[hwsv07]

If I want to simulate a very low turbulence problem, how should I know if I should select laminar or inviscid model in the fluid solver?

Thanks!

[FMDenaro]

Originally Posted by hwsv07

If I want to simulate a very low turbulence problem, how should I know if I should select laminar or inviscid model in the fluid solver?

Thanks!

Your question seems strange ... if you want to set a laminar model that's only for simulating turbulence in a DNS formulation. On the other hand, if you set inviscid flow you should be aware that you cannot simulate confined turbulence and you cannot perform a DNS.

So, explain in details what you want to simulate

[hwsv07]

Originally Posted by FMDenaro

Your question seems strange ... if you want to set a laminar model that's only for simulating turbulence in a DNS formulation. On the other hand, if you set inviscid flow you should be aware that you cannot simulate confined turbulence and you cannot perform a DNS.

So, explain in details what you want to simulate

maybe i am not framing the question well.

I want to compare RANS results with an analytical model (which does not include any turbulence dissipation).

so if i want to have CFD results that can be compared with my analytical model (which does not account for turbulence), how should I set my CFD model? invisicd or laminar?

thanks!

[FMDenaro]

Still not clear ...RANS provides a statistically averaged solution, what about the type of your analytical solution? I see you need for sure a viscous model

[hwsv07]

Originally Posted by FMDenaro

Still not clear ...RANS provides a statistically averaged solution, what about the type of your analytical solution? I see you need for sure a viscous model

yes my purpose is to show the difference between a RANS solution and a inviscid analytical model.

but i also want to check that when I select a invicid/laminar model in the fluid solver, the solution should equal or almost the same as the invisicid analytical model.

making sense here?

so my question is: whether i should choose a inviscid or laminar model in the fluid solver to compare to my inviscid analytical model.

[FMDenaro]

honestly, I don't see the sense in making comparison between RANS and an inviscid model.... But without more details I can't say much more...
I suppose you should compare an Euler simulation and your inviscid analytical solution

[hwsv07]

Originally Posted by FMDenaro

honestly, I don't see the sense in making comparison between RANS and an inviscid model.... But without more details I can't say much more...
I suppose you should compare an Euler simulation and your inviscid analytical solution

sorry if i was not clear.

forget about the RANS part.

I am trying to comparing between inviscid model to that from a fluid solver. so the question is what flow physics should I specify for the fluid solver: invisicid or laminar?

[FMDenaro]

The Euler inviscid model

[Antanas]

Originally Posted by hwsv07

sorry if i was not clear.

forget about the RANS part.

I am trying to comparing between inviscid model to that from a fluid solver. so the question is what flow physics should I specify for the fluid solver: invisicid or laminar?

In my understanding when you choose laminar flow that means that solver solve non-averaged N-S equations. Indeed this equations can resolve turbulent structures of different length scales depending on mesh resolution, that is called Direct Numerical Simulation and it's very costly for most practical (turbulent) flows. So it better be used for really laminar flows, i.e. flows where viscous forces are large compared with inertial forces, so characteristic Re number is low. Inviscid flows in contrast are those where viscous forces are neglible and Re is large. In this case we obtain Euler equations from N-S equations. If your CFD code doesn't have ablitiy to solve such equations you can try to set low viscosity in material props and use free sleep conditions on walls. And this won't be laminar flow.

Correct me if I'm wrong.

[hwsv07]

Originally Posted by FMDenaro

The Euler inviscid model

Originally Posted by Antanas

In my understanding when you choose laminar flow that means that solver solve non-averaged N-S equations. Indeed this equations can resolve turbulent structures of different length scales depending on mesh resolution, that is called Direct Numerical Simulation and it's very costly for most practical (turbulent) flows. So it better be used for really laminar flows, i.e. flows where viscous forces are large compared with inertial forces, so characteristic Re number is low. Inviscid flows in contrast are those where viscous forces are neglible and Re is large. In this case we obtain Euler equations from N-S equations. If your CFD code doesn't have ablitiy to solve such equations you can try to set low viscosity in material props and use free sleep conditions on walls. And this won't be laminar flow.

Correct me if I'm wrong.

i think both of your replies make sense to me.

for my case, since I am trying to validate results between a inviscid analytical model to that from a fluid code-solver, i should be using prescribing a fluid with very low viscosity or in this case, the inviscid NS equations.

thanks.