I have created a simulation of an echo pulse test (5 MHz) in k-wave. I used different element sizes (10 - 40 um) and time steps (3 - 12 ns). Unfortunately, I have noticed that the TOF changes in the range of almost 5% and the amplitudes by up to 40%.
I use the following solver: kspaceFirstOrder2D
Had someone similar problems or know what the error can be?