506
J. Apostolakis
simple model or parameterization is produced for the signal generation given the
energy deposition.
The simulation involves a level of detail beyond other Monte Carlo simulation
for HEP detectors. Some of the important aspects include:
• modeling the generation of all secondaries, without an energy threshold, in every
single inelastic collision in the gaseous volume;
• the effects of elastic collisions in the transport;
• for efficiency, pre-calculating the convolution of the effects of the resulting
diffusion and the drift in the electromagnetic fields of the detector;
• the effect of potential build-up of charge on detector elements, e.g. the space
charge in a gas.
Due to the need to simulate down to the eV scale, this simulation requires
detailed knowledge of the excitations of the molecular constituents of the mixture.
Specialised programs are necessary for this simulation. Garfield calculates the
electric field in many regular cell geometries. Then it combines it with the energy
deposition for each atomic shell via a specialized Photo Absorption Ionization
model integrated from Heed [101] to generate all secondary gammas and electrons.
Transport of the charged particles in the electromagnetic field is coupled with
diffusion, using pre-calculated transport coefficients generated by the Magboltz
code [102].
It is possible to simulate an avalanche near a sensor wire in order to accurately
model the signal arriving at the detector’s electronics. This is typically required
only for the detailed understanding of the effects of the shape of the signal and the
integration characteristics of the electronics, and requires three orders of magnitude
more computation than the simulation of the energy deposition in the gas volume.
Alternatively, a fraction of the track can be simulated, in order to determine key
characteristics, such as the arrival time of the signal. This reduces the computing
requirements by about an order of magnitude.
It is also possible to calculate the effect of charged particles on an integrated
circuit element [103]. Using the energy deposition to create electron–hole pairs,
an external Technical Computer Aided Design (TCAD) program simulates the
detailed response of the circuit. Applications of this technique have focused on
the simulation of single event upsets [104], in which a cosmic ray track results
in the flipping of a bit in a silicon circuit. Typically, the circuit response involves
proprietary TCAD programs.
11.4.3 Production Thresholds and Other Biasing Techniques
For many setups, computing resources for simulation in full detail are not available.
For example, large-scale experiments can require millions of events in order to
establish patterns related to rare processes, yet the full simulation of events is
prohibitive. In this case the choice must be made how to discard particular tracks in
J. Apostolakis
simple model or parameterization is produced for the signal generation given the
energy deposition.
The simulation involves a level of detail beyond other Monte Carlo simulation
for HEP detectors. Some of the important aspects include:
• modeling the generation of all secondaries, without an energy threshold, in every
single inelastic collision in the gaseous volume;
• the effects of elastic collisions in the transport;
• for efficiency, pre-calculating the convolution of the effects of the resulting
diffusion and the drift in the electromagnetic fields of the detector;
• the effect of potential build-up of charge on detector elements, e.g. the space
charge in a gas.
Due to the need to simulate down to the eV scale, this simulation requires
detailed knowledge of the excitations of the molecular constituents of the mixture.
Specialised programs are necessary for this simulation. Garfield calculates the
electric field in many regular cell geometries. Then it combines it with the energy
deposition for each atomic shell via a specialized Photo Absorption Ionization
model integrated from Heed [101] to generate all secondary gammas and electrons.
Transport of the charged particles in the electromagnetic field is coupled with
diffusion, using pre-calculated transport coefficients generated by the Magboltz
code [102].
It is possible to simulate an avalanche near a sensor wire in order to accurately
model the signal arriving at the detector’s electronics. This is typically required
only for the detailed understanding of the effects of the shape of the signal and the
integration characteristics of the electronics, and requires three orders of magnitude
more computation than the simulation of the energy deposition in the gas volume.
Alternatively, a fraction of the track can be simulated, in order to determine key
characteristics, such as the arrival time of the signal. This reduces the computing
requirements by about an order of magnitude.
It is also possible to calculate the effect of charged particles on an integrated
circuit element [103]. Using the energy deposition to create electron–hole pairs,
an external Technical Computer Aided Design (TCAD) program simulates the
detailed response of the circuit. Applications of this technique have focused on
the simulation of single event upsets [104], in which a cosmic ray track results
in the flipping of a bit in a silicon circuit. Typically, the circuit response involves
proprietary TCAD programs.
11.4.3 Production Thresholds and Other Biasing Techniques
For many setups, computing resources for simulation in full detail are not available.
For example, large-scale experiments can require millions of events in order to
establish patterns related to rare processes, yet the full simulation of events is
prohibitive. In this case the choice must be made how to discard particular tracks in