Selecting the correct phototube model

The command file PMT_SIMULATION.CMD can be used to select the desired complexity for the phototube simulation. As you might expect, there is the usual trade off between accuracy/complexity and speed. With reference to the command file, the first choice is the disk complexity - you get three choices: black, grey and transparent.

Black

means that all photons that intersect the disk are counted. If you select black disks be sure to set the Cerenkov correction factor to 1.0 in the MCMA bank.

Grey

has the average angular and wavelength response folded in, together with a simple simulation of reflections.

Transparent

disk gives access to the three-d full pmt simulation.

Note that, although this is stored as a geometry complexity switch it does not, in fact, control the geometry complexity!; it is stored here for historical reasons. The geometry is always full three-d; this switch is only used used to divert photons, incident on a PMT, to the black or grey model approximations. This means that if light is generated from within a PMT, it will be modelled in three-d within it regardless of the setting of this switch but, should it escape, will only `see' other PMTs of the selected complexity.

The second switch controls what happens at the photocathode. Again you get black, grey or transparent. Black means that any photon which reaches the photocathode is counted. Grey means that the the photon may either be transmitted, reflected or absorbed (ie a hit), however, the internal structure of the phototube is ignored so all transmitted photons are thrown away. Transparent is as grey, but the internal structure is included and transmitted photons will be followed. For general use there should be no reason to change this switch from its transparent setting, but it may be helpful in understanding some detector phenomena.

The reflectors also have two switches. The first covers whether the petals are included, or whether the smooth theoretical shape is used. The second covers whether the omega reflectivity is uniformly 0.9, or the wavelength and angle dependency is implemented. Once again the trade off is one of speed versus complexity.

The relative volumes of all regions internal to a PMT are set to unity. It is unnecessary to calculate them as they would only be used to get the correct weights if trying to pick points inside a PMT and outside a PMT simultaneously, something that makes no sense. Points can be picked within a specfic PMT by giving its serial number, or within any PMT by setting the serial number to zero.