- Nektar-users - imperial.ac.uk

Cannot find corresponding element face
by A.P. 19 Mar '18

19 Mar '18

Hello, I ran NekMesh utility and recived the error message: "Cannot find corresponding element face for 2D element ..." What can I do in this situation? Alexander.

2 1

initial condition for Turb_Channel Flow
by Mahdi Kooshkbaghi 13 Mar '18

13 Mar '18

Dear all, Is there any .rst file available for the initial condition of turbulent channel flow? I want to do DNS with Re_tua = 180 or 395 or 590. Those of benchmarks in MKM paper (https://doi.org/10.1063/1.869966) Best, Mahdi

1 0

Stability analysis of compressible flow
by devabratasahoo＠campus.technion.ac.il 11 Mar '18

11 Mar '18

Dear all, I would like to know if it is possible to use Nektar++ to analyze the stability of a compressible flow. To best of my knowledge, the functionality was unavailable as Nektar doesn't have an implementation of the linearised compressible NS equations. But, I would like to know if there is any development regarding this. Regards, Devabrata Sahoo

1 0

Problem in Robin BC in 3D1H case
by Asım Önder 08 Mar '18

08 Mar '18

Dear all, in a 3D turbulent channel flow simulation with Fourier expansions in one direction, I want to add some small wall imperfections using Robin boundary conditions. To this end, I specified the following simple cos disturbances into wall boundary: <R VAR="u" VALUE="0" PRIMCOEFF="(pertAmp*cos(2.0*PI/LZ*2*z+2.0*PI/LX*4*x))^(-1)"/> <R VAR="v" VALUE="0" PRIMCOEFF="(pertAmp*cos(2.0*PI/LZ*2*z+2.0*PI/LX*4*x))^(-1)"/> <R VAR="w" VALUE="0" PRIMCOEFF="(pertAmp*cos(2.0*PI/LZ*2*z+2.0*PI/LX*4*x))^(-1)"/> When I checked the data, it seems like the terms multiplied by z remains alway zero while x-related terms have no problem. In this regard, I was wondering whether 3D1H case does support functional expansions in the Fourier direction for Robin BC. Any feedback would be appreciated. Thanks, Asim

1 0

CCM module error
by Thomas Andreou 23 Feb '18

23 Feb '18

Dear All, I was trying to build the ccm module and got this message: CMake Warning at cmake/ThirdPartyCCM.cmake:29 (MESSAGE): We are seeking permission to distribute ccmio with Nektar++. If you are entitled to use libccmio please contact nektar-users(a)imperial.ac.uk and place the file ccmio-2.06.tar.bz2 in the director $NEKTAR/ThirdParty Call Stack (most recent call first): CMakeLists.txt:250 (INCLUDE) Could you please assist? Many Thanks, Thomas Kind Regards, Thomas Andreou t.andreou.1(a)research.gla.ac.uk PhD Student Aerospace Sciences Division University of Glasgow

1 0

Nektar++ Installation question
by Thomas Andreou 19 Feb '18

19 Feb '18

Dear All, I need some help please with installing Nektar++ on Ubuntu 16.04 . I've been trying to install it using the ccmake method as i was interested in using the NEKMESHGEN module. I have however been encountering this errror after "make install" : CMakeFiles/Makefile2:1023: recipe for target 'library/NekMeshUtils/CMakeFiles/NekMeshUtils.dir/all' failed make[1]: *** [library/NekMeshUtils/CMakeFiles/NekMeshUtils.dir/all] Error 2 make[1]: *** Waiting for unfinished jobs.... [ 38%] Linking CXX executable LocProject1D [ 38%] Built target LocProject1D [ 38%] Linking CXX executable LocProject_Diff2D [ 38%] Built target LocProject_Diff2D [ 38%] Linking CXX shared library libCollections.so [ 39%] Linking CXX executable LocProject3D [ 39%] Linking CXX executable LocalRegionsUnitTests [ 39%] Built target Collections [ 39%] Built target LocProject3D [ 39%] Built target LocalRegionsUnitTests Makefile:162: recipe for target 'all' failed make: *** [all] Error 2 I have also tried ''make VERBOSE=1'' and get this : [cid:5314fbf7-1cf2-4283-b059-7be9760ffb23] When i tried compiling with NEKMESHGEN off it seemed to work. Does anyone know how i can get around this issue? Also is there a way to install the mesh generator without cmake? could i do it via sudo apt-get install? Kind Regards, Thomas Andreou t.andreou.1(a)research.gla.ac.uk PhD Student Aerospace Sciences Division University of Glasgow

