Engin-X - Umbrella Ticket

[Peter Parker]

An "umbrella" ticket for any work to be done for Engin-X.

First task is to liaise with Joe Kelleher and Saurabh Kabra to see what needs to be done.

[Peter Parker]

Nick and I met with Joe and Saurabh on 2013/11/22:

• Both Joe and Saurabh were very enthusiastic about getting their existing reduction into Mantid.

• Their code is OpenGENIE based, and split into three main chunks.

• They were keen to have us work on the "underlying algorithms" while they spend time with Balsamiq to mock up some potential GUI interfaces which could then be used to wrap the functions.

• They suggested that it also might be helpful with the transition to have reader/writer algorithms that work with a current intermediate file type they use, called "G3 intermediate". A potential use for these algorithms would be to allow event data to be translated within Mantid to histo data, and then used in their current OpenGENIE reduction. Further inspection of the code will reveal how much of a priority this is. Nick mentioned that Freddie would be a good person to talk to about this format, should help be needed.

• Joe and Saurabh pointed out that there are things about their current reduction they'd like to change and improve upon. We stressed that we are more than happy to take things forward and make improvements once the initial "translation" has been done.

I also have in my possession a printout of some details they've typed up about their current reduction. I will add this in a future comment.

[Peter Parker]

Information provided to us about the current reduction:

There are three main functions in OpenGENIE at the moment.

1. Analyse Scan

The most extensive of the three programs. Linked to Calibrate.

Takes a range of runs. For each run, apply intensity normalisation, export to GSAS.EXP file with relevant calibration values, refine the lattice parameter in GSAS, read the result back in from the .EXP file, average the associated relevant log values, and collect these result for all runs in the range into a single structured .HDF.

2. Find Pin

The simplest of the three programs.

This takes a range of runs and averages the intensity value for each of the two banks. Plots this intensity against some logged value, usually a motor position. Fits a function, usually a Guassian, to the intensity value and reports the parameters of the fit. The current function takes its name from the usual use of finding the position at which a metal pin, when scanned across the beam, gives the maximum intensity.

3. Calibrate

The best place to start translating.

Takes one or both of two runs - a null scatterer which is used to measure the incident beam intensity, and a measurement of a known powder (one of 4 or 5 different possibilities, usually CeO2 in practice to determine the relation between the measured time of flight and the d-spacing in the sample. The calibration date derived from this procedure is subsequently used by any subsequent analyse_scan command.

In Mantid, they would envisage having an Engineering menu in Interfaces, with the above three options. The aim should not necessarily be to replicate the existing functionality of OpenGENIE, where obvious improvement can be made. The current interface is a console-based interactive utility -- in Mantid a GUI with a "wizard" or tabs for each section of the console-based program might be a good starting point.

[Nick Draper]

bulk move to assigned at the into of the triage step

[Nick Draper]

Moved to R3.5 at the R3.4 code freeze

[Federico M Pouzols]

Added new ENGIN-X tickets that have been opened in the last few months as 'blocked by'.

[Stuart Campbell]

This ticket has been transferred to github issue ​9213