DATABASE FOR TURBULENT PILOTED FLAMES WITH COMPOSITIONALLY INHOMOGENEOUS INLETS

INTRODUCTION:

Turbulent flames with inhomogeneous inlets are highly relevant for a range of practical applications such as gas turbines and engines where, either by design or due to instability, concentration gradients form upstream of the exit plane of the combustor so that the issuing mixture is compositionally inhomogeneous. Combustion in such environments may cover a range of modes such as premixed, stratified and non-premixed since mixture fractions may span the entire domain. There is need, therefore, for reliable generalized combustion models capable of accounting for the mixed-modes that are present in real combustors. The modified piloted burner presented here is capable of stabilizing flames where multi-mode combustion occurs and hence the data base is relevant for the development and validation of such generalised models. Distinction is made here between partially premixed flames with compositionally inhomogeneous inlet conditions and flows where the fluid is also partially premixed but with either homogeneous or near-homogeneous inlets.

This database presents measurements of flow, mixing and composition fields in a range of flames that have different inlet conditions and departures from blow-off. The objective is to facilitate the development of universal combustion models capable of accounting for all burning modes that may be present within a combustor. Some measurements of flow and mixing fields are presented for selected non-reacting cases to assist in validating this aspect of codes. Detailed boundary conditions are tabulated as relevant. The pilot stream usually consists of three gases (acetylene, hydrogen and air, referred to as 3GP) and has an adiabatic flame temperature that is higher than that of the parent fuel. To remedy this, a five-gas mixture (acetylene, hydrogen, carbon dioxide, nitrogen and air, referred to as 5GP) is also used in the pilot for some selected conditions. Data for flames stabilised with 3GP and 5GP pilots are presented here as shown in Table 1 below.

This site provides detailed description of the following sections:

1. Experimental set-up: showing details of the burner.
2. Stability characteristics: plotted for CNG fuel using the three-gas pilot and for methane fuel using both the 3GP and the 5GP configurations.
3. Photographs: showing the visible appearance of selected flames.
4. Non-reacting jets: showing boundary conditions as well as flow fields for a range of selected cases that are relevant to both 3Gp and 5GP cases (see Tables 2 and 3).
5. Reacting jets: There are a few sub-sections here. One presents details of the flow fields and boundary conditions for selected flames using CNG as fuel (see Tables 4 and 5). The second section provides flow measurements of reacting 3GP and 5GP pilots with an air jet. Another section presents composition measurements for THREE sets of data, all with pure methane as fuel. These are labelled as follows:
   1. I2013-3GP: Measurements in flames with the three gas pilot, performed in 2013 at normal resolution.
   2. I2013-5GP: Measurements in flames with the five gas pilot, performed in 2013 at normal resolution.
   3. I2015-5GP: Measurements in flames with the five gas pilot, performed in 2015 at high resolution and with the addition of cross-planar OH LIF to enhance the measurements of scalar dissipation to 3D.
      Note: Data Set I2015-5GP is recommended for the purpose of validating combustion models.
   A final sub-section presents samples of high-speed LIF-OH sequences collected at a repetition rate of 10 kHz are also shown for CNG flames with a three-gas pilot (see Table 10)

Note that measurements of flow fields, Rayleigh imaging of mixing fields at the jet exit plane (to obtain mixture fraction), and high-speed LIF-OH images are all performed at Sydney University using CNG as fuel. The line-Raman-Rayleigh measurements (I2013-3GP, I2013-5GP and I2015-5GP) are performed at Sandia's Combustion Research Facility using methane as fuel. Further details on this and the selected cases are given below.

---

EXPERIMENTAL SETUP:

The burner [1], shown in Figure 1, consists of two concentric tubes referred to hereon as the "inner" or "central" pipe with an inside diameter of d=4 mm (wall thickness of 0.25 mm) and an "outer pipe" or "annulus" with an inside diameter of D=7.5 mm (wall thickness of 0.25 mm). The outer tube is shrouded by the pilot stream which has an inside diameter of D_p=18 mm and a wall thickness of 0.2 mm. The burner assembly which is duplicated at Sydney and Sandia is centered in a wind tunnel supplying a co-flowing air stream at a fixed velocity, U_c=15 m/s. At Sydney, the wind tunnel has a square cross section of 15x15 cm while at Sandia the tunnel is larger at 30x30 cm. Fuel can be injected in the inner tube while air flows in the outer tube in which case the configuration is referred to as FJ. The inverse case where fuel issues from the annulus is referred to as FA. When both inner and outer tubes are flush with the exit plane, the flame is fully non-premixed while near-homogeneous mixing occurs when the inner tube is recessed by 300 mm (75 inner tube diameters).

