[Yaroslav]

Hello.
I don't know how to solve my problem.
I have a closed ring-shaped collector through which liquid fuel flows. the fuel is supplied from below and will fill the collector completely under pressure.
But I'm trying to make a calculation in which the collector will be filled by about half. that is, the fuel will not flow out of all the injectors.
I have already managed to build a grid (57 million cells), I also made a steady calculation of this collector at different operating pressures ( on these parameters, the collector is completely filled, the pressure is high >10 atm). The steqdy calculation was based on the pressure drop and the mass fuel consumption obtained from the calculations coincided with the practical one.
I have never performed a unsteady calculation and have never done multiphase calculations. so I started experimenting. A unsteady problem with the same parameters as a staeady problem is considered good. But if I connect a multiphase. then I count 5-6 iterations and everything collapses.
The main errors (1. a floating-point error occurred; 2. Amg coarsening halted....)
And now I'm experimenting and nothing is working. I'm not even sure what physics I need to use to complete my task. therefore, I ask you to help me with the task..