SANSILLAutoProcess v1¶

Summary¶

Performs complete SANS data reduction at the ILL.

Properties¶

Description¶

This algorithms performs complete treatment of SANS data recorded with the ILL instruments D11, D22 and D33. This high level algorithm steers the reduction and performs the full set of corrections for a given sample run; measured with one or more detector distances.

The sample measurement will be corrected for all the effects and converted to Q-space, producing by default the azimuthal average curve \(I(Q)\). One can use water reference measurement in order to derive the relative inter-pixel sensitivity map of the detector or the reduced water data for absolute normalisation of the subsequent samples. The regular output will contain fully corrected water run, and there will be an additional output containing the sensitivity map itself. The sensitivity map, as well as reduced water can be saved out to a file and used for sample reductions.

OutputBinning¶

This property can be used to set manual q-binning parameters per detector distance/configuration. This accepts a string, which must be : delimited sets of binning parameters per distance. The reason it is a string, is that at a given configuration the parameters set can be composed of , separated values (such as start, width, end). In fact, it could be even more complicated. See the semantics in SANSILLIntegration. If one parameter set is provided, i.e. there is no :, the same parameter set will be used for all the detector distances/configurations.

Caching with ADS¶

This algorithm does not clean-up the intermediate workspaces after execution. This is done intentionally for performance reasons. For example, once the transmission of a sample is calculated, it will be reused for further iterations of processing of the same sample at different detector distances. As other example, once the container is processed at a certain distance, it will be reused for all the subsequent samples measured at the same distance, if the container run is the same. The same caching is done for absorber, empty beam, container, sensitivity and mask workspaces. The caching relies on Analysis Data Service (ADS) through naming convention by appending the relevant process name to the run number. When multiple runs are summed, the run number of the first run is attributed to the summed workspace name.

Example - full treatment of 3 samples at 3 different distances in D11

beams = '2866,2867+2868,2878'
containers = '2888+2971,2884+2960,2880+2949'
container_tr = '2870+2954'
beam_tr = '2867+2868'
samples = ['2889,2885,2881',
           '2887,2883,2879',
           '3187,3177,3167']
sample_tr = ['2871', '2869', '3172']
thick = [0.1, 0.2, 0.2]

# reduce samples
for i in range(len(samples)):
    SANSILLAutoProcess(
        SampleRuns=samples[i],
        BeamRuns=beams,
        ContainerRuns=containers,
        MaskFiles='mask1.nxs,mask2.nxs,mask3.nxs',
        SensitivityMaps='sens-lamp.nxs',
        SampleTransmissionRuns=sample_tr[i],
        ContainerTransmissionRuns=container_tr,
        TransmissionBeamRuns=beam_tr,
        SampleThickness=thick[i],
        OutputWorkspace='iq_s' + str(i + 1)
    )

print('Distance 1 Q-range:{0:4f}-{1:4f} AA'.format(mtd['iq_s1_#1_d39.0m_c40.5m_w5.6A'].readX(0)[0],
                                                   mtd['iq_s1_#1_d39.0m_c40.5m_w5.6A'].readX(0)[-1]))
print('Distance 2 Q-range:{0:4f}-{1:4f} AA'.format(mtd['iq_s1_#2_d8.0m_c8.0m_w5.6A'].readX(0)[0],
                                                   mtd['iq_s1_#2_d8.0m_c8.0m_w5.6A'].readX(0)[-1]))
print('Distance 3 Q-range:{0:4f}-{1:4f} AA'.format(mtd['iq_s1_#3_d2.0m_c5.5m_w5.6A'].readX(0)[0],
                                                   mtd['iq_s1_#3_d2.0m_c5.5m_w5.6A'].readX(0)[-1]))

Output:

Distance 1 Q-range:0.001350-0.020028 AA
Distance 2 Q-range:0.007625-0.092061 AA
Distance 3 Q-range:0.032571-0.343635 AA

Example 2 - full treatment of 5 samples in D33

from mantid.simpleapi import *
import matplotlib.pyplot as plt

config.setFacility('ILL')
config['default.instrument'] = 'D33'
config.appendDataSearchSubDir('ILL/D33/')

absorber = '002227'

tr_beam = '002192'

can_tr = '002193'

empty_beam = '002219'

can = '002228'

mask = 'D33Mask2.nxs'

sample_names = ['H2O', 'D2O', 'AgBE', 'F127_D2O', 'F127_D2O_Anethol']
sample_legends = ['H$_2$O', 'D$_2$O', 'AgBE', 'F127 D$_2$O', 'F127 D$_2$O Anethol']

samples = ['002229', '001462', '001461', '001463', '001464']

transmissions = ['002194', '002195', '', '002196', '002197']

# Autoprocess every sample
for i in range(len(samples)):
    SANSILLAutoProcess(
        SampleRuns=samples[i],
        SampleTransmissionRuns=transmissions[i],
        MaskFiles=mask,
        AbsorberRuns=absorber,
        BeamRuns=empty_beam,
        ContainerRuns=can,
        ContainerTransmissionRuns=can_tr,
        TransmissionBeamRuns=tr_beam,
        CalculateResolution='None',
        OutputWorkspace=sample_names[i]
    )

fig, ax = plt.subplots(subplot_kw={'projection':'mantid'})

plt.yscale('log')
plt.xscale('log')

# Plot the result of every autoprocess
for wName in sample_names:
    ax.errorbar(mtd[wName][0])

plt.legend(sample_legends)
ax.set_ylabel('d$\sigma$/d$\Omega$ ($cm^{-1}$)')

#fig.show()

(Source code, png, hires.png, pdf)

Categories: AlgorithmIndex | ILL\SANS | ILL\Auto

Source¶

Python: SANSILLAutoProcess.py