The fuel used is either methane (CP grade, 99.0% CH4) or compressed natural gas (CNG which consists of 88% CH4, 7.8% C2H4, 1.9% CO2, and 1.2% N2 by volume with the remaining balance of 1.1% consisting of H2, H2O and other hydrocarbons). CNG is used for the experiments conducted at Sydney while methane is employed at Sandia for the Raman-Rayleigh line measurements. Two pilot gas mixtures are also used: a 3-gas mixture consisting C2H2, H2, and air and a 5-gas mixture of C2H2, H2, CO2, N2, and air with both mixtures having a C/H ratio of 1/4 matching that of methane. The 3-gas mixture produces excess heat with an adiabatic temperature of 2480 K while the 5-gas mixture is chosen to match the adiabatic equilibrium temperature of methane which is 2226 K. Conditions using the 3-gas pilot are referred to here on as 3GP while those using the 5-gas pilot as 5GP. Both pilot conditions are operated at stoichiometric and a heat release rate of 2.2kW. This corresponds to a bulk unburned pilot velocity of U=3.0m/s for the 3GP cases and U=3.7 m/s for the 5GP cases.

---

STABILITY CHARACTERISTICS:

As documented in [1], parameters that affect flame stability are: (i) the bulk jet velocity, U_J (ii) the ratio of bulk air to fuel injected through the outer and inner tubes, or bulk equivalence ratio at the jet exit plane, φ_J (the volumetric ratio is referred to as V_A/V_F while the mass ratio is M_A/M_F), (iii) the distance of recession of the inner tube, L_r; (iv) the heat release from the pilot, H_p; and (v) the velocity of the co-flowing air, U_c. Details presented in this section will highlight which of these are primary, controlling parameters. Turbulence levels in the co-flow, as well as the boundary layer thickness on the outer surface of the burner may also affect the flame and these will be specified in the boundary conditions. The bulk velocity, U_J is determined from the total flow of fuel and air in the outer and inner tubes divided by the exit area of the outer tube (which is larger than the sum of the areas of the inner tube and the annulus due to the wall thickness). The RMS fluctuations of velocity and mixture fraction at the jet exit plane are dictated by the extent of recession, L_r as well as the fuel and air flow through the inner and outer pipes.

Figure 2 shows the blow off limits plotted as bulk jet velocity U_J at blow-off, U_bo, versus recess distance, Lr. Results are shown for both the FA and FJ configurations with a volumetric ratio between the air and fuel streams, V_A/V_F=2.00 so that for CNG: φ_J=4.79, U_A/U_F=0.89, M_A/M_F =3.26, and ξ_J=0.235while for methane fuel: φ_J=4.76, U_A/U_F=0.89, M_A/M_F =3.6, and ξ_J=0.217. Subscripts “A” and “F” refer, respectively, to the air and fuel streams upstream of the jet exit plane so that U_A=bulk velocity in the air stream and U_F=bulk fuel velocity in the fuel stream (Note that, due to the wall thickness (0.25mm) of the inner tube, the area at the jet exit plane (corresponding U_J) is larger than the sum of the areas of the inner tube and the annulus(corresponding to U_A and U_F)

Figure 2a shows the blow-off limits for CNG using the 3-gas pilot while those for methane (also using the 3-gas pilot) are shown in Fig. 2b. Figure 2c shows the blow-off limits also obtained for methane but with the 5-gas pilot. In all cases, the FJ configuration shows the blow-off velocity peaking at some intermediate recess distance, Lr between 75mm and 100mm, while the FA cases experience a progressive increase in U_bo with recess distance. At Lr≈300 mm both configurations converge to the same blow-off velocity, which corresponds to the homogeneous limit. The difference in blow off limits between CH₄ and CNG (Figures 2a and 2b) is attributed to the ethylene content. This was confirmed by Meares [2,3], where, using the same burner, incremental addition of ethylene to pure methane fuel led to a gradual enhancement of the blow off velocity. In Figs. 2b and 2c, the difference in the blow-off limits between the 3GP and 5GP cases is expected due to the lower temperature content in the latter.

