Modification of the gravitationalDrop method of MatrixWorkspace

[Anton Piccardo-Selg]

Original email from Richard:

Dear all

We are not applying enough gravitational drop in the SANS data reduction, as I had assumed that the beam was horizontal at the sample. In fact it is actually horizontal half way along the beam defining collimation L1, before the sample. (Note here, that as in the Q resolution calculations, L1 is the incident collimation length, which varies 4m to 12m on SANS2d, L1 is not the source to sample distance ~19m on SANS2d).

We are currently using drop = gconst*(wav^2)(L2)2, but need instead gconst*(wav^{2)*(L1*0.5+L2)}2

Usually for the rear detector the sample-detector distance L2 is set the same as L1, so since drop ~ wav^2.L2 we could be off by as much as

{(L2+(L1)/2)/(L2) }^{2 ~ 9/4 = 2.25}

I have fit the beam centre at 11 and 5.5 angstrom for run 28222 (iron oxide) at L2=12m, despite some noisy curves, the difference in beam height is 5.1mm, even with “gravity on”, which is exactly what the correct formula predicts should be the case, as per table at bottom of attached spreadsheet.

I would like to correct the cpp routine at the end of this email by putting in a + LEXTRA to the calculation as shown in double MatrixWorkspace::gravitationalDrop, where LEXTRA defaults to zero and has a value passed from the user file and gui.

LEXTRA is likely L1*0.5 but may be less than this depending on where scraper baffles are inside the collimation.

Can this be done, urgently, please,

Thank you,

Richard

PS alas this does not solve any of our “background” at 12m issue, as the current beam centre is dominated by the long wavelengths, however it will improve the shape of the I(Q).

[Anton Piccardo-Selg]

Refs #11736 Add extra length option for gravity

Changeset: d7d330c4bb9365f6d828423b9f93922e639f6a44

[Anton Piccardo-Selg]

Refs #11736 Add validator for extra length

Changeset: fef75706833db8228f77a437c9ee444e02bdd0f0

[Anton Piccardo-Selg]

This is being verified as ​pull request #792.

[Anton Piccardo-Selg]

Refs #11736 Fix doxygen error

Changeset: 18a5c05e271c5bf9d09252c1b773c39cbb07be4f

[Anton Piccardo-Selg]

Refs #11736 Add system test

Changeset: 62a5344d0bb937954fd0ab723244c7d97ac44346

[Anton Piccardo-Selg]

Refs #11736 Add system test data

Changeset: a5ed868a4e8e34a2840dacf5defcf6a44e89f20e

[Anton Piccardo-Selg]

Refs #11736 Added newline for pylint

Changeset: 89695d4c2e6b0a41bf44d8ef588102d3a0c4c49d

[Anton Piccardo-Selg]

Refs #11736 Fix bug in gravity parsing

Changeset: 276ed88fdcd4d8bd967b5648eb7a8e4d8ec0aa7f

[Martyn Gigg]

Merge pull request #792 from mantidproject/feature/11736_modification_of_gravitational_drop

Modification of gravitational drop in SANS

Full changeset: ​5240467b4ebe996037c56b66fbc478ac6ac3facf

[Stuart Campbell]

This ticket has been transferred to github issue ​12574