Thank you @Anonymous for your time and attention, but sorry. I didn't comprehend exactly what happens in your file. 😁
In my computer, It seems that the upper part of the sphere creates particles that are being blocked by the face removed from the cone trunk (cup), while the particles generated in the lower part of the sphere go to the bottom of the glass, although it does not reach it nor fill the glass. 🤔
But the file helped to understand the limitation of the simulation: the emitter can fill the container only if it is not inside the liquid. That is, by this solver, if a pool is supplied by a pipe below the existing water level, it will not fill and overflow. But if the emitter is above, it will work.
😐
I attached an example file.
Other discoveries that I made:
The option Delete Exceeding Particles (Simulation Parameters > Solver Properties) makes a considerable difference in the container filling (time to fill and liquid volume).
Fluid velocity and blocker thickness also can influence the result.
Emitter works better with closed 3d Objects (opened faces are ignored) bigger than the voxel size.
What generated this post was that I had the misfortune of having almost all these problems in the same place and time: a tiny emitter with deleted faces on the bottom part of the container (I was thinking about letters being filled by water). And in another test, I should have used a fast emission on a thin blocker.
So it stays the suggestion for the developers (emission from bellow fluid surface).