Five flames were selected for further measurements for each of the 3GP and 5GP cases as shown by the stars (FJ configuration) and squares (FA configuration) on Figs. 2a-2c. Further details about the selected cases are listed in Table 1. The bulk velocity U_J is based on the total volumetric gas flow rate through the exit plane of the burner. The Reynolds number is based on U_J at the jet exit plane and the annulus diameter, while U_A and U_F refer to the bulk velocities in the air and fuel streams upstream of the jet exit plane. The potential heat release, Hr is calculated from the lower heating value and the bulk mass flow rate of methane. For the 3GP cases, three of the five selected flames (FJ200-3GP-Lr100-82, FJ200-3GP-Lr300-82, and FA200-3GP-Lr100-82) have a fixed U_J=82m/s but different levels of inhomogeneity at the inlet as well as different departures from blow-off. However, for the 5GP cases, three flames are selected to have the same departure from blow-off but different levels of inhomogeneity at the inlet as well as different values of U_J (FJ200-5GP-Lr75-80, FJ200-5GP-Lr300-59, and FA200-5GP-Lr75-45). Flames FJ200-3GP-Lr100-82, FJ200-3GP-Lr100-115, and FJ200-3GP-Lr100-139 have the same inlet configuration but increasing jet velocity and hence different departures from blow-off. This is also true for cases FJ200-5GP-Lr75-59, FJ200-5GP-Lr75-80, and FJ200-5GP-Lr75-103.

Measurements of the flow fields were performed at Sydney for selected flames using CNG fuel with both 3GP and 5GP pilots (see Tables 4 and 5). Flow field measurements were also made in selected non-reacting jets relevant to the 3GO and 5GO reacint counterparts (see Tables 2 and 3). High-speed LIF imaging of OH (HS-LIF-OH) was also performed at Sydney for the 3GP cases (see Table 10).

Using methane as fuel, both 3GP and 5GP cases were used in 2013 at Sandia for detailed measurements of composition but with regular resolution. The data sets resulting from this campaign are referred to hereon as I2013-3GP and I2013-5GP, respectively. Additionally, multi-scalar diagnostics of the 5GP cases was undertaken in 2015 with improved resolution and with cross-plane imaging of OH-LIF to enable and estimate of the three dimensional scalar dissipation. This data set is referred to here on as I2015-5GP.

---

PHOTOGRAPHS:

Photographs of the flame series for different departures from blow off are shown below. Approximate values are noted for U_J/U_bo since the appearance of the flame is similar within these bands. The following explanation is given for the naming terminology:

• FA200-3GP-Lr100-82: refers to a flame in the FA configuration (fuel in the annulus), V_A/V_F=2.00, recess, Lr=100mm, bulk velocity of U_J=82m/s, and with a 3-gas pilot.
• FJ200-3GP-Lr300-139: refers to FJ configuration (fuel in the central pipe), with V_A/V_F=2.00, recess L_r=300mm, U_J=139m/s, and with a 3-gas pilot.

---

NON-REACTING JETS:

This section provides measurements of velocity fields performed at the University of Sydney in selected jets of CNG and air. Standard laser Doppler velocimetry (LDV) is used for the measurements of velocity by seeding the inner and outer jets with sub-micron aluminium oxide particles. Flow field measurements for selected non-reacting (fresh fuel and air in the jet and pilot) 3GP and 5GP cases are available and their details are given in Table 2 and Table 3 respectively. Flow field measurements are also available for selected 5GP cases with a non-reacting jet (air only to match bulk jet velocity) with a reacting pilot and their details are given in Table 4.

In a separate experiment, the Rayleigh scattering technique was used to provide a quantitative and instantaneous measure of mixture fraction as close as possible to the jet exit plane for selected 3GP cases (Table 2). This was possible due to the significant differences in the Rayleigh cross sections of methane and air.

The boundary conditions are also provided for the co-flowing stream including the boundary layer on the outer surface of the burner.

Practical limitations of the measurement techniques dictate that the boundary conditions for velocity were measured at x/D = 0.2 and 0.133 for the 3GP and 5GP cases respectively, and the boundary conditions for mixture fraction were measured at x/D = 0.5. It is expected that these are sufficiently representative of conditions at the jet exit plane.

