How to use the measured background feature in MAUD - luttero/maud GitHub Wiki

In some experimental setups background from e.g. sample environments can make the background for a diffraction experiment not suitable for modeling by polynomials. In these cases, a measured background can be used in addition to a polynomial background. This document describes the procedure for HIPPO data, but the procedure should work for other applications with slight variations.

• In the Hippo wizard, load data files as usually plus a data file containing the measured background. Load the background data set first to save a few mouse clicks in the next step. As always, make sure for single pattern (5 rings instead of 45 panels) no histogram specific background parameters are defined. The "Treat rotations as separate datasets" option must NOT be checked.
• Edit each bank and check the "As background" flag for the background run in "Datafiles" for all of them. If the background file is the first file, it is automatically selected and one just has to check the "As background" checkbox and hit enter.
• Add phase(s)
• Make sure in Graphic - Plot Options, the checkbox for "Subtract background" is NOT checked. "Plot background" may be checked.
• Calculate the pattern
• Edit the Incident Intensity scale (parameter _pd_proc_intensity_incident) to values bringing the peak intensities above background in the experimental data closer to peak intensities above background in the calculated patterns for each dataset
• For the background run, adjust the "Monitor counts" in "Datafiles" (_pd_meas_counts_monitor for the GDA object corresponding to background run) to roughly match the strength of the background
• In Parameter list, click Fix all parameters
• Refine _pd_proc_intensity_incident, _pd_meas_counts_monitor for the background gda file.
• Free background parameters but fix background parameters for the measured background dataset if they are present in a 45 panel texture fit.
• Run refinement, should match background peaks.
• From here on refinement strategy should be normal.

Relevant settings in preferences are "backgroundSpectrum":

• "smoothing_polinomial_degree": the degree of a polynomial to interpolate the measured background.
• "smoothing_pol_additional_points": by default the interpolation with the polynomial used the same number of points as the degree of the polynomial plus 1, e.g. a 2nd degree polynomial by default with zero additional points uses 3 points. If this number is larger, e.g. 10, then the 2nd degree polynomial would be fit for 13 data points and make the experimental background smoother.