Python based Monte Carlo simulation of a test beam setup. An EUDET-type beam telescope is implemented. The telescope consists of MIMOSA26 planes and an ITkPix as a time reference, but other devices can easily be added. Numba allows for performant calculations, drastically reducing compilation time. All physics cases are greatly simplified.
The particle beam consists of electrons with a given rate, beam profile, dispersion, angle, and total number of particles. Each particle loses energy according to a Landau distribution and scatters according to a Gaussian distribution, with the width calculated using the Highlander formula. Scattering occurs at an infinitely thin sheet at the device surface, and there is no scattering in the air between the devices.
The devices have rectangular pixels and a mostly uniform material budget, and they are placed in the center of the beam, though they can be shifted off-center. The clusters are calculated by sampling a Gaussian distribution around the particle intersection point, with a width taking diffusion and Coulomb expansion into account. Each device can be either triggered or untriggered; this only affects part of the hit data collection, as each beam particle has a given chance to be triggered.
pytestbeam and all the requirements are installed with:
git clone https://github.com/rpartzsch/pytestbeam
cd pytestbeam
pip install -e .Create your test beam setup in setup.yml, and then execute the following command:
python pytestbeam.pyMaterials are characterized in materials.yml. To add, for example, an absorber to the setup, include a device with a single row and column of pixels made from the specified material in setup.yml.
The output data is saved in HDF5 format.