[vedmak]

Greetings my friends,


I`m working on a somewhat unique problem and as you might imagine, help is needed



I`ve got a tank, which will be filled with gas. The gas flows in through a valve which is mounted directly to the tank and has an approximately 140mm long injection pipe with a constant cross area inside the tank. The flow path inside the valve is quite comlicated. The valve is always open completely.



The only things I know: Initial state (p, T) of the entire system and the inlet pressure at the valve inlet (given by a arbitary constant pressure ramp, delta p/delta t).


In order to save simulation time, I think about calculating the mass flow though the valve separated from the tank. Then use that mass flow data as inlet BC for the tank (with ONLY an injection pipe).



But now, the question emerges as to what outlet boundary condition might be appropriate, as I don`t know anything about the mass flow rates.



Inital idea: use the ideal gas eq. p=rho*R*T as the outlet pressure, where p is the pressure in the tank. rho=(mass_initial+mass_inflow)/Volume_tank.


The question is: is this legit? And if so: would it be legit for any pressure ramp, or is there a limit, since if increasing the pressure at the valve inlet too fast, I imagine Ma number at the valve outlet will reach 1.0 (accroding to my superviser we`ll even see supersonic flow??) But in this case there will be a pressure shock at the valve outlet (like with an underexpanded nozzle) since the pressure in the tank can`t homogenize fast enough, or can it?




Sorry, if anything is formulated in a confusing way. Will explain again if wished.



I really really would appreciate any help on that topic.