Abstract This study experimentally investigates the combustion stability, flame macrostructure, and emissions of diffusion-assisted partially premixed compressed natural gas (CNG) flames in a model gas turbine combustor equipped with a perforated plate burner featuring a central diffusion hole. The investigation focused on the effects of fuel diffusion percentage (DP) and oxygen fraction (OF) on flame stability across a range of equivalence ratios (Φ) and oxidizer Reynolds numbers (ReOxidizer). The first set of experiments established the baseline stability map for CNG-air flames by defining the upper and lower flammability limits in Φ-ReOxidizer domain across DPs of 0%, 5%, 10%, and 15%. The subsequent sets investigated the effects of oxygen enrichments (OF = 22%, 23%, and 24%) on the same stability framework, quantifying the shift in operability limits and the onset of flashback phenomena. Increasing DP and OF broadened the stable operating limits, reaching ∼30% broader limits. While the lower flammability limit was relatively unaffected by OF, the risk of flashback increased markedly with OF, especially at reduced ReOxidizer. Thermal measurements revealed a rise in peak flame temperature by approximately 4.4% at 10% DP compared to the nondiffusion case. Emission analyses showed that O2 and CO2 concentrations remain largely unaffected by DPs, whereas carbon monoxide (CO) shows a significant increase from 25 to 75 ppm at 0% DP and from 63 to 91 ppm at 10% DP, indicating incomplete combustion. The findings provide operability maps and emission trends that are directly relevant to the design and safe operation of gas turbine combustors with oxygen enrichment.

The stability, macrostructure, and emissions of diffusion-assisted premixed oxy-CNG (compressed natural gas) flames, stabilized over a burner of a perforated plate with a diffusion central hole, were experimentally investigated in a model gas turbine combustor. Four sets of experiments were conducted with the first three sets characterizing stability maps of the combustor, in terms of upper and lower blow-off limits, in the Φ–Reoxidizer domain, over ranges of fuel diffusion percentage (DP: 0%, 5%, 10% and 15%) at different oxygen fractions (OFs) of 29%, 32% and 36%. The fourth set characterizes the combustor near stoichiometry (Φ = 0.85) stability maps, in terms of flashback and blow-off limits, in the OF–Reoxidizer domain over ranges of the same range of DPs. The results showed better flame stability with extended upper and lower blow-off limits at higher DPs and OFs because of the excess CNG ejecting in the burner and high kinetics at high O2 concentrations. Overall flame length was elongated with OF; however, adding a diffusion flame at high DP fades the outer flame of the cone due to high O2 consumption and low reactivity. At Φ = 0.85, close to stoichiometry, operability of the combustor was constrained within the OF range of 25% (at blowout) to 46% (at flashback). The operating OF value should not surpass 36% at a DP of 15% in order to prevent flashback. The developments of concentrations of CO, CO2, and O2 are presented within the combustor domain for better characterization of the flame stratification.

CompressedNatural Gas (CNG), predominantly composed of methane (CH4) with minor constituents of higher hydrocarbons and inert gases, is renowned for its clean-burning properties due to its favorable hydrogen-to-carbon ratio. However, under ultra-lean conditions, achieving stable and efficient combustion is chemically challenging because of limited radical formation and suboptimal reaction kinetics. This study presents an in-depth chemical analysis of CNG-Air combustion using an innovative dual-stream burner design. The system integrates an annular pre-mixed CNG-Air stream with a centrally injected diffusion stream of pure CNG, a configuration that alters the local chemical environment and reaction pathways. By varying the diffusion ratio and adjusting the fuel flow rates at different overall equivalence ratios, the study revealed notable chemical and operational enhancements. At a 15% diffusion ratio, the upper limit of the equivalence ratio increased to phi = 1.328 by achieving an increase of 11% compared to a configuration without central diffusion. Conversely, a 5% diffusion ratio resulted in a dramatic decrease to phi = 0.039 and a percentage decrease up to 91.5% in the lower equivalence limit. These chemical modifications enhance the lower limits due to the presence of sufficient oxygen for improved chemical reactions, effectively broadening the operability range of ultra-lean combustors. Computational results indicate that the minimum axial temperature and minimum CO2 concentration were achieved at a 15% diffusion ratio with a recorded decrease of 21.86% and 26.59%, respectively, compared to a 0% diffusion ratio. This is explained by the fact that at greater equivalence ratio values, there is less oxygen available for combustion. However, at a 5% diffusion ratio, the minimum axial temperature and minimum CO2 concentration were recorded a decrease of 14.75% and 16.89%, respectively, compared to a 0% diffusion ratio. Environmentally, these improvements lead to significantly lower fuel consumption, lower maximum temperature and reduced emissions of greenhouse gases and nitrogen oxides, paving the way for cleaner and more sustainable combustion technologies.

