Defining the location of a detector, or any generic object, is potentially challenging. It needs to be agreed upon how to define the location of an object, specifically detectors, and their phsyical extent. Proposals should be able to define a position unambigiously in a way that is easily interpreted by analysis and plotting software.
Taken from the McStas manual p45:
The instrument is given a global, absolute coordinate system. In addition, every component in the instrument has its own local coordinate system that can be given any desired position and orientation (though the position and orientatino must remain fixed for the duration of a single simulation). By convention, the z axis points in the direction of the beam, the x axis is perpendicular to the beam in the horizontal plane pointing left as seen from the source, and the y axis points upwards (see figure 5.1). Nothin in McStas enforces this convention, but if every component used different conventions the user would be faced with a severe headacha! It is therefore recommended that this convention is followed by users implementing new components.
The x axis points in the direction of the beam, the z axis up, and the y axis to the left while facing down-stream to complete the right handed coordinate system.
Orientation is given by two angles which define a component’s normal direction. phi is the angle between the x axis and the normal, projected onto the xy plane. theta is the angle between the xy plane and the normal. This does assume that a normal direction is well defined.
Peter Peterson Thursday, 28 August 2003, 12:04:51 pm
01/2015 This got nowhere. It is also now obsolete as NIAC is under the impression that the CIF-style coordinate description allows us to define the position and orientation of detector elements in space in an adequate and exact way.