SPH zero time increment estimate

Does anyone have any idea what reasons can cause the SPH modelling to run with "zero time increment estimate" error? I read some posts saying that "zero time increment estimate" may caused by the distortion of element such that at least one element because too small. The FEM analysis requires the wave to not transport over 1 element in one time increment, thus a very small element may cause very small time increment.

I don't know how to analogy this error to SPH analysis. Anyone who knows the theory behind it and how to avoid such an error?


Thanks,