Boundary Conditions:
Non-reacting_3GP_Velocity_Boundary_Condition.zip
Non-reacting_5GP_Velocity_Boundary_Condition.zip
coflow.csv
Non-reacting_3GP_Mixture_Fraction_Boundary_Condition.zip

Velocity Profiles:
Non-reacting_Velocity-3GP.zip
Non-reacting_Velocity-5GP.zip
Reacting_Pilot_Only_Velocity-5GP.zip

---

REACTING JETS:

There are a few sub-sections here due to the fact that the velocity and Rayleigh imaging of mixture fraction, as well as the high-speed LIF-OH imaging were performed at The University of Sydney using CNG as fuel. Additionally, line-Raman-Rayleigh measurements were performed at Sandia using methane as fuel with both three-gas and five-gas pilot. Tables 4 and 5 list the 3GP and 5GP cases respectively with velocity measurements taken at Sydney using CNG as fuel. Table 8 lists the 3GP caes measured at Sandia using methane as fuel (Data Set: I2013-3GP). Table 9 lists the 5GP cases measured at Sandia whcih have different blow-off limits and different jet velocities. The cases listed in Table 6 have been used both in 2013 (with normal resolution, Data Set: I2013-5GP) and 2015 (with high resolution, Data Set: I2015-5GP).

The first sub-section presents details of the jet flow fields and boundary conditions (CNG fuel), this is followed boundary condition measurements of the pilot and co-flow. Sandia composition measurements from three separate datasets for 3GP and 5GP flames but with pure methane are presented in the next sub-section. The final sub-section presents samples of high-speed LIF-OH sequences collected at a repetition rate of 10 kHz are also shown for CNG-3GP flames (see Table 10).

Jet Velocity Profiles and Boundary Conditions:

Cases with 3GP:

Table 5 shows the conditions for the 3GP cases which velocity and boundary condition measurements are presented here.Three of the five selected flames (FA200-3GP-Lr100-82, FJ200-3GP-Lr100-82 and FJ200-3GP-Lr300-82) have a fixed U_J=82 m/s but different levels of inhomogeneity at the inlet as well as different departures from blow-off. Flames FJ200-3GP-Lr100-82, FJ200-3GP-Lr100-115, and FJ200-3GP-Lr100-139 have the same inlet configuration but increasing jet velocity and hence different departures from blow-off.

The following conditions are fixed for all 3GP cases shown here:

• Stoichiometric pilot with an unburnt bulk velocity, U_pu=3.0m/s which corresponds to a heat release rate of H_p=2.2 kW. Note that the pilot is a mixture of acetylene, hydrogen and air with a C/H of 1/4 matching that of methane but with a higher adiabatic flame temperature 2480 K.
• The velocity ratio V_A/V_F=2.0 so that for CNG: φ_J=4.79, U_A/U_F=0.89, M_A/M_F =3.26, and ξ_J=0.235 while for methane fuel: φ_J=4.76, U_A/U_F=0.89, M_A/M_F =3.6, and ξ_J=0.217.
• The coflow air velocity is fixed at: U_C=15m/s.

Note: The boundary condition for mixture fraction for the 3GP cases may be obtained either from the Rayleigh measurements listed in Table 5 or, for the reacting cases, from the most upstream location for reactive scalars (Table 7).

Boundary Conditions:

Reacting_Velocity_Boundary_Condition-3GP.zip
Reacting_Mixture_Fraction_Boundary_Condition-3GP.zip

Velocity Profiles:
Reacting_Velocity-3GP.zip

Cases with 5GP:

Species Profiles
Table 6 shows the conditions the 5GP case for which velocity and boundary conditions measurements are presented here. Case FJ200-5GP-Lr300-59 has a recession distance of Lr = 300mm and is near homogeneous at the jet exit. The other two cases (FJ200-5GP-Lr75-57 and FJ200-5GP-Lr300-59) are both inhomogeneous at the jet exit but are at different departures from blow-off. However, FJ200-5GP-Lr300-59 and FJ200-5GP-Lr75-80 are both 30% departure from their respective blow-off velocities.

