March 2018 - Nek5000-users - mailman3.cels.anl.gov

Small random perturbations
by nek5000-users＠lists.mcs.anl.gov 09 Mar '18

09 Mar '18

Dear Dr.Fischer, Actually, I am doing 2-D simulation, at least for now, so no z-dimension is up there. Am I wrong to do DNS in 2-D? Also, I just started doing so, and I think my files are not useful at this point. Let me try more, maybe I am missing something. Thank you for your awesome free code and your continuous support. Ali https://lists.mcs.anl.gov/pipermail/nek5000-users/2018-March/005133.html

1 1

Small random perturbations
by nek5000-users＠lists.mcs.anl.gov 08 Mar '18

08 Mar '18

Yes, the length is 5D and 40 elements are in the x-y plane only. sincerely, Ali

1 1

Small random perturbations
by nek5000-users＠lists.mcs.anl.gov 08 Mar '18

08 Mar '18

Dear Dr.Fischer, Thank you for your time. I simply change lx1 in the SIZE from 7 to 9 in order to increase the resolution and I am not moving to a new mesh. Also, I just started with 40 elements with lx1 =7, clustered to the wall, and my Re is around 6000. Best Regards, Ali https://lists.mcs.anl.gov/pipermail/nek5000-users/2018-March/005129.html

1 1

Small random perturbations
by nek5000-users＠lists.mcs.anl.gov 08 Mar '18

08 Mar '18

Dear all First, many thanks for your response. I couldn't be more thankful. Either by starting with high Reynolds number or by using initial noise is working now. However, once I increase the resolution the turbulence decays and disappears completely, and I don't know why? I just wanted to make grid independence and my geometry is very small, 10 cm in axial direction & 1 cm along the radius. Any idea is greatly appreciable. Ali https://lists.mcs.anl.gov/pipermail/nek5000-users/2018-March/005126.html

1 1

Non-uniform mesh
by nek5000-users＠lists.mcs.anl.gov 07 Mar '18

07 Mar '18

Dear Neks, I'm simulating the flow in a 90 degree bent pipe. I am trying to build a non-uniform mesh which is finer in the bent section and coarser in the straight sections. For this purpose I am writing this code in the subroutine usrdat2: do i = 1, nelv iel=lglel(i) if (iel.le.960) then z = 0.8*z elseif ((iel.gt.960).and.(iel.le.1728)) then z = 3.*z elseif (iel.gt.1728) then z = 0.5*z endif enddo The idea is to dilate the coordinates in the straight sections and to shrink them in the bent section so that the mesh would be non-uniform. However, this doesn't work. Is this the correct procedure? Thanks. Best regards.

1 0

Small random perturbations
by nek5000-users＠lists.mcs.anl.gov 07 Mar '18

07 Mar '18

Thank you for responding. It is axisymmetric with periodic conditions at the ends. what geometry are you interested in? Philipp On 2018-02-20 15:20, nek5000-users at lists.mcs.anl.gov<https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users> wrote: > > Hello everyone, > > I am just wondering about the validity of my way to initiate > turbulence. My question is that instead of introducing random > disturbances to the initial velocity, can I introduce them to the > pressure gradient? Hint, I am using Nek5000 to simulate statistically > non-stationary turbulent flow. > > Thank you in advance > Ali > Embry-Riddle Aeronautical University

1 2

Outflow boundary conditions
by nek5000-users＠lists.mcs.anl.gov 07 Mar '18

07 Mar '18

Dear all, are outflow advective boundary conditions available in nek5000? Or is there any work around in order to impose outflow boundary conditions (for the velocity and the passive scalars) that can grantee that there is no flow coming back from the outflow boundary? Thanks, Dante

1 2

Complex number as mesh velocity
by nek5000-users＠lists.mcs.anl.gov 06 Mar '18

06 Mar '18

Hi Neks, I was doing pitching and plunging simulation of airfoil. I could successfully rotate the foil with smoothing function. Thank to you all who helped me. Now as a part of validation, I have to impose a forced motion. For that, I found a good paper where a complex exponential function( e^ iwt ) as the velocity of the foil. My question is, is it possible to give a complex number to give as an input to the mesh velocity ? Sorry if its a foolish question. Thanks Sijo

1 1

Generating Turbulence
by nek5000-users＠lists.mcs.anl.gov 06 Mar '18

06 Mar '18

Dear Nek5000-users, Could you please help me to solve my problem? My geometry is axisymmetric with the following dimensions (R=0.01 m, L=10 R). Well, I did DNS and got something indicating turbulence which is consistent with experiments. However, once I increase either the spatial and/or temporal resolution even a little bit, the flow is laminarized completely. Any help would be greatly appreciated Ali

1 0

gmsh and nek5000
by nek5000-users＠lists.mcs.anl.gov 04 Mar '18

04 Mar '18

Dear all, For a different project (with OpenFOAM), we have started to use gmsh for meshing, and it seems to be quite powerful for our purposes, and also provide high-order meshing capabilities, which come in handy for Nek5000. I have seen the tool by Praveen Chandrashekar on bitbucket, but it appears to only handle 2D geometries so far: https://bitbucket.org/cpraveen/nek5000 Is there any other converted being developed perhaps which could handle 3D? Thanks, Philipp

1 0