A toroidal-topology traveling-wave thermoacoustic electric generator (TWTEG) is developed. It consists of a traveling-wave thermoacoustic engine, two linear alternators connected in parallel, and sets of variable resistive-capacitive (R-C) external electric loads, in conjunction with accessories and instrumentation required for experimental investigations. The working medium is helium with a static absolute pressure that varies from 25 bars to 30 bars. A detailed description of the thermal design of the heat exchangers is presented. Sustainable operation of the TWTEG is achieved over a range of external R-C loads at different imposed hot-side temperatures and mean gas pressures. The performance parameters are measured for different experimental conditions and compared with a developed lumped-element model. The comparison between the experimental results and predictions reveals a good agreement. The impedance matching between the thermoacoustic engine and the linear alternators is investigated experimentally over a wide range of external R-C loads. The external R-C loads play a crucial role in the operation of the TWTEG. The mean gas pressure changes the operating frequency; however, it has no significant influence on the operating range of the TWTEG on the R-C load map. Increasing the hot-side temperature improves the thermal-to-acoustic efficiency and extends the operating region into larger regions.

The current study reports an experimental investigation of flow regimes for acoustically oscillating flows in the near-wall region of a resonator. Particle Image Velocimetry (PIV) is used to measure axial and transverse velocity components inside an empty square resonator over the acoustic cycle for a wide range of Reynolds numbers and at a high Womersley number. The measurements are compared with LDV measurements in a circular duct. A good agreement between the PIV and LDV measurements is observed. The measured transverse distribution of the axial velocity agrees well with the Stokes boundary layer expectations up to the Reynolds number (based on acoustic velocity and viscous boundary layer depth) of 240. At higher Reynolds numbers, the measured velocity distribution deviates from the theoretical expectations, suggesting a change of regime at a Reynolds number between 240 and 272. To assert the critical Reynolds number, the velocity fluctuations and Reynolds stress are estimated. At Reynolds number of 272, an increase in the value of the Reynolds stress is observed at the edge of the viscous penetration depth which is regarded as an indication of changing the flow regime. To better qualify the high amplitude acoustic flow, the spectrum of the turbulent kinetic energy and evolution of the integral length scale over the acoustic cycle are investigated. The decay of turbulent kinetic energy with the wavenumber complies with the universal slope (−5/3). The integral length scale increases during the deceleration stage and decreases at the beginning of the acceleration stage.

The application of the oxy-fuel combustion technique could tackle the combustion process's environmental issues. Experiments were conducted on partially premixed air- and oxy-methane combustion flames stabilized over a novel perforated burner in the present work. The burner has a premixing ratio of 7.0. In oxy-fuel combustion, the experiments were performed at oxygen fractions (OF%: volumetric percentage of O-2 in the oxidizer mixture) of 29%, 32%, and 36% and over a range of operating conditions necessary for a stable flame. The results of oxy-combustion flames were compared with the corresponding air-combustion flames at the same operating conditions. Two sets of statistical analyses were performed for further confirmation of the experimental results. The first set investigated the operating parameters' effect, including OF and oxidizer Reynolds number (Re), on the upper flammability limits (UFL). Simultaneously, the second set studied the impact of OF and equivalence ratio on flame length. The experimental results revealed that the flammability limits get wider as the OF increases due to the resulting flame speed rise with O-2-enrichment. The statistical analysis is conducted by analysis of variance (ANOVA) technique, which carries innovation and confirms that OF and Re significantly impacted the UFL. The visual flame length of oxy-flames was longer than its correspondents of air-flames due to the reduction of flame speed associated with the negative influence of CO2 dilution in oxy-flames. The statistical analysis showed a significant effect of OF and equivalence ratio on the visible flame appearance.

