About This dataset contains infra red recordings from a series of drop tower experiment conducted at ZARM. Thin PMMA samples were combusted under controlled atmospheric conditions and forced opposed flow. An overview of all experiment conditions and sample types is given in the index.csv file. Two kinds of samples were investigated: "full samples", i.e. continuous fuel strips, and "gap samples", i.e. fuel strips separated by metal barriers of different widths. Consequently, if the `Gaps [Sequence]` or `Gap Material [Material]` fields are empty, the "full sample" was tested. For the "gap samples" the sequence of gaps is given as the individual widths of each gap in mm separated by slashes, e.g. '3/4/5' to refer to a sample with 3, 4 and 5 mm gaps. IR data The infrared data is stored in matlab (.mat) files. The radial lens distortion of the individual frames was corrected as well as the temperature. The frames are stored as 3 dimensional arrays with the first two dimensions being the spatial and the third the time (or frame count) dimension. The value of each pixel corresponds to the measured temperature in °C at the pixel's location. Visual data Visual data is not included in this dataset (due to storage space) but can be provided on request. Figure data This data is published as part of a publication in the Proceedings of the Combustion Institute. In addition to the raw infrared files, the data shown in the figures is included in this data set as well. Each figure from the publication (only those that show processed data) has three corresponding files: a .png-file, a .svg-file, and a .csv-file. The latter contains the data shown in the figure in tabular format. Usage Experiment Overview The experiment overview is given in the index.csv file. Here, the metadata for each experiment is given. The corresponding IR data can be found in the `IR Data File [Path]` column. Loading the IR frames The frames are stored in matlab files. An example for loading the files in Python is given below: from scipy.io import loadmat exp_id = 166 mat_data = loadmat("{}.mat".format(exp_id)) # individual IR frames as a 3 dimensional array of shape (x, y, t) ir_data = mat_data["data_IR_undist"] In addition to the IR frames, "Sample_Shapes.csv" contains the borders of the sample, fuel strips, and metal barriers, in the recording. The "type" column of the shape denotes whether the border is for "PMMA" (fuel), "Stainless Steel"/"Aluminium" (metal barrier), or "global" (entire sample). import numpy as np import pandas as pd import matplotlib.pyplot as plt sample_shapes = pd.read_csv("Sample_Shapes.csv", delimiter=";") exp_shapes = sample_shapes.loc[sample_shapes.ExpID == exp_id] fig, ax = plt.subplots() frame = np.max(ir_data, axis=2) # or, e.g., ir_data[:, :, 50] img = ax.imshow(frame, cmap="jet", vmin=30, vmax=450) plt.colorbar(img, ax=ax, label="Temperature in °C", extend="both") ax.set_xlabel("x") ax.set_ylabel("y") # show the sample outline for _, s in exp_shapes.iterrows(): ax.plot([s.x0, s.x0], [s.y0, s.y1], color="white") ax.plot([s.x1, s.x1], [s.y0, s.y1], color="white") ax.plot([s.x0, s.x1], [s.y0, s.y0], color="white") ax.plot([s.x0, s.x1], [s.y1, s.y1], color="white") plt.show() The "sample shapes" can be used to crop the IR recordings to the relevant area like so: fig, ax = plt.subplots() # get "global" shape s = exp_shapes.loc[exp_shapes.type == "global"].iloc[0] frame = np.max(ir_data, axis=2)[s.y0:s.y1, s.x0:s.x1] img = ax.imshow(frame, cmap="jet", vmin=30, vmax=450) plt.colorbar(img, ax=ax, label="Temperature in °C", extend="both") ax.set_xlabel("x") ax.set_ylabel("y") plt.show() Acknowledgments The authors thank Jan Heißmeier and Michael Peters for technical support during the design phase and the experimental campaigns. This research has been funded by the German Federal Ministry of Research, Technology and Space through the German Space Agency at DLR in the framework of FLARE-G II & III (grants 50WM2160 & 50WM2456).