Ticket #6569 (closed: fixed)
add option in SassenaFFT to fourier transform only the real part of the structure factor
| Reported by: | Jose Borreguero | Owned by: | Jose Borreguero |
|---|---|---|---|
| Priority: | major | Milestone: | Release 2.5 |
| Component: | Mantid | Keywords: | |
| Cc: | Blocked By: | ||
| Blocking: | Tester: | Martyn Gigg |
Description
This option is advantageous when I(Q,t)=Real+i*Imag but we know that Imag should be zero due to spherical average in Q-vector space. Sassena does a numerical spherical average which could result in Imag residual part in I(Q,t). This residual introduces artifacts in S(Q,E). By setting this option to True, we enforce that I(Q,t) is real.
Attachments
Change History
comment:2 Changed 8 years ago by Jose Borreguero
REfs #6569 Added option
modified: SassenaFFT.cpp
Changeset: 913a0b2d5ea235d83e629148b05aaebe8e431e67
Changed 8 years ago by Jose Borreguero
- Attachment example.h5 added
Sassena output file containing I(Q,t>0)
comment:3 Changed 8 years ago by Jose Borreguero
Refs #6569 Added wiki header and FFTonlyRealPart property
Changeset: ecf552deee2073e6679e35dd2869451656b05745
comment:4 Changed 8 years ago by Jose Borreguero
Refs #6569 Updated the header
Changeset: e85ad625db9b4f46da38f7c983a9e9c054dc753d
comment:5 Changed 8 years ago by Jose Borreguero
Refs #6569 Updated header
Changeset: 566be8350179b029a9a9cd91c90f9cda6fc7d1d8
comment:6 Changed 8 years ago by Jose Borreguero
To test: download attached files to /tmp. Run example.py in a Mantid session to load file example.h5 and apply SassenaFFT. The workspaces to look at after the script is ran are "transformed_of_real_part" and "transformed_of_real_plus_imaginary".
The input intermediate structure factor I(Q,t) of example.h5 is composed of a real part (a Gaussian in time, panel a in expected.jpg) and a imaginary part ( time*Gaussian(time), , panel b in expected.jpg ). The Fourier transform of only the real part is another Gaussian. Workspace "transformed_of_real_part" should be a Gaussian (panel c in expected.jpg). The Fourier transform of the real plus the imaginary part, workspace "transformed_of_real_plus_imaginary",is a Gaussian with a dip at negative energies (panel d in expected.jpg)
Plot these two workspaces and compare with expected shapes of figure expected.jpg.
comment:7 Changed 8 years ago by Jose Borreguero
- Status changed from accepted to verify
- Resolution set to fixed
comment:8 Changed 8 years ago by Martyn Gigg
- Status changed from verify to verifying
- Tester set to Martyn Gigg
comment:9 Changed 8 years ago by Martyn Gigg
- Status changed from verifying to closed
Verified that the transform is free of the "dip" to negative energy when the new option is checked.
comment:10 Changed 5 years ago by Stuart Campbell
This ticket has been transferred to github issue 7415