This work investigates the effects of fuel and oxidizer flexibility on the operability of a partially premixed combustor stabilizing H-2-enriched compressed natural gas (CNG) oxy-flames over a perforated plate burner. Three sets of experiments were conducted over ranges of equivalence ratio (Phi), oxygen fraction (OF), and hydrogen fraction (HF). The first set was performed at OF = 29% to investigate the effects of H-2 addition (HFs: 0.0%, 10%, 20%, and 30%) on the combustor operability and flame macrostructure over ranges of (I) and inlet flow Reynolds (Re). In the second set, the experiments of the first set were repeated for different OFs (29%, 32%, and 36%) to study the effects of oxidizer flexibility on combustor operability and visual flame appearance. In the last set, the combustor operability was examined near stoichiometry (Phi = 0.85) over ranges of OF, HF, and Re. At Re = 1481 and OF = 29%, the lean blowout limit was extended from Phi of 0.55 for HF of 0.0% to Phi of 0.38, 0.35, and 0.30 for HFs of 10%, 20%, and 30%, respectively. However, the flashback limit was narrowed with H2 addition at Re = 1481 and OF = 32% from Phi = 1.1 for HF = 0.0 to Phi = 1.05 and 1.0 for HF of 20% and 30%, respectively. The flame becomes whiter, shorter, more intense within the inner cone and more stable with strong attachment to the plate with H-2 addition. Flames of similar Phi and OF, i.e., similar adiabatic flame temperature (T-ad), resulted in similar flame shape. The operability of the combustor near stoichiometry (Phi = 0.85) was not possible outside the range of OF from 25% (at blowout) to 46% (at flashback). The lean blowout limit of the combustor near stoichiometry was extended to lower OFs with H-2 addition. The combustor was able to drop down with the OF from 28% to 25% when the HF was raised from 0.0% to 30% at Phi = 0.85 and Re of 1481. However, it is advised to keep OF below 36% at HF of 30% to avoid flame flashback. The results showed that the flow Re becomes the dominant parameter controlling the combustor blowout limit, and reaction kinetics rates dominate the control of the flashback limit. Having similar inlet flow characteristics, i.e., at fixed Re, T(ad )becomes the most relevant parameter choice when designing a combustion system in order to avoid flame flashback when operated at medium to full load conditions whatever the flame type (partially or fully premixed flames) whatever the stabilization mechanism (swirl or perforated plate).

Axial impulse turbines have been utilized for oscillating water column wave-energy conversion because of their wide operating range without stalling point. Previous studies in the oscillating-flow environment gave no information on the performance of these turbines at high frequencies. This study extends the use of these turbines to conditions close to those encountered in thermoacoustic power generators, which require the turbine to operate at a large frequency in a closed duct. The use of bi-directional turbines as acoustic-to-mechanical power converters brings several advantages, such as their low acoustic impedance, in comparison with linear alternators, which facilitates integration with thermoacoustic engines. In the study, a variable-frequency test rig is set up using atmospheric air. Then, the performance of the bi-directional impulse turbine is studied at different rotor inlet/exit angle, rotor space-to-chord ratio, stator space-to-chord ratio, stator exit angle, and tip clearance. The dependence of the conversion efficiency on flow coefficient is reported over a set of constant input gas parcel velocities. The study also introduces an impedance segment to simulate the turbine on Delta-EC. The peak efficiency of the improved case is found to be 35.2% at a flow coefficient of 0.22, resulting an improvement of 38% over the reference case.

Standing-wave thermoacoustic engines and refrigerators utilize gas oscillations and stacks to produce thermoacoustic effects. The flow morphology at the inlets/exits of the stack affects the heat transfer processes and the viscous flow losses in the heat exchangers. In this work, the flow morphology and the size of the disturbance zone are investigated experimentally for different plate-end shapes (rectangular, circular and triangular) at different drive ratios, using Particle Image Velocimetry. The plates are placed inside a resonator filled with air at atmospheric conditions and the oscillations are generated by a loudspeaker operates at the resonance frequency of the system. The size of the disturbance zone is identified as the distance between the furthest axial location the vortex reaches and the location at which it forms. The results reveal that the size of the disturbance zone generally increases with the increase of the drive ratio for all plate-end shapes. At the same drive ratio, the use of circular-end plates reduces the size of disturbance zones with respect to rectangular-end plates. The use of triangular-end plates with 30 O cone angle causes further reduction in the disturbance zone size. The flow morphology in all cases is presented and analyzed qualitatively. At a drive ratio of 3 %, the disturbance zone extends to a distance of 5.6 mm, 3.3 mm and 1.7 mm for the rectangular, circular and triangular ends, respectively. These values are 14.7 mm, 12 mm and 10.6 mm at a drive ratio of 7 %.