3 5

Re: [Nektar-users] [Follow-up] Some problems when using SFD method in nektar++ 4.4.1 ( non-convergence)
by Sherwin, Spencer J 18 Feb '18

18 Feb '18

Dear Yiran, Sorry I have been ratter behind on my emails and so am only finally getting to the bottom of them. I am forwarding your message to the user list to see if anyone else has experienced similar issues. One point I would make however is that the SFD methods is good for unsteady flows (like vortex shedding) but cannot find the steady state of a problem that has a real leading eigenvalue such as a hopf bifurcation. The reason is that the flow needs to contain some instance of the steady state. Your problem of a separation bubble might be more similar to the second type of problem and this could be the cause behind the issue. Have you had a chance to look at the paper by Bastien Jordi? Jordi BE, Cotter CJ, Sherwin SJ, 2014, Encapsulated formulation of the selective frequency damping method<http://dx.doi.org/10.1063/1.4867482>, Physics of Fluids, Vol: 26, ISSN: 1089-7666 Where we try to motivate this issue. Cheers, Spencer On 1 Feb 2018, at 03:47, 李依然 <by1504151(a)buaa.edu.cn<mailto:by1504151@buaa.edu.cn>> wrote: Dear Professor Sherwin, My name is Yiran Li, and I am currently a PhD student at Beihang University in the School of Energy and Power Engineering. I am sincerely sorry for bothering you again but I want to follow up with you to be sure you received my email that I sent on 11th January, 2018 with the title being "Some problems when using SFD method in nektar++ 4.4.1 ( non-convergence)". Recently, I have been trying to use the Selective Frequency Damping (SFD) method implemented in Nektar++4.4.1 to obtain the base flow of an experiment-specified separation bubble, but after weeks of numerical tests, my calculations still cannot converge. I do not know if there is something wrong with my case settings or the case I try to figure out here might fundamentally beyond the scope of the SFD method. For details, a review of my works and results over the past few weeks together with some personal thinking and confusions was attached at that email as the PDF file called “Review on results and some confusions using SFD method” for your reference. Besides, the related session files and field files for some typical cases mentioned in that review were also attached there in the package called “cases”. Now, in case the attachments at that email have been out of date, I have attached them again in this email for your reference. Thank you very much for your patience and look forward to your kindly reply! Best Regards, Yiran Li -----原始邮件----- 发件人:"李依然" <by1504151(a)buaa.edu.cn<mailto:by1504151@buaa.edu.cn>> 发送时间:2018-01-11 12:24:08 (星期四) 收件人: s.sherwin(a)imperial.ac.uk<mailto:s.sherwin@imperial.ac.uk> 抄送: b.jordi11(a)imperial.ac.uk<mailto:b.jordi11@imperial.ac.uk> 主题: Some problems when using SFD method in nektar++ 4.4.1 ( non-convergence) Dear Professor Sherwin and Dr. Jordi, My name is Yiran Li, and I am currently a PhD student at Beihang University in the School of Energy and Power Engineering. Recently, I have been trying to use the Selective Frequency Damping (SFD) method implemented in Nektar++4.4.1 to obtain the base flow of an experiment-specified separation bubble which undergoes a laminar separation while a turbulent reattachment under adverse pressure gradient on a flat plate. However, after weeks of numerical experiments, the calculations still cannot converge. I was wondering if there is something wrong with my comprehension and settings or the case here might fundamentally beyond the scope of the SFD method. For details, a review of my works and results over the past few weeks together with some personal thinking and confusions is attached as the PDF file called “Review on results and some confusions using SFD method” for your reference. Besides, the related session files and field files for some typical cases mentioned in that review are also attached in the package called “cases”. Thank you very much for your patience and any response will be appreciated! Best Regards, Yiran Li ________________________________ 从buaa.edu.cn<http://buaa.edu.cn>发来的超大附件 cases.zip<https://mail.buaa.edu.cn/coremail/viewDownloadFile.jsp?key=1U31SsvkjqIgh48C…> (66MB, 2018年01月18日 12:04 到期) 查看下载信息<https://mail.buaa.edu.cn/coremail/viewDownloadFile.jsp?key=1U31SsvkjqIgh48C…> ________________________________ 从buaa.edu.cn<http://buaa.edu.cn>发来的超大附件 cases.zip<https://mail.buaa.edu.cn/coremail/viewDownloadFile.jsp?key=1U31SsvkjqIgh48C…> (66MB, 2018年02月08日 10:48 到期) 查看下载信息<https://mail.buaa.edu.cn/coremail/viewDownloadFile.jsp?key=1U31SsvkjqIgh48C…> <Review on results and some confusions using SFD method .pdf> Spencer Sherwin FREng, FRAeS Head, Aerodynamics, Professor of Computational Fluid Mechanics, Department of Aeronautics, Imperial College London South Kensington Campus, London, SW7 2AZ, UK s.sherwin(a)imperial.ac.uk<mailto:s.sherwin@imperial.ac.uk> +44 (0)20 7594 5052 http://www.imperial.ac.uk/people/s.sherwin/

