench (PQ), and uncooled probes were sized using non-reacting two-dimensional axisymmetric CFD. The uncooled-metal (UC-M) probe serves as a hot-wall reference. The uncooled-quartz (UC-Q) probe isolates the wall-material effect under the same geometry. Each probe was assessed experimentally and through two-dimensional probe-resolved reacting CFD in ANSYS Fluent. A residence-time method quantified the probe-induced upstream shift of the sampling location for the designed probes. A staged correction methodology recovers the sampled-gas composition from the conditioned analyzer reading. It corrects the mass flow meter reading for its gas-property mismatch. It then removes the imposed dilution and restores the moisture removed during conditioning. An independent route back-calculates the dilution ratio from paired CO₂ measurements. The methodology propagates the input uncertainties through the correction chain using the GUM framework. Reported values quote the expanded uncertainty on the corrected species. The corrected CO and NO were evaluated against a one-dimensional chemical kinetic simulation of the unperturbed flame. The comparison accounts for the probe-induced upstream shift of the sampling location. Each designed probe draws its sample from a location displaced toward the burner relative to its tip. The three designed probes were compared against the reference at their displaced sampling locations. The pressure-quench probe is the most consistent for both species. The water-cooled probe is a close alternative. The vacuum pump performance degraded due to moisture condensation in the pump head. Therefore, it held the AQ suction above its design value throughout the campaign. The AQ probe is therefore off-design and cannot be ranked on equal terms. The uncooled-metal probe shows a large NO deficit in the rich flames from hot-wall surface reactions. The uncooled-quartz probe isolates the metal wall as the origin of this deficit. These rankings measure consistency with the kinetic reference rather than absolute accuracy. This work delivers outcomes of broader use to the combustion community. The framework offers documented guidelines to the community for ex-situ sampling experiments. It provides a comprehensive assessment of how each quenching strategy performs and which probe suits a given measurement. The corrected CO and NO data are suitable for chemical kinetic mechanism development and validation. Many detailed chemistry mechanisms exist, yet reacting CFD of practical combustion devices requires reduced mechanisms that reproduce the target species profiles. The present setup allows reduced mechanisms to be validated quickly across flames spanning fuel-lean to fuel-rich conditions. A validated reduced mechanism can then support detailed CFD analysis of practical combustion systems.

Speaker :  SARVAGYA SHARMA

Research Supervisor :  Irfan Ahmed Mulla