- Nektar-users - imperial.ac.uk

Use of equispacedoutput in Quasi-3D Simulations
by İlteber Özdemir 14 Jul '23

14 Jul '23

******************* This email originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list https://spam.ic.ac.uk/SpamConsole/Senders.aspx to disable email stamping for this address. ******************* Hi, Hope everybody is having a nice summer vacation. But the simulations must go on! I am running a quasi-3D simulation. In my test case I have a tetra 2D mesh and 8 Fourier planes. So the elements are triangular prisms. As discussed in many earlier posts, use of equispacedoutput normally decreases the output file size. However, when I use equispacedoutput it not only interpolates my 2D mesh, but also creates triangles in z-direction. Thus, the final output is not comprised of triangular prisms but tetrahedrons. Although decreasing the file size of the 2D mesh, this increases the total file size, because I basically have double the amount of elements along z-direction compared to prior. With the hope to solve this problem, I tried tetonly flag. I am unsure if this is the correct option to use. I get the attached error message, when I run FieldConvert with this option: FieldConvert -f -v -m equispacedoutput:tetonly timeSteps/out.xml timeSteps/out_3.chk/ timeSteps/fcFiles/out_3.plt Here out.xml is my mesh, out_3.chk is the time step I want to convert, and out_3.plt is the name of the output. I know that FieldConvert fails specifically when I use tetonly, as it works when I don't use it. I would very much appreciate any help to solve this problem or a suggestion for a workaround. Thank you in advance for your help. With kind regards, Ilteber -- İlteber R. Özdemir

1 0

Integral filter
by Guillermo Suárez 26 Jun '23

26 Jun '23

******************* This email originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list https://spam.ic.ac.uk/SpamConsole/Senders.aspx to disable email stamping for this address. ******************* Hello, I would like to use the integral filter in nektar++ but I am experiencing some difficulties. As a test case for testing the filter I am using a slightly modified version of the forward step case from the tutorials - I have modified the inlet to 2 length units high instead of just 1. Besides, at the inlet I have set the horizontal velocity component to 1, I no longer use a parabolic profile. Then, I have defined the Integral filter to monitor the composites 2 and 3, which correspond to the "wall" and "inlet". Regarding the results: At the "wall" composite, non-slip boundary conditions are applied, meaning that velocity is set to zero. However, the integral filter outputs values quite different from zero. For the "inlet" composite, The output of the integral filter for the vertical velocity provides a value quite close to zero, O(-7), but not exactly zero. I understand that there might be numerical errors in the calculations but I do not know if this accuracy (or error) is expected. Regarding the integral results for the horizontal velocity, I would expect the output to be a constant value of 2 - the horizontal velocity is 1 and the height of the inlet is 2. However, it outputs a value of 3.1568. Could you please indicate to me whether I am misinterpreting the functionality of the integral filter? Have I maybe incorrectly set certain parameters? I enclose the session and mesh files. Thank you in advance. Cheers, Guillermo

1 0

Time-average of a gradient
by Ehsan Asgari 19 Jun '23

19 Jun '23

******************* This email originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list https://spam.ic.ac.uk/SpamConsole/Senders.aspx to disable email stamping for this address. ******************* Dear All, How may I calculate the time-average of a gradient, for example velocity gradient? Kind regards syavash

2 3

Some questions about the modifiedArnoldi solver
by 王舟阳 18 Jun '23

18 Jun '23

******************* This email originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list https://spam.ic.ac.uk/SpamConsole/Senders.aspx to disable email stamping for this address. ******************* Dear nektar++ users and developers. Sorry to bother everyone here, however, I need some help the modifiedArnoldi solver. In the linear stability solver, there are NUMSTEPS and KDIM to be filled in, what is the relationship between the setting of these two variables and the frequency of the unstable mode to be solved? For example, if the frequency of the unstable mode is approximately 0.1 Hz (0.2pi for the circle frequency) and TIMESTEP = 0.002, should NUMSTEPS and KDIM be set relatively large or small in order to capture this unstable mode? I wonder if anyone has any relevant experience in setting this up and I would appreciate any help. Thanks again to the users and developers of nektar++. Best Regards, Wang Zhouyang

2 1

result problems in Aerodynamics forces
by 王同快 14 Jun '23

14 Jun '23

******************* This email originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list https://spam.ic.ac.uk/SpamConsole/Senders.aspx to disable email stamping for this address. ******************* Dear All, I've calculated the drag forces on a cylinder and a sphere and compared the results with the standard drag model and find that the force in .fce file is approximately 8 times larger than the model prediction.And it is also the case for lift force and momentum.I wonder if there is something wrong with the force integral code. Kind regards Wang

2 1

Calculating time-averaged wall shear stress
by Ehsan Asgari 06 Jun '23

06 Jun '23

******************* This email originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list https://spam.ic.ac.uk/SpamConsole/Senders.aspx to disable email stamping for this address. ******************* Dear All, I have a time-averaged field, say "time-average.fld". How should I proceed to calculate time-averaged wall shear stress? I was wondering if I need to use "wss" first. Kind regards syavash

2 2

Re: [Nektar-users] use 2-d initial condition for quasi-3d simulation
by Ehsan Asgari 29 May '23