The following conditions are fixed for all 5GP cases shown here:

• Stoichiometric pilot with an unburnt bulk velocity, U_pu=3.7m/s which corresponds to a heat release rate of H_p=2.2 kW. Note that the pilot is a mixture of acetylene, hydrogen, carbon dioxide, nitrogen and air with a C/H ratio and adiabatic flame temperature of 1/4 and 2226K respectively, matching those a stoichiometric methane/air flame.
• The velocity ratio V_A/V_F=2.0 so that for CNG (velocity measurements): φ_j=4.79, U_A/U_F=0.89, M_A/M_F=3.26, and ξ=0.235 while for methane fuel (mixture fraction measurements): φ_j=4.76, U_A/U_F=0.89, M_A/M_F=3.6, and ξ=0.217.
• The co-flow air velocity is fixed at: U_c=15m/s.

Note:

1. Since there are no Rayleigh measurements of mixture fraction for the selected cases with 5GP-CH₄. The boundary conditions are obtained from the most upstream measurement location for reactive scalars as shown in Table 9. This may be deemed sufficiently close to the jet exit plane.
2. The blow-off limits are not provided here for the CNG flames with 5GP since the intention of the velocity measurements is simply to match the flowrates of the CH₄ 5GP flames measured at Sandia. Note that this applies to LDV measurements only as mixture fraction boundary conditions are from Sandia measurements with methane as fuel.

Boundary Conditions:
Reacting_Velocity_Boundary_Condition-5GP.zip
Reacting_Mixture_Fraction_Boundary_Condition-5GP.zip

Velocity Profiles:
Reacting_Velocity-5GP.zip

Pilot and Co-flow Boundary Conditions:

This sub-section presents measurements of boundary conditions peformed as close as possible to the exit plane of the jet. Velocity measurements were taken at Syndey while composition and temperature measurements from Raman-Rayleigh-CO-LIF measurements were taken at Sandia. For velocity measurements of the reacting pilot, the jet used air only with a volume flowrate, Q = 100 l/min, which corresponds to a bulk jet velocity, U_j=3.7.7m/s. The composition and temperature measurements were taken at Sandia using Raman-Rayleigh-CO-LIF. Specific details of the measurements conditions for velocity and multi-scalar experiemnts can be seen in Table 7 below. Flow rates of pilot reactants and mean values of the pilot products can be found in "Pilot_Species_Temp_Boundary_Conditions_Mean_Values.xlsx". The boundary layer of the co-flow was measured without any pilot or jet flow at an axial height x/D = 0.133 from the brner exit.

Boundary Conditions:
Pilot_Species_Temp_Boundary_Conditions_Detailed.zip
Pilot_Velocity_Boundary_Conditions_Reacting.zip
Pilot_Species_Temp_Boundary_Conditions_Mean_Values_Reacting_Non-reacting.xlsx
coflow.csv

Notes

1. File "Pilot_Species_Temp_Boundary_Conditions_Mean_Values.xlsx" contains mean values of composition and temperature near the exit plane for the reacting pilots (3GP and 5GP). Also given are the flowrates and mole fractions of reactants upstream of the pilot flames holder.
2. Velocity measurements of the co-flow were taken without any pilot or jet flow and are the same as those provided in the non-reacting section.

Species Profiles:

The reader is referred here to a number of publications discussing the compositional structure of these flames [3-4]. Three datasets of temperature and major species are presented, all collected at Sandia National Laboratories using methane as fuel. One data set consists of five flames with a 3 gas pilot (I2013-3GP) and the other two consist of five flames with 5 gas pilots (I2013-5GP and I2015-5GP). Two 5GP data sets consist of the same five flames, one collected in 2013 at a resolution of 100 microns, and another collected in 2015 with a higher resolution of 20 microns and additional cross planar OH LIF measurements. The 3GP flame species data was also collected at 100 microns without the OH PLIF.

Points of note and references to the experimental setup for each dataset are noted in their individual subsections. Note that despite attempts to preserve experimental conditions, discrepancies in flame structures may be observed between the two 5GP datasets due to small differences in boundary conditions for which partially premixed flames are sensitive. For validation of numerical models we recommend use I2015-5GP due to improvements in coaxial symmetry, processing techniques, and additional axial locations.

The following conditions are fixed for all cases and datasets:

• The pilots are as described in the experimental setup section. Both 3 and 5 gas pilots are stoichiometric, have a heat release rate of H_p=2.2 kW, and C/H ratio of 1/4. Adiabatic flame temperatures are 2480K and 2226K for the 3GP and 5GP respectively. The velocity ratio V_A/V_F=2.0 so that for methane fuel: φ_J=4.76, U_A/U_F=0.89, M_A/M_F =3.6, and ξ_J=0.217.
• The coflow air velocity is fixed at: U_C=15m/s.
• For all cases there are some differences in results from the positive and negative sides of the burner due to asymmetry of the inner jet tube. For this reason, only the positive side data is used to calculated conditional mean and RMS profiles. The level of asymmetry changes from case to case and can be observed in the radial profiles.

