Temperature Shift in MLR Curves from TGA Simulations #14862
Description
Describe the bug
Hello, I’ve been testing some old TGA simulations (originally run prior to FDS 6.7.7) in FDS 6.10.1 and noticed a temperature shift in the MLR curves. I am aware that the old kinetic parameters need to be adjusted for the updated reaction rate expression by recalculating the pre-exponential factor.
To Reproduce
Run the provided input file (TGA_wood) and compare the output to the experimental curve used to derive the original kinetic parameters. The recalculation of the old pre-exponential factor is included directly in the FDS input file below each material.
A Python script is also provided to plot the comparison between FDS, the experimental data, and the optimized kinetics. The original set of kinetic parameters is provided in the table below.
If there were a typo or another error in the parameters, it would be unlikely for the curve shape to match so precisely.
| Material | Parameter [Unit] | Final R1 | Final R2 | Final R3 |
|---|---|---|---|---|
| OSB | log Aᵢ [log s⁻¹] | 8.881 | 25.928 | 2.013 |
| OSB | Eₐ,ᵢ [kJ mol⁻¹] | 132.473 | 348.991 | 50.817 |
| OSB | nᵢ [–] | 1.267 | 1.451 | 2.824 |
| OSB | ν_{A,ᵢ} [–] | 0.377 | 0.316 | 0.307 |
| OSB | ν_{residue,ᵢ} [–] | 0.22 | 0.22 | 0.22 |
Python plotter, original experimental and optimization data
Plotter.txt
MLRsmooth_PYRO_OSB_30_900_N2_05Kmin_recal_02.txt
FINAL_MOD_DATA1_OSB.txt
Expected behavior
I would expect FDS to reproduce the experimental data without a temperature shift of approximately 25 °C.
Screenshots
Desktop
OS: Win10 Enterprise LTSC
FDS: FDS-6.10.1-403-ge3a306c-nightly
Thank you for your help as always. I sincerely apologize if this turns out to be a simple oversight on my part.