1 0

IncNavierStokesSolver - Geometry section not read from config file
by Oliviu Sugar 08 Feb '18

08 Feb '18

Dear all, The problem I'm facing is probably trivial, but I just cannot figure it out.The simulation is simply a test case for an under-resolved DNS of a flow at Re=3000 inside a circular pipe, with 1D homogeneity along the flow direction, a body force to drive the flow and some white noise to stimulate transition to turbulence. The problem is that the IncNS solver throws the following error: Cannot find element 'GEOMETRY'.The config file is attached here.Any advice would be much appreciated. Cheers,Oliviu

2 4

Re: [Nektar-users] Regarding expansion types
by Abhishek Kumar 07 Feb '18

07 Feb '18

Dear David, Thanks a lot for making the changes. I will do regress test of RBC stability, and update you soon. With regards Abhishek On Wed, Feb 7, 2018 at 6:07 PM, David Moxey <D.Moxey(a)exeter.ac.uk> wrote: > Hi Abhishek, > > Okay, that issue was pretty straightforward to solve -- advection was not > being calculated for the scalar field. I have pushed that fix up to the > repository, and I now get a positive eigenvalue of 2.1124 with your XML > file, so pretty close to your reference value. > > Thanks, > > Dave > > > On 7 Feb 2018, at 15:24, Abhishek Kumar <abhishek.kir(a)gmail.com> wrote: > > > > Dear David, > > > > Thanks for your response. > > > > I have attached the XML file for RBC stability. I am running for Ra=2000 > with k=3.117. The critical value of Rayleigh number for this configuration > is 1708. > > > > The growth rate for Ra=2000 of the most unstable mode should be 2.1464. > However, I am getting -9.8699, a negative growth rate, which is not > possible as Ra > Ra critical. > > > > With regards > > Abhishek > > > > > > > > > > > > On Wed, Feb 7, 2018 at 2:50 PM, David Moxey <D.Moxey(a)exeter.ac.uk> > wrote: > > Hi Abhishek, > > > > Thanks for the update, glad to hear we are producing correct results on > that front. > > > > Regarding the linear stability; can you share your XML file? It might > just be that the base flow needs to be properly imposed in the appropriate > scalar -- we probably haven't taken scalar fields into account yet in the > setup. > > > > Thanks, > > > > Dave > > > > > On 6 Feb 2018, at 13:20, Abhishek Kumar <abhishek.kir(a)gmail.com> > wrote: > > > > > > Dear David, > > > > > > Thanks for your answer. Now, I got it. > > > > > > You would be happy to know that the Rayleigh-Benard convection (RBC) > solver is working properly with Nektar++. I have tested the solver and also > did a benchmark with Clever & Busse JFM 1974 for the Nusselt number > calculation. There is an excellent agreement. > > > > > > Now, I am trying to do the direct linear stability for RBC. But, I > find there is a problem with the base flow addition for the buoyant scalar. > I have reported this issue on the forum. Could you kindly also check this > issue. > > > > > > Again, thanks for your response. > > > > > > With regards > > > Abhishek > > > > > > > > > > > > On Mon, Feb 5, 2018 at 5:15 PM, David Moxey <D.Moxey(a)exeter.ac.uk> > wrote: > > > Hi Abhishek, > > > > > > These define the various choice of basis functions; the first is the > modified C^0 basis that's defined in