I2013-3GP:
The I2013-3GP dataset provides multi-scalar measurements for five flames, three at a fixed bulk velocity, U_j=82 m/s (FJ200-3GP-Lr100-82, FJ200-3GP-Lr300-82 and FA200-3GP-Lr100-82) but with either different burner configuration or recession, and two higher velocity cases (FJ200-3GP-Lr100-115 and FJ200-3GP-Lr100-139) with the same level of inhomogeneity. The three FJ-3GP-Lr100 cases have the same inlet configuration to demonstrate the effect of increasing velocity and different departures from blow-off. More specific information about these flames can be found in Table 5.

The diagnostic system has been described in detail elsewhere [5,6] and uses combined Raman and Rayleigh scattering and laser induced fluorescence of CO to obtain temperature and major specie's mass fractions along a 6mm probe length. Groups of 500 shots were taken at radial locations 3mm apart giving at least 1000 samples for most radial locations.

I2013-3GP_Scatter.zip
I2013-3GP_Mean_RMS.zip

I2013-5GP:
Table 6 shows the conditions for the I2013-5GP measurements presented here, all using methane as fuel. Three of the five selected flames (FA200-5GP-Lr75-45, FJ200-5GP-Lr75-80 and FJ200-5GP-Lr300-59) have a fixed departure from blow-off (70%) but different levels of inhomogeneity at the inlet as well as different jet velocities. Flames FJ200-Lr75-57, FJ200-Lr75-80, and FJ200-Lr75-103 have the same inlet configuration but increasing jet velocity and hence different departures from blow-off.

The diagnostic system is the same as used for I2013-3GP and uses combined Raman and Rayleigh scattering and laser induced fluorescence of CO to obtain temperature and major specie's mass fractions along a 6mm probe length. For a more detailed description of the laser and optical setup refer to the following [5,6].

Notes:

1. Only measurements at the axial locations marked in bold in Table 9 are available in data set I2013-5GP.
2. For the 5GP flames listed in Table 9, velocity field measurements are available for selected cases (Table 6).

I2013-5GP_Scatter.zip
I2013-5GP_Mean_RMS.zip

I2015-5GP:
The flames presented in the I2015-5GP data set are the same as those of I2013-5GP, the details of which may be found in Table 9. The reason for this newer dataset was to allow a closer look at regions of interest in these flames and as such data was collected at three additional axial locations, x/D=2, 3 and 7 and a greater number of shots (~2000) were taken at specific radial locations. Despite best efforts to maintain flowrates between the data collection in 2013 and 2015 there are slight differences between the two datasets which are related to high sensitivity of these partially premixed to small changes in boundary conditions. Flame symmetry is also improved in 2015 dataset over the 2013 ones.

The experimental setup used for the 2015 measurements is described in more detail in the following publication [5] but a brief description will be given here. The laser diagnostic setup for the Raman/Rayleigh/COLIF portion is unchanged from the earlier measurements of these flames bar reducing the on-chip radial bin width of the Raman camera. The use of a wavelet denoising algorithm enabled the finer spatial resolution [7]. In addition to the increased resolution, planar laser induced fluorescence measurements of OH were used to provide information on the three dimensional structure of the flames.

Knowledge of the instantaneous flame orientation in combination with multi-scalar line measurements would allow calculation of full three dimensional scalar dissipation rates. However measurement of the flames normal along the full probe length is extremely difficult, so as a compromise, iso-surfaces from OH PLIF measurements are used to approximate scalar gradients. This places limitations on samples where these measurements are viable, particularly to areas with viable levels of OH signal and where the iso-contour tangent is far from the radial axis. Additionally flame fronts close to parallel with the laser axis have increased uncertainties [8,9] and so, in the files below, three dimensional scalar dissipation for samples where the angle between the iso-surface normal and the laser axis, θ>50°, are not included. The reporting of samples with unusable OH data or flame normal angles above the threshold is explained in the notes below and in more detail in [10].

