Fire on the Web

Building and Fire Research Laboratory

Diffusion Flame Measurements

Notes to Modelers

(A) Wolfhard-Parker burner; 2-D rectilinear geometry

Methane fuel flows from a central slot (width = 8 mm) and dry air from each of two outside slots (width = 16 mm). This three-slot burner configuration produces two identical, two-dimensional flame sheets of length = 40-45 mm. The fuel and air chambers are filled with 1 mm diameter glass beads with fine copper screens at the bottom and top, giving uniform velocity profiles at the burner exit. Cold flow velocities for the methane and dry air are 11.0 cm/s and 21.7 cm/s, respectively. These re-calibrated values are somewhat higher than those given in our original paper (K.C. Smyth, J.H. Miller, R.C. Dorfman, W.G. Mallard, and R.J. Santoro, Combustion and Flame 62:157-181 (1985)).

For the steady methane/air diffusion flame, there is mild pre-heating of the fuel and air, as evidenced by the temperature profiles nearest to the burner surface.

Stabilizing screens. Two curved pieces located within a rectangular screen box stabilize the flame by serving as a heat sink. These curved screens intercept the high-temperature reaction zones at height of approximately 45 mm above the burner surface.

Measurement uncertainties. A detailed discussion of measurement uncertainties and database consistency checks can be found in T.S. Norton, K.C. Smyth, J.H. Miller, and M.D. Smooke, Combustion Science and Technology 90:1-34 (1993). In addition, a discussion of radical concentrations (specifically H atom, O atom, OH, CH, triplet CH₂, and CH₃) has been presented in K.C. Smyth, Combustion Science and Technology 115:151-176 (1996).

(B) Mitchell/Santoro burner; 2-D axisymmetric geometry

Fuel flows from a central tube (diameter = 11.1 mm) and air from a surrounding annulus (diameter = 102 mm). For the base case methane/air flame, the cold flow area-averaged fuel and air velocities are 7.8 cm/s and 7.9 cm/s, respectively. The fuel tube is an open pipe without screens or beads, with a sufficiently large length/diameter ratio (L/D = 11.8) that fully developed pipe flow occurs at the fuel tube exit. In contrast, the air chamber is filled with glass beads followed by several fine wire mesh screens and a 2.54-cm thick ceramic honeycomb section with 0.13-cm square cells, providing a uniform air flow velocity.

In the base case methane/air flame (designated CH4-78), the fuel and air are preheated to 550 K and 330 K, respectively, based upon experimental thermocouple measurements for a series of heights just above the burner under steady flame conditions. These data were extrapolated to estimate the temperature at the burner exit. For the methane flame with the higher fuel flow rate (CH4-101, 10.1 cm/s methane flow), the fuel is preheated to 450 K. For the non-smoking ethylene/air flame (C2H4-40), the fuel is preheated to approximately 625 K. See also L.R. Boedeker and G.M. Dobbs, Twenty-First Symposium (International) on Combustion, pp. 1097-1105 (1986).

Additional boundary condition considerations are described in the modelling study of C.R. Kaplan, C.R. Shaddix, and K.C. Smyth, Combustion and Flame 106:392-405 (1996). These include flat thermal profiles for the fuel and air streams, symmetry at the left-hand boundary, free-slip wall at the right-hand boundary, and zero-gradient (simple continuative) outflow condition at the top of the computational domain.

Measurement uncertainties. For the soot measurements a detailed discussion of data collection and analysis methods (using extinction and laser-induced incandescence) as well as the measurement uncertainties can be found in C.R. Shaddix and K.C. Smyth, Combustion and Flame 107:418-452 (1996). Note that the laser-induced incandescence measurements were calibrated against extinction results using m = 1.57 - 0.56i as the refractive index of soot.

NOTE: Although identical burners were used at NIST and in the laboratory of Bob Santoro at Penn State, our flames were unconfined, whereas a chimney was utilized by Santoro and co-workers. This may alter the flow fields, such that a rigorous comparison between the two sets of measurements is not possible.