CorelliPowderCalibrationCreate v1¶

Summary¶

Adjust bank positions and orientations to optimize peak determination in d-spacing

Properties¶

Description¶

Standard powder samples with Bragg scattering diffraction rings of accurately known lattice spacings can be used to adjust the position and/or orientation for the components of an instrument (moderator and detector bank) such that the observed peak intensities yield lattice plane spacings as close as possible to the standard reference values.

This algorithm adjusts the position of the moderator as well as the position and orientation of each bank of detector pixels. The goal of these adjustments is to produce neutron paths and scattering angles that lead to optimal comparison between observed and reference lattice plane spacings

The execution workflow is as follows:

Usage¶

For best results, a tube calibration should be applied to the input event workspace, before attempting to adjust the position and orientation of each bank.

• Example - adjust banks 42 and 87

from mantid.simpleapi import *
from corelli.calibration import load_calibration_set, apply_calibration
#
# Loading the runs for the standard powder sample
LoadNexus(Filename='/tmp/CORELLI_124036_banks42_87.nxs', OutputWorkspace='LaB6')
#
# Apply the tube calibration
table, mask = load_calibration_set(input_workspace, '/SNS/CORELLI/shared/tube_calibration/', 'table', 'mask')
apply_calibration('LaB6', table)
#
# Adjust the position of the moderator. Adjust the position and orentation of banks 42 and 87
CorelliPowderCalibrationCreate(InputWorkspace='LaB6',
                               OutputWorkspacesPrefix='LaB6_',
                               TubeDatabaseDir='/tmp',
                               TofBinning=[3000,-0.001,16660],
                               PeakFunction='Gaussian',
                               PeakPositions=[1.3143, 1.3854,1.6967, 1.8587, 2.0781, 2.3995, 2.9388, 4.1561],
                               SourceMaxTranslation=0.1,
                               ComponentList='bank42/sixteenpack,bank87/sixteenpack',
                               ComponentMaxTranslation=0.02,
                               ComponentMaxRotation=3.0)

A set of output workspaces are created, including the adjustment diagnostic workspaces.

All workspaces are prefixed by LaB6_, which is the value we set for option OutputWorkspacesPrefix when we invoked the algorithm. A description of the output workspaces follows:

Workspace LaB6_adjustments is the main result, a table containing adjustments for different instrument components. In our example, we have adjusted the moderator and banks 42 and 87.

• Xposition, Yposition, Zposition: coordinates for the center of ComponentName in the lab’s frame of reference. Units are in Angstroms.
• XdirectionCosine, YdirectionCosine, ZdirectionCosine: direction cosines in the lab’s frame of reference. They define a rotation axis to set the orientation of ComponentName.
• RotationAngle: rotate this many degrees around the previous rotation axis to set the orientation of ComponentName.

The diagnostics workspaces are stored within WorkspaceGroup LaB6_bank_adjustment_diagnostics. These are:

• LaB6_PDCalibration_peaks_original and LaB6_PDCalibration_peaks_adjustments contains one fitted-intensity spectrum per bank versus d-spacing before and after the banks are adjusted.

As one can see, intensity is displayed only in the neighborhood of those values of d-spacing where peaks are found. In addition, we don’t show the raw intensity, but we substitute each raw peak with a Gaussian fit of the peak. This substitution takes place on every pixel of the bank. Later, we add up all these fitted-intensity peaks for all pixels within one bank.

• LaB6_peak_deviations_original and LaB6_peak_deviations_adjustment finds for each peak the difference between the peak center and the reference d-spacing value we know from the standard. These peak center deviations are collected for all peaks found in the pixels of the bank, and then histogrammed.

Adjusting the bank removes any observed systemic deviation, thus centering the histogram around the no-deviation value. The FHWM of these histograms is only modestly reduced, indicating that the remaining observed deviations are not due to misalignment in the position and orientation of the bank.

• LaB6_percent_peak_deviations_original and LaB6_percent_peak_deviations_adjustment histograms the unitless peak center deviations, this achieved dividing each deviation by its reference d-spacing, i.e. \(\frac{d_{observed} - d_{reference}}{d_{reference}}\). This largely eliminates the scaling of the peak center deviation with the value of the peak center.

Adjusting the bank removes the previously observed systemic deviation and modestly reduces the FWHM of the histogram.

• LaB6_percent_peak_deviations_summary presents basics statistics for the histograms of LaB6_percent_peak_deviations_original and LaB6_percent_peak_deviations_adjustment, namely the average and FWHM of each histogram, as well as the average of the absolute value of the deviations.

From the picture, peaks in bank87 deviated from their reference value between \(mean - FWHM \simeq -1\%\) and \(mean + FWHM \simeq +4.5\%\) before adjustment. After adjustment the peaks deviated between \(-2\%\) and \(+2\%\)

Categories: AlgorithmIndex | Diffraction\Reduction

Source¶

Python: CorelliPowderCalibrationCreate.py (last modified: 2021-01-29)