Explore: check units to be used in QENS models

[celinedurniak]

Check if the current units used in the models are the most appropriate.
Investigate if some conversion tools should be provided.
This point should also be raised during our user test's sessions.

[gonzalezma]

Attached a personal proposition of what to do about the units.
QENSmodels_units.pdf

[Anders-Markvardsen]

@gonzalezma nice exploration of units

[Anders-Markvardsen]

Celine, Tom and Anders agree on Miguel proposal

[Anders-Markvardsen]

In terms a tool for converting units (when we to that) could be to use what is done in MDMC

[Anders-Markvardsen]

Further to above Miguel writes "If we decide to offer some conversion mechanism, we need to find how to handle it. One proposition:
In the script file, the user adds a line e.g.
input_units = [‘inv_angstrom’, ‘meV’] to indicate the units employed in the data for Q and ħomega/omega.
Acceptable units could be:
‘inv_angstrom’, ‘inv_nm’, ‘inv_cm’, ‘inv_m’ for Q
‘meV’, ‘microeV’, ‘rad/ps’, ‘rad/s’, ‘Hz’, ‘THz’, ‘GHz’, ‘inv_cm’, ? for ħomega/omega
And another line specifying the preferred output units for length and time, e.g.
output_units = [‘m’, ‘s’], so in this case the self-diffusion coefficient would be returned in m2/s.
Or output_units = [‘angstrom’, ‘inv_meV’] and D would be returned in Å2*meV.
Valid length units could be ‘angstrom’, ‘pm’, ‘nm’, ‘microm’, ‘mm’, ‘cm’, ‘m’.
Valid time units could be ‘s’, ‘ps’, ‘ns’, ‘inv_meV’, ‘inv_microeV’
This would require that for each output parameter we know its dimensions, e.g. we need to know that the dimensions of D are L2 T^-1. Note that in the document, I’ve differentiated between energy and 1/time for correctness, but internally I think that we could work simply with L, T, L^-1, and T^-1.
"