I'm trying to simulate a monopole moving back and forth in air along an arc. I can get the simulation to run fine when the timestep is made by kgrid.makeTime, but the issue is I would like it to take a certain amount of time (0.01 seconds) for the monopole to move from one end of the arc to the other. Even when the CFL condition is pushed so that Cmax = 0.9, the timestep made by kgrid.makeTime is too small such that dt * the number of nodes along the arc << 0.01 seconds. I have already increased the number of points along the arc to be the maximum allowable by the grid size. I could decrease the grid size to allow for more points along the arc, but this would necessitate a smaller dt from the CFL condition.
My questions are as follows:
1. The first and most promising solution I thought of was to decrease the speed of sound in the simulation. Unfortunately, in order for the times to match, I would need to decrease the sound speed to ~1m/s. The simulation runs, but is this reduction of the sound speed too un-physical? What are the ramifications of decreasing the sound speed to such an extent?
2. The second solution I thought of was to have the monopole linger at each node for long enough for the times to match. However, this leads to ripples that occur each time the source moves nodes after lingering for ~200 timesteps. Is there a better approach to accomplish this?
Thank you!