\(\renewcommand\AA{\unicode{x212B}}\)

D7YIGPositionCalibration v1

../_images/ImageNotFound.png

Enable screenshots using DOCS_SCREENSHOTS in CMake

Summary

Performs D7 position calibration using YIG scan and returns D7 IPF readable by LoadILLPolarizedDiffraction.

See Also

LoadILLPolarizedDiffraction, PolDiffILLReduction, D7AbsoluteCrossSections

Properties

Name

Direction

Type

Default

Description

Filenames

Input

list of str lists

The file names with a single YIG scan.

InputWorkspace

Input

MatrixWorkspace

The name of the workspace containing the entire YIG scan.

YIGPeaksFile

Input

string

C:/Repos/GitHub/conda/mantid/instrument/D7_YIG_peaks.xml

The file name with all YIG peaks in d-spacing. Allowed extensions: [‘.xml’]

ApproximateWavelength

Input

number

3.1

The initial guess for the neutrons’ wavelength

MinimalDistanceBetweenPeaks

Input

number

3

The minimal allowable distance between two YIG peaks (in degrees 2theta).

BankOffsets

Input

dbl list

3,3,0

List of values of offset for each bank (in degrees).

BraggPeakWidth

Input

number

2

An initial guess for the width of YIG Bragg peaks used for fitting (in degrees).

MaskedBinsRange

Input

dbl list

-35,15

The lower and upper bound for the masked region around the direct beam (in degrees).

CalibrationOutputFile

Input

string

The output YIG calibration Instrument Parameter File. Allowed values: [‘xml’]

ClearCache

Input

boolean

True

Whether to clear the intermediate fitting results.

FitOutputWorkspace

Output

TableWorkspace

The table workspace name that will be used to store all of the calibration parameters.

FittingMethod

Input

string

None

Option to provide the initial guesses or perform fits for Bragg peaks either individually or globally. Allowed values: [‘None’, ‘Individual’, ‘Global’]

Description

This is the algorithm that performs wavelength and position calibration for both individual detectors and detector banks using measurement of a sample of powdered \(\text{Y}_{3}\text{Fe}_{5}\text{O}_{12}\) (YIG). This data is fitted with Gaussian distributions at the expected peak positions. The output is an Instrument Parameter File readable by the LoadILLPolarizedDiffraction algorithm that will place the detector banks and detectors using the output of this algorithm. The algorithm implements method described in Ref. [1].

It is crucial for a reliable calibration to first run a test without setting the BankOffsets parameter and with FittingMethod set to None. This allows checking the positions of the initial guesses for the positions of YIG Bragg peaks in the conjoined_input workspace. Then, the BankOffsets can be obtained by comparing the peak positions coming from the detectors with their expected location, coming from known d-spacings for YIG.

Currently the algorithm is focused on the D7 instrument that has three detector banks and a monitor, and all of them can move individually. The detector position and wavelength calibration have to be checked each time the used wavelength is changed, since the instrument has to be manually moved into a different parking position around the monochromator.

The property MaskedBinsRange allows to mask multiple \(2\theta\) ranges of the \(2\theta\) scan. If the number of provided arguments is even, arguments will be paired and the angular range between the first and the second element of the pair will be masked, and for all pairs. In the case where the number of arguments is odd, the first argument is assumed to be a lower cut-off, and all \(2\theta\) detector positions below that value will be masked.

The values of the fitted wavelength, three bank gradients, and the \(chi^{2}\) value are logged on the notice level.

Calibration method

A YIG scan data is loaded into a single 2D Workspace where the X axis contains detector positions during each step of the scan, and the Y axis the measured intensities normalized to monitor. The Y axis data is also 2D with each row representing full scan for a single detector. It is recommended to mask scan points located between -35 and 15 degrees to remove the impact of the direct neutron beam.

The provided YIG d-spacing values are loaded from an XML list. The default d-spacing is coming from Ref. [2]. The peak positions are converted into 2theta positions using the initial assumption of the neutron wavelength. All peaks that would require the \(2\theta\) to be above 180 degrees are removed.

YIG peaks in the detector’s scan are fitted using a Gaussian distribution. The Bragg peaks are fitted either separately, when the FittingMethod property is set to Individual, or all peaks measured by a detector are fitted together, when the FittingMethod property is set to Global. The fitting results are stored in a new workspace, where the Y axis contains the fitted peak centres and the X axis the calculated peak positions.

The workspace containing the peak fitting results is then fitted using a Multidomain function of the form:

\[2\theta_{\text{fit}} = m \cdot (2.0 \cdot \text{asin} ( \lambda / 2d ) + offset_{\text{pixel}} + offset_{\text{bank}}),\]

where m is the bank slope, \(offset_{\text{pixel}}\) is the relative offset to the initial assumption of the position inside the detector bank, and \(offset_{\text{bank}}\) is the offset of the entire bank. This function allows to extract the information about the wavelength, detector bank slopes and offsets, and the distribution of detector offsets.

Usage

Note

To run these usage examples please first download the usage data, and add these to your path. In Mantid this is done using Manage User Directories.

Example - D7YIGPositionCalibration - calibration at the shortest wavelength

approximate_wavelength = '3.1' # Angstrom
D7YIGPositionCalibration(Filenames='402652:403041', ApproximateWavelength=approximate_wavelength,
                         YIGPeaksFile='D7_YIG_peaks.xml', CalibrationOutputFile='test_shortWavelength.xml',
                         MinimalDistanceBetweenPeaks=1.5, BankOffsets=[3, 3, 1], ClearCache=True,
                         FittingMethod='Individual', FitOutputWorkspace='shortWavelength')

print('The calibrated wavelength is: {0:.2f}'.format(float(approximate_wavelength)*mtd['shortWavelength'].column(1)[1]))
print('The bank2 gradient is: {0:.3f}'.format(1.0 / mtd['shortWavelength'].column(1)[0]))
print('The bank3 gradient is: {0:.3f}'.format(1.0 / mtd['shortWavelength'].column(1)[176]))
print('The bank4 gradient is: {0:.3f}'.format(1.0 / mtd['shortWavelength'].column(1)[352]))

Example - D7YIGPositionCalibration - calibration at the intermediate wavelength

Load('ILL/D7/396442_396831.nxs', OutputWorkspace='intermediateWavelengthScan')
approximate_wavelength = '4.8' # Angstrom
D7YIGPositionCalibration(InputWorkspace='intermediateWavelengthScan', ApproximateWavelength=approximate_wavelength,
                         YIGPeaksFile='D7_YIG_peaks.xml', CalibrationOutputFile='test_intermediateWavelength.xml',
                         MinimalDistanceBetweenPeaks=1.5, BankOffsets=[3, 3, -1],
                         MaskedBinsRange=[-50, -25, 15],
                         FittingMethod='Individual', FitOutputWorkspace='intermediateWavelength')

print('The calibrated wavelength is: {0:.1f}'.format(float(approximate_wavelength)*mtd['intermediateWavelength'].column(1)[1]))
print('The bank2 gradient is: {0:.1f}'.format(1.0 / mtd['intermediateWavelength'].column(1)[0]))
print('The bank3 gradient is: {0:.1f}'.format(1.0 / mtd['intermediateWavelength'].column(1)[176]))
print('The bank4 gradient is: {0:.1f}'.format(1.0 / mtd['intermediateWavelength'].column(1)[352]))

Output:

The calibrated wavelength is: 4.8
The bank2 gradient is: 1.0
The bank3 gradient is: 1.0
The bank4 gradient is: 1.0

Categories: AlgorithmIndex | ILL\Diffraction

Source

Python: D7YIGPositionCalibration.py