The random dipole cloud and time monitors are to make sure that all modes are excited and the related signals are picked up. In case a dipole is locagted at the dark spot (zero intensity) of an mode, it cannot be excited; similarly, if a time monitor is located at the dark spot, it cannot be used to calculate the band. the randomness is to avoid such scenarios. That said, you may need to modify the locations of the dipoles and time monitors to get better results.
In addition, the band structure is calculated from the Fourier transform or findpeaks. If the simulation time is too short, there might be too many points; if the simulation time is too long, the band structure may miss some points, since the signals for small Q modes are only small amount, and the bandstructure is ploted as the intensity of the modes. Sometimes at specific frequency the mode indeed has small Q, then there can be an issue. This is the drawback of time-domain methods fi=or this kind of simulation. fortunately you may adjust dipoles and time monitors.
So three important factors affect better band structure: locatons of dipoles, locations of time monitors and proper simulation time.
You are navigating away from the AIS Discovery experience