EB/CNS Test

[jramirezcr]

Hello dear AMRex community.

I want to use AMR Framework in my code, first I am trying to understand the EB/CNS tutorial. So I would like to know if you have any reference about the numerical scheme used in this test.

Thank you very much
Regards
JRC

[asalmgren]

Jorge, Welcome to the AMReX community! Can you tell us a little bit about what type of problem you want to solve with AMReX? Will it be compressible flow? An

…

On Sat, Jul 3, 2021 at 7:18 PM Jorge Ramírez Cruz ***@***.***> wrote: Hello dear AMRex community. I want to use AMR Framework in my code, first I am trying to understand the *EB/CNS* tutorial. So I would like to know if you have any reference about the numerical scheme used in this test. Thank you very much Regards JRC — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#2146>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACRE6YTCMYAYKUVRSVHWJF3TV7AITANCNFSM47YWRHQA> .

-- Ann Almgren Senior Scientist; CCSE Group Lead Pronouns: she/her/hers

[jramirezcr]

Hi An

It is a compressible flow (multiple jets injected into a mixing chamber).

JRC

[asalmgren]

JRC, I would suggest looking at the PeleC code (based on AMReX) for compressible reacting flow with EB geometries -- I think they probably have some problems with multiple jets even set up already ... no point reinventing the wheel :-)

…

On Sat, Jul 3, 2021 at 7:28 PM Jorge Ramírez Cruz ***@***.***> wrote: Hi An It is a compressible flow (multiple jets injected into a mixing chamber). JRC — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#2146 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACRE6YXXMSUHIHY5UNE3BNDTV7BOFANCNFSM47YWRHQA> .

-- Ann Almgren Senior Scientist; CCSE Group Lead Pronouns: she/her/hers

[drummerdoc]

Happy to help. Let know if you run into problems. There’s even a paper that describes the code (although it didn’t make the cut at SC this year, unfortunately).

[jramirezcr]

Dear Ann and Marc.

The thing is that our team has been developing different solvers for years. However, they are curvilinear and use immersed boundaries to represent the geometries. Now, we are trying to port them using AMR techniques. Currently, we have three options:

-Develop the entire code (not the best idea).
-Use PARAMESH
-Use the AMRex framework.

We decided to use your framework AMRex. The first solver we wanted to port is a fully explicit compressible code. That's why I ask for references on the EB/CNS test to use it as a guide (just the numerical scheme implemented). The documentation of the framework is pretty good but we are trying to understand in depth this case.

Thank you so much
Saludos Cordiales
JRC

[asalmgren]

Understood! Tests/EB/CNS is a compressible solver using method-of-lines with Riemann solver (not sure which one). Unfortunately it's not the best example of how to code in AMReX since it still uses f90 code which we have tried to get away from. (We've been meaning to migrate it to our C++ style but just haven't had time yet.) It does have AMR with subcyling in time, EB and flux redistribution (and re-redistribution at coarse-fine boundaries) Hope that helps! Ann p.s. If you'd like to be able to ask questions as you go along, we'd be happy to invite you to our amrex slack channel...

…

On Sun, Jul 4, 2021 at 4:51 PM Jorge Ramírez Cruz ***@***.***> wrote: Dear Ann and Marc. The thing is that our team has been developing different solvers for years. However, they are curvilinear and use immersed boundaries to represent the geometries. Now, we are trying to port them using AMR techniques. Currently, we have three options: -Develop the entire code (not the best idea). -Use PARAMESH -Use the AMRex framework. We decided to use your framework AMRex. The first solver we wanted to port is a fully explicit compressible code. That's why I ask for references on the EB/CNS test to use it as a guide (just the numerical scheme implemented). The documentation of the framework is pretty good but we are trying to understand in depth this case. Thank you so much Saludos Cordiales JRC — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#2146 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACRE6YWTBUGUINA7W3HXNO3TWDXWLANCNFSM47YWRHQA> .

-- Ann Almgren Senior Scientist; CCSE Group Lead Pronouns: she/her/hers

[jramirezcr]

It helps to me, thanks. And it would be nice to be in the AMRerx slack channel :) .

[asalmgren]

Jorge - can you email me your preferred email for the slack invite? thanks Ann

…

On Mon, Jul 5, 2021 at 8:39 PM Jorge Ramírez Cruz ***@***.***> wrote: It helps to me, thanks. And it would be nice to be in the AMRerx slack channel :) . — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#2146 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACRE6YS65KSGZ4TKMKC25G3TWJ3G3ANCNFSM47YWRHQA> .

-- Ann Almgren Senior Scientist; CCSE Group Lead Pronouns: she/her/hers

[jramirezcr]

Sure, jramirezcr@iingen.unam.mx

Thanks.
JRC