29 May '23

Hi Stanislaw I tried your suggestion but I get some error. Do you have any idea how I can fix this? Initial Conditions: > - Field u: from file inlflow2D_restart.fld > - Field v: from file inlflow2D_restart.fld > - Field w: 0.0 > - Field p: 0.0 > CFL (zero plane): 1.22293e-06 (in elmt 6555) > Fatal : Level 0 assertion violation > m_timestep < 0 > Fatal : Level 0 assertion violation > m_timestep < 0 > Fatal : Level 0 assertion violation > m_timestep < 0 > Fatal : Level 0 assertion violation > m_timestep < 0 > Fatal : Level 0 assertion violation Here is my initial condition: <FUNCTION NAME="InitialConditions"> > <F VAR="u,v" FILE="inflow2D_restart.fld" /> > <E VAR="w,p" VALUE="0.0" /> > </FUNCTION> Kind regards syavash On Fri, May 26, 2023 at 11:07 AM Stanisław Gepner < stanislaw.gepner(a)pw.edu.pl> wrote: > Hi, > > It is, all you need is an XML file for the aquasi-3D simulation with > velocity components u,v,w and pressure p, just as you would in quasi 3D, > than in initial conditions you do: > > <FUNCTION NAME="InitialConditions"> > <F VAR="u,v" FILE="geom2D.fld" /> > <E VAR="w" FILE="0" /> > <E VAR="p" VALUE="0" /> > </FUNCTION> > > You could also initialize w from file, eg. using solution from the ADR > solver or any other fld file. > > Cheers! > Stan > > > *Faculty of Power and Aeronautical Engineering* > Nowowiejska 21/25 Str. > 00-665 Warsaw, Poland > phone +48 (22) 234 51 70 > > <https://www.pw.edu.pl/engpw> > On 26.05.2023 08:30, Ehsan Asgari wrote: > > This email from eh.asgari(a)gmail.com originates from outside Imperial. Do > not click on links and attachments unless you recognise the sender. If you > trust the sender, add them to your safe senders list > <https://spam.ic.ac.uk/SpamConsole/Senders.aspx> to disable email > stamping for this address. > > > Dear All, > > Is it possible to run a 2-d simulation and then use it solution for the > quasi-3d simulation of the same mesh? If so, how can I achieve this? > > Kind regards > syavash > > _______________________________________________ > Nektar-users mailing listNektar-users@imperial.ac.ukhttps://mailman.ic.ac.uk/mailman/listinfo/nektar-users > >

2 5

Implementing rough wall condition
by Selman BAYSAL 29 May '23

29 May '23

******************* This email originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list https://spam.ic.ac.uk/SpamConsole/Senders.aspx to disable email stamping for this address. ******************* Dear all,I am relatively new to CFD and Nektar++. Although there are much studies perfomed by Nektar++ on flows over rough surface/bed, I could not find detailed information on&nbsp;how to define a rough surface.I intended to define a rough surface/boundary (not a single roughness element or hump), and then, to investigate the boundary layer developed over this surface. For this purpose, do I need an external program like GMSH or is there any parameters (such as Nikuradse equivalent sand roughness or roughness height) or&nbsp;modules implemented in Nektar++? Any help would be greatly appreciated!Many thanks,Selman BaysalR. A. in HydraulicsDept. of Civil Eng'gÇanakkale Onsekiz Mart University--ITU Civil Eng., BSc.ITU Coastal Sci. &amp; Eng., MSc.

2 1

Re: [Nektar-users] use 2-d initial condition for quasi-3d simulation
by Ehsan Asgari 26 May '23

26 May '23

Thank you. It is what I was looking for. syavash On Fri, May 26, 2023 at 11:07 AM Stanisław Gepner < stanislaw.gepner(a)pw.edu.pl> wrote: > Hi, > > It is, all you need is an XML file for the aquasi-3D simulation with > velocity components u,v,w and pressure p, just as you would in quasi 3D, > than in initial conditions you do: > > <FUNCTION NAME="InitialConditions"> > <F VAR="u,v" FILE="geom2D.fld" /> > <E VAR="w" FILE="0" /> > <E VAR="p" VALUE="0" /> > </FUNCTION> > > You could also initialize w from file, eg. using solution from the ADR > solver or any other fld file. > > Cheers! > Stan > > > *Faculty of Power and Aeronautical Engineering* > Nowowiejska 21/25 Str. > 00-665 Warsaw, Poland > phone +48 (22) 234 51 70 > > <https://www.pw.edu.pl/engpw> > On 26.05.2023 08:30, Ehsan Asgari wrote: > > This email from eh.asgari(a)gmail.com originates from outside Imperial. Do > not click on links and attachments unless you recognise the sender. If you > trust the sender, add them to your safe senders list > <https://spam.ic.ac.uk/SpamConsole/Senders.aspx> to disable email > stamping for this address. > > > Dear All, > > Is it possible to run a 2-d simulation and then use it solution for the > quasi-3d simulation of the same mesh? If so, how can I achieve this? > > Kind regards > syavash > > _______________________________________________ > Nektar-users mailing listNektar-users@imperial.ac.ukhttps://mailman.ic.ac.uk/mailman/listinfo/nektar-users > >

1 0

use 2-d initial condition for quasi-3d simulation
by Ehsan Asgari 26 May '23

26 May '23

******************* This email originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list https://spam.ic.ac.uk/SpamConsole/Senders.aspx to disable email stamping for this address. ******************* Dear All, Is it possible to run a 2-d simulation and then use it solution for the quasi-3d simulation of the same mesh? If so, how can I achieve this? Kind regards syavash

1 0