Sub-grid scale turbulent micro-mixing with infinitely fast chemistry in the context of fire modelling

Combustion modelling with the Eddy Dissipation Model (EDM) requires closure of the mixing time between fuel and oxidizer occurring on a sgs level. Six different models for estimating the micro-mixing times, in the context of fire modelling, are investigated: (1) equal to a constant fraction of the integral time scale (i.e., default EDM approach); (2) allow for this fraction to vary in space and time as a function of the standard deviation of the mixture fraction; (3) use of the geometric mean of the integral time scale and the Kolmogorov time scale; (4) use of the geometric mean of the sub-grid velocity stretching time and the Kolmogorov time scale; (5) calculation of the mixing time scale based on the local variation of the scalar dissipation rate and the mixture fraction variance; (6) use of the minimum of the mixing times for diffusion, sub-grid scale advection and buoyant acceleration. Overall, the performance of all models on fine grids is comparable. However, models 2 and 6 perform better for all the grid sizes and test cases considered. Model 1 predicts the highest temperatures, the performance of models 3 and 4 is similar while the predictions with model 5 are highly grid-dependent. •Six different micro-mixing models with infinitely fast chemistry are tested.•The model using an effective diffusion time scale was highly grid-sensitive.•The differences between the models were more pronounced on coarser grid sizes.