[pedramtx]

Hi All,
I am trying to implement a numerical tripping by applying a Gaussian force in the wall-normal direction in OpenFOAM. The reason for this implementation is that I'd like to force the transition to turbulence in flat plate boundary layer flow. This idea can be done by adding a term to the momentum equation of the solver. The volume force has been defined in Schlatter and Orlu (2012) "Schlatter, P., & Örlü, R. (2012). Turbulent boundary layers at moderate Reynolds numbers: inflow length and tripping effects. Journal of Fluid Mechanics, 710, 5-34."
I am seeking help regarding implementation of this force defined in spectral domain (the ineverse Fourier) into the momentum equation in physical domain. Any help is greatly appreciated.

Thanks,
Pedram

[mAlletto]

Just from the picture you showed it seems that the force is a function of x,y,z,t. The random numbers h are function of z and t. Can you provide some more detail?

[pedramtx]

Originally Posted by mAlletto

Just from the picture you showed it seems that the force is a function of x,y,z,t. The random numbers h are function of z and t. Can you provide some more detail?

Hi Michael,

You are right. The random force g is a function of z and t. I want to apply a 1D wavenumber in the spanwise direction. Then the whole term F is expressed by a local Gaussian distribution in z plane.

[mAlletto]

Maybe I did not und understand correctly the question.

[mAlletto]

I found by accident this post of Schlatter where he explains how he generated the random harmonic signal: https://lists.mcs.anl.gov/pipermail/nek5000-users/2018-February/005087.html

in order to generate the random source term in order to simulate the boundary layer tipping.

[pedramtx]

Originally Posted by mAlletto

I found by accident this post of Schlatter where he explains how he generated the random harmonic signal: https://lists.mcs.anl.gov/pipermail/nek5000-users/2018-February/005087.html

in order to generate the random source term in order to simulate the boundary layer tipping.

Thank you, Michael. This is so helpful.

[mAlletto]

You're welcome

Let me know how it goes. I find the topic interesting

[agustinvo]

Very interesting topic! I will follow it. It is also interesting for me. For the moment, what I do is to use swak4Foam and introduce some disturbances in the wall.

[pedramtx]

Originally Posted by agustinvo

Very interesting topic! I will follow it. It is also interesting for me. For the moment, what I do is to use swak4Foam and introduce some disturbances in the wall.

Agustín,
Thank you for your suggestion. Still trying to make it work.

[syavash]

Originally Posted by agustinvo

Very interesting topic! I will follow it. It is also interesting for me. For the moment, what I do is to use swak4Foam and introduce some disturbances in the wall.

Why not use a more sophisticated method like LeMOS or SEM at the inlet? Or a step further ahead, use the method introduced by Lund et al. (JCP 1998) in your simulation.


Regards,
Syavash

[agustinvo]

Originally Posted by syavash

Why not use a more sophisticated method like LeMOS or SEM at the inlet? Or a step further ahead, use the method introduced by Lund et al. (JCP 1998) in your simulation.


Regards,
Syavash

Hi Syavash,
my case is a natural convection boundary layer, not a forced convection one. The boundary layer starts developing when the wall is heated. If I introduce an inlet flow, I'd have a mixed convection case.


I will check later the different approaches you said, I have some colleagues who may be interested.


Agustín

[syavash]

Originally Posted by agustinvo

Hi Syavash,
my case is a natural convection boundary layer, not a forced convection one. The boundary layer starts developing when the wall is heated. If I introduce an inlet flow, I'd have a mixed convection case.


I will check later the different approaches you said, I have some colleagues who may be interested.


Agustín

Agustin,


Having a naturally developing boundary-layer to turbulence should be very interesting, and challenging! Actually I got very interested in this problem. Could you please provide me a snapshot of your simulation? Also would you introduce me a paper dealing with the same problem using LES?


Regards,
Syavash

[pedramtx]

Originally Posted by syavash

Why not use a more sophisticated method like LeMOS or SEM at the inlet? Or a step further ahead, use the method introduced by Lund et al. (JCP 1998) in your simulation.


Regards,
Syavash

Syavash,

Thank you for your suggestion. We already captured the turbulent BL using Lund method, but the current simulation aims at resolving the transition behavior of BL. As you may know, rescaling procedure in Lund method necessitates equilibrium state (developed turbulent field) at recycling station.


Thanks,
Pedram

[sjlouie91]

Originally Posted by pedramtx

Agustín,
Thank you for your suggestion. Still trying to make it work.

Hi Pedram,

I also find this numerical tripping quite interesting. Although it's been a long time since you last posted, I am still very curious about how this method is implemented in OpenFOAM. Have you already solved this problem? I would appreciate it if you could share your code or give me some clues on how to implement it. Thanks.

kind regards,
Jin