The structure and lean extinction of premixed liquefied petroleum gas air flames seated on conductive perforated plates were examined experimentally, with focus on the effects of plate material, thickness, and hole diameter. The lean extinction limit was determined by gradually reducing the fuel-flow rate for a given air-flow rate, until extinction occurred. Flame structure was quantified by mapping the local mean temperature and species concentrations and by imaging the average visual length of the flame plume. Pyrometer measurements of the temperature of the upper plate surface were made to estimate the heat transfer through the plate. It was found that the flames stabilized on plates with higher thermal conductivity were shorter and more stable (i.e., have lower lean extinction limits). This was attributed to preheating of fresh reactant mixture by greater heat transfer through the plate. Increasing the hole diameter (percentage open area) was found to enhance flame stability by reducing the reactant jet velocity for a given flow rate of reactant mixture. Heat transfer through the plate deteriorated with increasing hole size. However, the positive effect of smaller jet velocity on flame stability overpowered the negative effect of reduced heat transfer, and the net result was enhanced stability with larger hole sizes. Plate thickness, on the other hand, was found to have a weak effect on flame stability and structure. Thicker plates showed slightly better stability characteristics because of greater heat transfer through them. Nonetheless, plate heat transfer did not affect flame stability as significantly as jet velocity did.

The current work investigates experimentally the combined effects of oxy-combustion and partial premixing of the oxidizer (O2 plus CO2) and the fuel (compressed natural gas, CNG) on flame stability considering new burner configuration, namely perforated plate burner. The work explains and quantifies the limits of flammability, visual flame appearance as well as the extinction mechanisms and the visual flame length on ranges of operating conditions. The work investigates the flame stability under two different sets of experiments. The first set was conducted over a range of equivalence ratio to obtain the upper and lower flammability limits in terms of oxygen fraction (percentage of O2 in the oxidizer mixture). The second set was conducted to quantify the range of equivalence ratio within which stable flames can be obtained at constant oxygen fraction of 36%. Visual flame appearance and extinction mechanisms are also investigated. The results showed that operation with an oxygen fraction of less than 29% is not possible over the considered range of equivalence ratio. Extinction at the upper flammability limit occurs by flashback when oxygen fraction exceeds 42% However, at the lower flammability limit, extinction occurs by blow-off when oxygen fraction drops beyond 29%.

This work investigates the feasibility and thermal comfort of using natural ventilation in order to achieve thermal comfort in a handball arena with realistic dimensions and a full occupation of 4300 persons in the Gulf area. The work numerically simulates the temperature and velocity fields inside the full arena using computational fluid dynamics techniques at different internal loads, prevailing wind speeds, prevailing wind temperatures and prevailing wind angles. The work generates certain air opening configuration to be used for natural ventilation and the results show that natural ventilation is feasible if the following conditions are met simultaneously: the occupation density is 25% or less, sitting in the prevailing wind side, the lighting load does not exceed 50% of its full capacity, the prevailing wind temperature does not exceed 30 °C and the prevailing wind velocity is in range 3-4 m/s, where the upper limit arises from the requirement to avoid high velocities in the playing area. These conditions can be met during the training time and during parts of the day and over parts of the year hours making this method conditionally feasible.

Thermoacoustic power converters consist of thermoacoustic heat engine and linear alternator. The linear alternator converts the acoustic power generated by the thermoacoustic engine to electric output. Efficient and stable operation of a thermoacoustic power converter requires acoustic matching between the engine and the alternator. It also requires matching between the linear alternator and the connected load. An experimental setup was built to measure and analyze the linear alternator performance under different thermoacoustic power converter operating conditions. The effects of the different design and operation factors on the key linear alternator performance parameters such as mechanical stroke, the generated electric power, the acoustic-to-electric conversion efficiency, the mechanical motion loss, the electric loss, and the fluid-seal loss were investigated experimentally and numerically. The experimental results were simulated using DeltaEC and reasonable agreement was obtained.

Thermoacoustic power converters consist of thermoacoustic engines that convert thermal energy into acoustic energy and linear alternators that convert the generated acoustic energy into electric energy. The conditions required for best acoustic-to-electric power conversion include that linear alternators operate under mechanical and electrical resonance simultaneously causing the acoustic impedance of the linear alternator to become purely real. Electrical resonance is achieved by balancing the linear alternator inductor’s impedance by using a power-factorcorrecting capacitor. However, the exact capacitance value depends on the mechanical stroke, which in turn depends on the load seen by the linear alternator, including the value of the capacitance used. Thus, if operation takes place at off-design conditions, the mechanical stroke in operation and the capacitance used may not lead to electrical resonance. This work experimentally investigates the linear alternator performance indices, namely the mechanical stroke, the dynamic pressure at the face of the linear alternator’s piston, the output electric power, the generated volt, the generated current, the acoustic-to-electric conversion efficiency, the mechanical-motion loss, the Ohmic loss, and the fluid-seal loss when operating at electrical resonance and when operating at different levels of off-electrical resonance for two types of loads: a linear (resistive load) and a non-linear constant-voltage DC electronic load. Increases in the acoustic to-electric conversion efficiencies of up to 27.8% and 54.7% can take place when operating at electrical resonance in the linear and non-linear cases, respectively. The effects of operation at and off-electrical resonance conditions on the harmonic generation and on the acoustic impedance under linear and non-linear loadings are presented.

