The NeXus coordinate system is defined as being identical to McStas.
In P. Willendrup, and K. Lefmann, Journal of Neutron Research, vol. 23, no. 1, pp. 7-27, 2021 you can see that is defined as so:
The component coordinate system can be chosen freely as to whatever is the easy system of the device/problem at hand. However, the usual starting condition in McStas sources is to have z along the neutron beam and y vertical, and since our coordinate system is right-handed x is horizontally transverse to the neutron beam, pointing left as seen from the source.
If the beam travels orthogonally to gravity there is no misinterpretation of this system. However, there are some cases where this is not the case (e.g. incident beams on liquid samples).
The angle of the incoming beam can be described via NXbeam, but by definition the coordinate system ensures the beam always has direction [0 0 1].
NXcoordinate_system could be used to interpret the system, but it assumes the orthogonal assumption always holds (see point 0 at https://manual.nexusformat.org/classes/base_classes/NXcoordinate_system.html ), and there is no other NeXus-standard way to define the gravity vector. Anecdotally, most software using the generated NeXus files is always assuming an exactly-vertical component for gravity too. Additionally, it seems point 0 also states that the direction of Z can be different to that of the beam if it' is not defined in the NXbeam instance'.
I strongly suggest this situation is clarified. Functionally, there should be no difference with a definition that is defined by y, where:
y exactly opposes gravity, positive y pointing upwardsz points along the ideal beam such that x = 0 along a length of beam before the samplex is the remaining orthogonal axis to this system using a RHS.
Alternatively, the transformation should be described for the case where the beam does not travel orthogonally to gravity and instead y is rotated, with some method to record the direction of gravity w.r.t. the beam.
In the case of a vertical neutron source (firing directly downwards rather than e.g. FIGARO at ILL) it should be immediately clear why clarity would be good.
The NeXus coordinate system is defined as being identical to McStas.
In P. Willendrup, and K. Lefmann, Journal of Neutron Research, vol. 23, no. 1, pp. 7-27, 2021 you can see that is defined as so:
The component coordinate system can be chosen freely as to whatever is the easy system of the device/problem at hand. However, the usual starting condition in McStas sources is to have z along the neutron beam and y vertical, and since our coordinate system is right-handed x is horizontally transverse to the neutron beam, pointing left as seen from the source.
If the beam travels orthogonally to gravity there is no misinterpretation of this system. However, there are some cases where this is not the case (e.g. incident beams on liquid samples).
The angle of the incoming beam can be described via
NXbeam, but by definition the coordinate system ensures the beam always has direction [0 0 1].NXcoordinate_systemcould be used to interpret the system, but it assumes the orthogonal assumption always holds (see point 0 at https://manual.nexusformat.org/classes/base_classes/NXcoordinate_system.html ), and there is no other NeXus-standard way to define the gravity vector. Anecdotally, most software using the generated NeXus files is always assuming an exactly-vertical component for gravity too. Additionally, it seems point 0 also states that the direction of Z can be different to that of the beam if it' is not defined in the NXbeam instance'.I strongly suggest this situation is clarified. Functionally, there should be no difference with a definition that is defined by
y, where:yexactly opposes gravity, positiveypointing upwardszpoints along the ideal beam such thatx = 0along a length of beam before the samplexis the remaining orthogonal axis to this system using a RHS.Alternatively, the transformation should be described for the case where the beam does not travel orthogonally to gravity and instead
yis rotated, with some method to record the direction of gravity w.r.t. the beam.In the case of a vertical neutron source (firing directly downwards rather than e.g. FIGARO at ILL) it should be immediately clear why clarity would be good.