https://knowledge.autodesk.com/support/cfd/learn-explore/caas/CloudHelp/cloudhelp/2019/ENU/SimCFD-UsersGuide/files/GUID-9C1E28D5-B32E-4546-8620-7F428982184A-htm.html

There are some misunderstandings. Mass is always conserved and the inlet+outlet + accumulation (d(rho)/dt) = 0. In an incompressible flow inlet =outlet and the TOTAL mass flow rate can be specified at outlet equals to inlet. It is physically right.

The difficult in two-phase flow is WHICH part of outlet each portion of mass outflows.They will depends on the volume fraction and velocities fields.

The outlet is projected to be a "passive condition" if some prescription is made concerning to mass or volume fractions, the solution will change to respected some profile defined by you, that is, you need to give the outlet profile to volume fractions - what can be made to channel flows, or free-surface flows in general. But is not ease or can be applied to any multi-phase flow. Sometimes it will be impossible.

Opening condition is a outlet + inlet condition. If the fluids flows out from domain the average - or a hydro-static profile - pressure condition is enough. But in a presence of a back flow the code need the volume fractions, temperature, turbulence properties, etc values because in that situation, the boundary behaves such as inlet. The solver does not know how the fluid backflows! That is a big issue if you do not know the BC in that place.

Well, if we have no idea concerning to BC?? The advice is: take experimental measures to feed the needed values, or change your domain till a region where you really know the BCs, extending the domain, including farfields, steps, etc. Otherwise your BCs are a guess.

Thanks Mohsen Isaac Nimvari

Thanks Ricardo Vicente de Paula Rezende .

What would you say about the 'Bulk mass flow rate' available for Outlet boundary condition.

For that, no volume fraction information is asked in CFX-Pre.

Should I conclude that, if I use that option: the liquid+vapor=bulk; it would mean that it is necessarily kept constant (0.3kg/s both for inlet and outlet)?

If so, is it Physically possible as well?

P.S.: my fluids are liquid water and water vapor (wall boiling case).

Interesting question

Yes, Soumitra Vadnerkar , the total mass flow rate is constant (liquid + vapor = cte.). But the volume fractions are unknown. All possibles are solutions 0 vapor and 1 of liquid, 1 of vapor and 0 of liquid or any other combination. The positions ( the profile of volume fractions are unknown as well. Hence, faced to a back flow you need to take some assumptions. To impose a backflow profile for example. Or change the boundaries positions. I've attached a figure to illustrate what can be done to avoid answers things that we do not have to know without change the model response with some postulated assumption.

Thank you Ricardo Vicente de Paula Rezende

So do you think that the bulk mass flow rate can same at outlet as it was at inlet?

I have this confusion because, I thought although, not liquid but vapor being compressible might be reducing the bulk mass flow rate at outlet.

I think, that for a compressible flow (and a multiphase flow with at least one compressible phase - here the vapor - is compressible flow too) it is a mathematically wrongly posed problem, if math flow rate is specified at both inlet and outlet. Even for an incompressible flow it is not a good practice and can easily lead to very bad convergence behavior of the solver, if done so.

As mentioned in other answers above, suitable combinations of inlet and outlet boundary conditions are:

* inlet mass flow rate (equivalent with specified bulk velocity or velocity profile and densities) + averaged static pressure outlet boundary condition

* same for inlet + opening boundary condition at outlet

* mass flow rate at outlet + a pressure type BC at inlet

* pressure boundary conditions at both inlet and outlet

Thank you Dr.Tomas Frank

Soumitra Vadnerkar

even if the vapor is physically compressible, the isochoric flow can be considered because pressure changes may be negligible under certain conditions - I do not know if that would be your case. Generally for Mach

Thanks Ricardo Vicente de Paula Rezende

Dear Soumitra

Your response is very useful for me

Thank you

Ricardo Vicente de Paula Rezende : In case of multiphase flows not only fast movement with higher Ma number is a cause for compressible effects. Imagine a bubble column of 10m height, where gas bubbles are injected with spargers at the bottom. The water column height of 10m is sufficient to cause substantial decompression of the gaseous phase while rising from bottom to top. It is a 10 bar difference.

Thomas Frank, yes, good example! I agree with you, this is another important case with compressible effects. Regarding hydrostatic pressure effect is important to account for the compressible effects in the bubbles, (even though in your example It is 1 bar, I think was a typo error) - but it is still important, to assure more precision.

Ricardo Vicente de Paula Rezende : Yes, you are right. Typo on my end.

The math for the pressure calculation for 10m water column can e.g. be found here:

https://www.quora.com/How-is-1-atmospheric-pressure-10-33-meter-of-water

Ricardo Vicente de Paula Rezende ,referring to the picture above you shared to explain backflow problem,

Don't you think due to change in the geometry of the extended domain at the original outlet will cause messing with the physics of the main domain of interest?

I mean, of course having a diverging (extended) outlet will help reduce the backflow, but doesn't it affect the outlet pressure and velocity (Venturi Effect)?

If so, how can I can I mitigate this unwanted effect?