Karniadakis & Sherwin. The other two > are Lagrange basis functions: the first is a choice with the number of > quadrature points as one higher than the number of points; the other is > where they are equal (i.e. the collocated basis with diagonal mass matrix). > > > > > > I believe all of those basis functions should work fine for 2D > elements. For 3D, Lagrange only presently works for hex elements, whereas > the C^0 has full support for hex/tet/prism/pyramid. > > > > > > Hope that helps, > > > > > > Dave > > > > > > > On 1 Feb 2018, at 17:42, Abhishek Kumar <abhishek.kir(a)gmail.com> > wrote: > > > > > > > > Dear friends, > > > > > > > > I am new to NEKTAR++. > > > > > > > > Can someone tell me the difference between MODIFIED, GLL_LAGRANGE, > GLL_LAGRANGE_SEM expansions type? > > > > > > > > > > > > With regards > > > > Abhishek Kumar > > > > > > > > ------------------------------------------------------------ > --------------------------- > > > > Abhishek Kumar > > > > Post Doctoral Researcher > > > > Applied Mathematics Research Centre > > > > Coventry University, Coventry CV15FB > > > > The United Kingdom > > > > > > > > Email # ac7600(a)coventry.ac.uk, abhishek.kir(a)gmail.com > > > > ------------------------------------------------------------ > --------------------------- > > > > _______________________________________________ > > > > Nektar-users mailing list > > > > Nektar-users(a)imperial.ac.uk > > > > https://mailman.ic.ac.uk/mailman/listinfo/nektar-users > > > > > > > > > > > > > > > -- > > > With regards > > > Abhishek Kumar > > > > > > ------------------------------------------------------------ > --------------------------- > > > Abhishek Kumar > > > Post Doctoral Researcher > > > Applied Mathematics Research Centre > > > Coventry University, Coventry CV15FB > > > The United Kingdom > > > > > > Email # ac7600(a)coventry.ac.uk, abhishek.kir(a)gmail.com > > > ------------------------------------------------------------ > --------------------------- > > > _______________________________________________ > > > Nektar-users mailing list > > > Nektar-users(a)imperial.ac.uk > > > https://mailman.ic.ac.uk/mailman/listinfo/nektar-users > > > > > > > > > > -- > > With regards > > Abhishek Kumar > > > > ------------------------------------------------------------ > --------------------------- > > Abhishek Kumar > > Post Doctoral Researcher > > Applied Mathematics Research Centre > > Coventry University, Coventry CV15FB > > The United Kingdom > > > > Email # ac7600(a)coventry.ac.uk, abhishek.kir(a)gmail.com > > ------------------------------------------------------------ > --------------------------- > > <rbc_stability.xml>_______________________________________________ > > Nektar-users mailing list > > Nektar-users(a)imperial.ac.uk > > https://mailman.ic.ac.uk/mailman/listinfo/nektar-users > > -- With regards Abhishek Kumar --------------------------------------------------------------------------------------- Abhishek Kumar Post Doctoral Researcher Applied Mathematics Research Centre Coventry University, Coventry CV15FB The United Kingdom Email # ac7600(a)coventry.ac.uk, abhishek.kir(a)gmail.com ---------------------------------------------------------------------------------------

