I‘m confused about the difference between GenericPropagation and TransientPropagation in the simulation of a single pixel.
The drift time is nearly the same using the two methods. But the amplitude of the pulses in CSA are different.
I will appreciate it if someone could answer my questions and tell me which method is more accurate in the simulation of pulse signal.
The Drift Time in GenericPropagation (left) and TransientPropagation (right).
there indeed is a difference between the two modules, tl;dr the TransientPropagation module is the one expected to be more precise.
While both modules emply a Runge-Kutta step-wise integration of the charge carrier motion, it differs what they do apart from the motion itself:
GenericPropagation moves the charge carriers around the sensor until they reached a stop condition such as a sensor boundary. At that point, a PropagatedCharge object is created and the final time of the mtion is assigned to the object. Subsequently, the PulseTransfer module finds this charge carrier, sees that no explicit pulse information is attached to it and builds a total current pulse by using the final arrival time of the charge carrier at its position. This is an approximation which might provide reasonable results in some cases.
TransientPropagation also moves around the charge carriers, however, for every single step of the motion, the induced current in allsurrounding electrodes is calculated by means of the Shockley-Ramo theorem. That is the reason why a weighting potential is required for this module. After the motion has stopped, the PulseTransfer module finds the charge carriers, detects the explicit pulse information and builds the final electrode current pulse by summing the individual pulse contributions from all charge carriers.
This is why the drift time (using very similar mechanics) is almost the same (there is the tiny difference that GenericPropagation uses an adaptive time step size while for TransientPropagation it is fixed) while the resulting pulse looks different.
Thanks for your answer, which solves my question accurately. Allpix Squared is really great, but I need to learn some knowledge about signal formation in order to understand it more deeply.