Notes:

1. Samples with insufficient OH signal to determine flame front angle are given a NaN value in the theta column. Those samples with angles above the threshold to give reliable three dimensional gradient corrections are given NaN in the X3d column.
2. Due to issues with the burner traverse, offsets have been applied the radial values to account for drift during axial movement. A more in depth explanation and a table of the exact values used may be found in this PDF.

I2015-5GP_Scatter.zip
I2015-5GP_Mean_RMS.zip

Image Sequences for HS-LIF-OH:

Planar laser induced florescence of OH (HS-LIF-OH) was performed at a repetition rate of 10kHz. This imaging is performed at the University of Sydney using CNG as fuel. Sample images are presented here for selected flames shown in Table 10 and for selected axial locations. The full data set is huge and may be made available on request.

Each page of images shows samples for the five flames listed in Table 8 labelled (a) to (e) on the LHS of each band of images referring to the following flames (from top to bottom): (a) FA200-3GP-Lr100-82, (b) FJ200-3GP-Lr300-82, (c) FJ200-3GP-Lr100-82, d) FJ200-3GP-Lr100-115, e) FJ200-3GP-Lr100-139. Note that images in a given row are time-sequenced with a separation of 0.1ms between images (10kHz) but images in different rows corresponding to different axial locations (x/D=8 to 20) are not in time-sequences and are stacked together for convenience.

Reacting-OH-3GP.zip

---

REFERENCES:

1. Meares, S. Masri, A.R. "Turbulent Diffusion Flames of Hydrocarbon Fuels Stabilised on a Bluff Body", Combustion and Flame, 161(2), 2014, pp. 484-495
2. Meares, S. "Turbulent Piloted Jet Flames with Compositionally Inhomogeneous Inlets", PhD Thesis, The University of Sydney, 2014
3. Meares, S. Prasad, V.N. Magnotti, G. Barlow, R.S. Masri, A.R. "Stabilization of piloted turbulent flames with inhomogeneous inlets", Proceedings of the Combustion Institute, 35:1477-1484 (2014).
4. Barlow, R.S. Meares, S. Magnotti, G. Cutcher, H. Masri, A.R. "Local Extinction and Near Field Scalar Structure in Piloted Turbulent CH4/air Flames with Inhomogeneous Inlets", Combustion and Flame, 192, 2015, pp. 3516-3540
5. M.S. Sweeney, S. Hochgreb, M.J. Dunn, R.S. Barlow, "The Structure of Turbulent Stratified and Premixed Methane/Air Flames I: Non-swirling flows" Combustion and Flame, 159, (9) (2012) 2896-2911.
6. R.S. Barlow, G.H. Wang, P. Anselmo-Filho, M.S. Sweeney, S. Hochgreb, "Application of Raman/Rayleigh/LIF diagnositcs in Turbulent Stratified Flames" Proceedings of the Combustion Institute, 32, (1) (2009) 945-953.
7. M.S. Sweeney, S. Hochgreb, M.J. Dunn, R.S. Barlow, "Multiply Conditioned Analyses of Stratificatio0n in Highly Swirling Methane/Air Flames" Combustion and Flame, 160, (2) (2013) 322-334.
8. M.S. Sweeney, S. Hochgreb, R.S. Barlow, "The Structure of Premixed and Statified Low Turbulence Flames" Combustion and Flame, 158, (5) (2011) 935-948.
9. A.N. Karpetis, T.B. Settersten, R.W. Schefer, R.S. Barlow, "Laser imaging system for determination of three-dimensional scalar gradients in turbulent flames" Opt. Lett., 29, (4) (2004) 355-7.
10. H. Cutcher, R.S. Barlow, G. Magnotti, A.R. Masri, "Statistics of Scalar Dissipation and Reaction Progress in Turbulent Flames with Compositional Inhomogeneities" (Under Review).

ADDITIONAL REFERENCES:

1. H.C. Cutcher, R.S. Barlow, G. Magnotti, A.R. Masri, "Turbulent flames with compositionally inhomogeneous inlets: Resolved measurements of scalar dissipation rates", Proceedings of the Combustion Institute, 36, (2017), 1737-1745
2. R.S. Barlow, G. Magnotti, H.C. Cutcher, A.R, Masri, "On defining progress variable for Raman/Rayleigh experiments in partially-premixed methane flame", Combustion and Flame, 179 (2017) 117-129

---

DISCLAIMER: No responsibility is assumed by the suppliers of this data for any injury and/or property damage as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas based on these data.