1 0

Re: [Nektar-users] Regarding expansion types
by Abhishek Kumar 07 Feb '18

07 Feb '18

Dear David, Thanks for your response. I have attached the XML file for RBC stability. I am running for Ra=2000 with k=3.117. The critical value of Rayleigh number for this configuration is 1708. The growth rate for Ra=2000 of the most unstable mode should be 2.1464. However, I am getting -9.8699, a negative growth rate, which is not possible as Ra > Ra critical. With regards Abhishek On Wed, Feb 7, 2018 at 2:50 PM, David Moxey <D.Moxey(a)exeter.ac.uk> wrote: > Hi Abhishek, > > Thanks for the update, glad to hear we are producing correct results on > that front. > > Regarding the linear stability; can you share your XML file? It might just > be that the base flow needs to be properly imposed in the appropriate > scalar -- we probably haven't taken scalar fields into account yet in the > setup. > > Thanks, > > Dave > > > On 6 Feb 2018, at 13:20, Abhishek Kumar <abhishek.kir(a)gmail.com> wrote: > > > > Dear David, > > > > Thanks for your answer. Now, I got it. > > > > You would be happy to know that the Rayleigh-Benard convection (RBC) > solver is working properly with Nektar++. I have tested the solver and also > did a benchmark with Clever & Busse JFM 1974 for the Nusselt number > calculation. There is an excellent agreement. > > > > Now, I am trying to do the direct linear stability for RBC. But, I find > there is a problem with the base flow addition for the buoyant scalar. I > have reported this issue on the forum. Could you kindly also check this > issue. > > > > Again, thanks for your response. > > > > With regards > > Abhishek > > > > > > > > On Mon, Feb 5, 2018 at 5:15 PM, David Moxey <D.Moxey(a)exeter.ac.uk> > wrote: > > Hi Abhishek, > > > > These define the various choice of basis functions; the first is the > modified C^0 basis that's defined in Karniadakis & Sherwin. The other two > are Lagrange basis functions: the first is a choice with the number of > quadrature points as one higher than the number of points; the other is > where they are equal (i.e. the collocated basis with diagonal mass matrix). > > > > I believe all of those basis functions should work fine for 2D elements. > For 3D, Lagrange only presently works for hex elements, whereas the C^0 has > full support for hex/tet/prism/pyramid. > > > > Hope that helps, > > > > Dave > > > > > On 1 Feb 2018, at 17:42, Abhishek Kumar <abhishek.kir(a)gmail.com> > wrote: > > > > > > Dear friends, > > > > > > I am new to NEKTAR++. > > > > > > Can someone tell me the difference between MODIFIED, GLL_LAGRANGE, > GLL_LAGRANGE_SEM expansions type? > > > > > > > > > With regards > > > Abhishek Kumar > > > > > > ------------------------------------------------------------ > --------------------------- > > > Abhishek Kumar > > > Post Doctoral Researcher > > > Applied Mathematics Research Centre > > > Coventry University, Coventry CV15FB > > > The United Kingdom > > > > > > Email # ac7600(a)coventry.ac.uk, abhishek.kir(a)gmail.com > > > ------------------------------------------------------------ > --------------------------- > > > _______________________________________________ > > > Nektar-users mailing list > > > Nektar-users(a)imperial.ac.uk > > > https://mailman.ic.ac.uk/mailman/listinfo/nektar-users > > > > > > > > > > -- > > With regards > > Abhishek Kumar > > > > ------------------------------------------------------------ > --------------------------- > > Abhishek Kumar > > Post Doctoral Researcher > > Applied Mathematics Research Centre > > Coventry University, Coventry CV15FB > > The United Kingdom > > > > Email # ac7600(a)coventry.ac.uk, abhishek.kir(a)gmail.com > > ------------------------------------------------------------ > --------------------------- > > _______________________________________________ > > Nektar-users mailing list > > Nektar-users(a)imperial.ac.uk > > https://mailman.ic.ac.uk/mailman/listinfo/nektar-users > > -- With regards Abhishek Kumar --------------------------------------------------------------------------------------- Abhishek Kumar Post Doctoral Researcher Applied Mathematics Research Centre Coventry University, Coventry CV15FB The United Kingdom Email # ac7600(a)coventry.ac.uk, abhishek.kir(a)gmail.com ---------------------------------------------------------------------------------------

1 0