Thermoacoustic power converters consist of thermoacoustic engines and linear alternators. The former convert thermal energy into acoustic energy and the latter convert the acoustic energy into an electric output. In industrial applications, thermoacoustic power converters are connected to the electric grid, which is a non-linear load that imposes a certain impedance condition to this combined system. However, during initial testing in the laboratory and prior to connection to the grid, a certain electric load must be used to dissipate the electric output generated by the linear alternator and to provide a stable and controllable operating point to the combined system made of the thermoacoustic power converter and the load. This work experimentally examines the characteristic curves of the supply (thermoacoustic power converter) and dissipation (load) systems for linear and non-linear loading in terms of supplied/dissipated powers versus mechanical stroke. The issues that must be considered when using a linear load are presented. The effects of using non-linear versus linear loads on the performance indices of linear alternators are presented and discussed for a range of operating frequencies, mean gas pressures and gas mixture compositions. The experiments are carried-out at a range of values selected to reflect thermoacoustic-power-conversion conditions. The results show that non-linear loading can provide an un-expensive, easily-applied method to provide a stable and controllable method to test linear alternators in the development phase under off-grid conditions. The performance indices of the linear alternator are compared using linear and non-linear loading. For example, when the break-down voltage of the zener diode used to induce the non-linearity in the load was 25% of the open-circuit voltage, the acoustic-to-electric conversion efficiency decreased by 2.8%.

A linear alternator lie at the interface between the acoustic power generated by a thermoacoustic engine and an electric load, and thus, its performance is significantly controlled by its matching with the thermoacoustic engine and with the electric load. Under thermoacoustic power conversion conditions, small unavoidable changes in the operating conditions may incur significant effects on the alternator performance. In this work, the sensitivities of several alternator performance indices, namely, the acoustic-to-electric conversion efficiency, the mechanical stroke, and some of the main alternator losses (mechanical damping loss, seal loss, and electric copper loss), are examined to small changes in four operating conditions. Using the methodologies of the design of experiments and sensitivity analysis, a scheme of experiments is designed and carried-out to analyze the sensitivity of these indices to ± 10% changes in operating conditions at mechanical resonance. The operating conditions considered are the gas mixture composition, the mean gas pressure, the dynamic pressure ratio acting on the alternator and the electric load. The results reveal how variations in each of these operating variables as well as variations in their combined interactions affect the alternator’s performance with respect to the results obtained in a reference experiment.

Concerns about global warming have encouraged the interest in hydrocarbon combustion techniques that allow easy capture of carbon dioxide. One technique for achieving this objective is through the use of pure oxygen instead of air for combustion or what is called oxy-combustion carbon capture technology. The main goal of the manuscript is to study flammability limits, visual flame appearance and exhaust emissions of diffusion flame stabilized over a bluff body over ranges of operating and design conditions. The operating conditions include flow Reynolds number, equivalence ratio and oxidizer composition. The design parameter considers the change of blockage ratio (BR) of the bluff body namely, BR = 036, 0.5, 0.67 and 0.82. Based on this, three sets of experiments were performed utilizing compressed natural gas (CNG) as a fuel to be burned with three different oxidizers including air, oxygen enriched-air and oxy-fuel mixtures (O-2 plus CO2 with a controlled oxygen fractions, OF). The three sets of experiments were performed to identify ranges for stable flame operation considering different oxidizers under different operating conditions. Stability limits, visual flame appearance and extinction limits of these flames are quantified and analyzed. Furthermore, three different regions were observed; precisely, jet flames, central jet dominated flames and recirculation zone flames, depending on the ratio between oxidizer and fuel momentum. The flame color changed from yellow for air combustion, to bright white for oxygen-enriched-air combustion and finally to blue with yellow tips for oxy-combustion. The flame length was the highest for air combustion, then lower for oxy-combustion and the lowest for oxygen enriched-air combustion. This was attributed to the effect of oxygen-enrichment which results in increase in the flame speed making flame length shorter. For the sake of comparison, the flammability limits of the three sets were reported and the results revealed that oxygen-enriched-air-flames have higher stability than air-flames and oxy-flames, respectively. (C) 2016 Elsevier Ltd. All rights reserved.

A thermoacoustic power converter consists of a thermoacoustic heat engine and a linear alternator. Integration of linear alternators into thermoacoustic power converters is complicated since it requires acoustic matching with the thermoacoustic engine and matching with the load connected to it. In order to fully understand this process, this work presents an experimental setup designed and built to test linear alternators under different thermoacoustic-power-conversion conditions. The setup supplies the acoustic power to the linear alternator in a controllable and stable form, using an acoustic driver. Results indicate that introduction of a low-resistance in parallel to the linear alternator can provide over-stroke protection on a time scale of few milliseconds. Results on how the key performance indices (mechanical stroke, acoustic-to-electrical conversion efficiency, mechanical-motion loss, fluid-seal loss and Ohmic loss) of the linear alternator are affected by the operating frequency, the mean gas pressure and the working gas mixture composition are presented and discussed. Under the conditions employed, the proportionality constant between the mechanical stroke and the generated voltage is found to increase linearly with the operating frequency (even across the mechanical resonance frequency point) and to decreases linearly with the mean gas pressure and to be almost independent on the gas mixture composition. The effects of the acoustic gas impedance at different mean gas pressures and different gas mixture compositions on the acoustic matching between the acoustic power supplied and the linear alternator are quantified.

Linear alternators convert acoustic power into an electric output and thus are integral parts of thermoacoustic power converters. Their performance is significantly affected by their matching with the thermoacoustic engine and their matching with the electric load. This critical operation may cause any slight change in the design or operating conditions to induce a significant effect on the linear alternator performance. In this work, this performance is monitored by examining the acoustic-to electric conversion efficiency, the mechanical stroke, the mechanical-motion loss, the fluid-seal loss and the Ohmic loss. A reference experiment is carried-out to document the performance of the linear alternator under certain operating conditions typical in thermoacoustic power conversion conditions at the linear alternator's mechanical resonance frequency. Then, using the methodologies of design-of-experiments and sensitivity analysis, a scheme of experiments is designed and implemented to analyse the changes in the alternator performance indices to 10% changes in four selected operating conditions, namely the helium molar fraction in the gas mixture composition, the mean gas pressure, the electric load resistance and the pressure ratio. The results reveal how variations in each of these operating variables as well as variations in their combined interactions affect the linear alternator's performance indices with respect to the results obtained in the reference experiment. For example, the mechanical stroke is mostly affected by the pressure ratio, followed by the mean gas pressure and then by the load resistance. The most significant combined interaction arises from the product of the mean gas pressure and the helium molar fraction.

In this work, two sets of experiments were performed including air and oxy-combustion premixed flames stabilized over a perforated-plate burner. A set of experiments was performed considering air-fuel combustion in order to identify a range of equivalence ratio for stable flame operation at fixed oxidizer Reynolds number and over a range of premixing ratio (namely 7, 25, 45, 67 and 128). Also, effects of premixing ratio (LID) on lower and upper flammability limits over a range of oxidizer Reynolds number were investigated. Another set of experiments was performed considering oxy-fuel combustion in order to identify a range of equivalence ratio for stable flame operation at fixed oxidizer Reynolds number. The study documents the visual flame length, appearance and color and identifies the extinction mechanism outside the flammability limits for both air and oxy-combustion flames over wide ranges of operating equivalence and premixing ratios. For air combustion flames, the results showed wide ranges of flammability limits at lower premixing ratios and tight ranges of flammability limits at higher premixing ratios. This can be attributed to the increase in the degree of flame diffusivity while reducing the premixing ratio and, as a result, more stable flame is obtained. Reductions in the emissions of NO and CO were observed while increasing the premixing ratio for the air combustion case. For the oxy-combustion flames, the results showed stable flame operation at oxygen fraction of 36%; however, the flammability limits were approximately 20% lower as compared to those of air combustion flames. Flash back was observed when operating oxygen fraction exceeded 40% and flame stability was affected badly with the decrease of oxygen fraction. The visual flame length was longer in cases of oxy-combustion flames as compared to those of air combustion. (C) 2016 Elsevier Ltd. All rights reserved.