From nek5000-users at lists.mcs.anl.gov Wed May 6 11:08:39 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 6 May 2015 16:08:39 +0000 Subject: [Nek5000-users] >9 passive scalars Message-ID: Hello Neks. We are working on a problem where will need more than 9 passive scalars, and 9 appears to be the default maximum number. I know we can go beyond that, but I'm looking for advice on how to make that happen. I suspect will need to modify some of the main routines. (FYI -- we're working the .rea format). Anyone have advice? Thanks. Mike From nek5000-users at lists.mcs.anl.gov Wed May 6 11:19:41 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 6 May 2015 16:19:41 +0000 Subject: [Nek5000-users] >9 passive scalars In-Reply-To: References: Message-ID: This can be done in the .usr file, no mod to source. Set ldimt sufficiently high. There is, in fact, an "add_temp" routine that should work --- it's not been exercised a lot, however, so it may or may not fill the bill. Look for it in navier5.f Paul ________________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Wednesday, May 06, 2015 11:08 AM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] >9 passive scalars Hello Neks. We are working on a problem where will need more than 9 passive scalars, and 9 appears to be the default maximum number. I know we can go beyond that, but I'm looking for advice on how to make that happen. I suspect will need to modify some of the main routines. (FYI -- we're working the .rea format). Anyone have advice? Thanks. Mike _______________________________________________ Nek5000-users mailing list Nek5000-users at lists.mcs.anl.gov https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users From nek5000-users at lists.mcs.anl.gov Thu May 7 16:03:08 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 7 May 2015 17:03:08 -0400 Subject: [Nek5000-users] Compiling nek5000 with MPI on Mac. Message-ID: Hi, I'm trying to use nek5000 with MPI on a Mac. OS is Yosemite. Fortran compiler = gfortran: nikolai:eddy wenhaowu$ gfortran --version GNU Fortran (GCC) 5.1.0 Copyright (C) 2014 Free Software Foundation, Inc. C compiler = mpicc from mpich: nikolai:eddy wenhaowu$ mpicc --version Apple LLVM version 6.1.0 (clang-602.0.49) (based on LLVM 3.6.0svn) Target: x86_64-apple-darwin14.3.0 Thread model: posix And I followed the instruction 'QuickStart': Here is the beginning of my 'makenek': ----------------------------------------------------------------- #!/bin/bash # Nek5000 build config file # (c) 2008,2009,2010 UCHICAGO ARGONNE, LLC # source path SOURCE_ROOT="/Users/wenhaowu/Documents/testcases/nek5/trunk/nek" # Fortran compiler F77="gfortran" # C compiler CC="mpicc" # pre-processor symbol list # (set PPLIST=? to get a list of available symbols) #PPLIST="?" # plug-in list PLUGIN_LIST="" # OPTIONAL SETTINGS # ----------------- # enable MPI (default true) #IFMPI="false" ... ----------------------------------------------------------------- Here is my command: ./makenek eddy_uv Here is an example what goes wrong: ------------------------------------------------------------------------- mpicc -c -O2 -DPTRSIZE8 -DMPI -DLONGINT8 -DUNDERSCORE -DGLOBAL_LONG_LONG -DPRE FIX=jl_ /Users/wenhaowu/Documents/testcases/nek5/trunk/nek/nek_comm.c -o obj/nek _comm.o In file included from /Users/wenhaowu/Documents/testcases/nek5/trunk/nek/nek_com m.c:17: /opt/local/include/mpich-mp/mpi.h:116:56: error: expected identifier static const MPI_Datatype mpich_mpi_char MPICH_ATTR_TYPE_TAG(char) = MPI_CHAR; ^ /opt/local/include/mpich-mp/mpi.h:24:106: note: expanded from macro 'MPICH_ATTR_TYPE_TAG' # define MPICH_ATTR_TYPE_TAG(type) __attribute__((type_tag_for_datatype(MPI,type))) ^ :2:13: note: expanded from here #define MPI 1 ------------------------------------------------------------------------- The problem is nek5000 has #defined MPI to be 1, but the mpich header file 'mpi.h' uses the symbol 'MPI' (for a completely different reason). How do other people use nek5000 with mpi on a Mac? Thanks. Sincerely, Wenhao Wu -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Thu May 7 16:28:15 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 7 May 2015 21:28:15 +0000 Subject: [Nek5000-users] Compiling nek5000 with MPI on Mac. In-Reply-To: References: Message-ID: Hi Nicolai, Try C compiler cc and IFMPI="false" first. (Otherwise you need to be consistent and use mpif77 installed for gfortran) Aleks ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Thursday, May 07, 2015 4:03 PM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] Compiling nek5000 with MPI on Mac. Hi, I'm trying to use nek5000 with MPI on a Mac. OS is Yosemite. Fortran compiler = gfortran: nikolai:eddy wenhaowu$ gfortran --version GNU Fortran (GCC) 5.1.0 Copyright (C) 2014 Free Software Foundation, Inc. C compiler = mpicc from mpich: nikolai:eddy wenhaowu$ mpicc --version Apple LLVM version 6.1.0 (clang-602.0.49) (based on LLVM 3.6.0svn) Target: x86_64-apple-darwin14.3.0 Thread model: posix And I followed the instruction 'QuickStart': Here is the beginning of my 'makenek': ----------------------------------------------------------------- #!/bin/bash # Nek5000 build config file # (c) 2008,2009,2010 UCHICAGO ARGONNE, LLC # source path SOURCE_ROOT="/Users/wenhaowu/Documents/testcases/nek5/trunk/nek" # Fortran compiler F77="gfortran" # C compiler CC="mpicc" # pre-processor symbol list # (set PPLIST=? to get a list of available symbols) #PPLIST="?" # plug-in list PLUGIN_LIST="" # OPTIONAL SETTINGS # ----------------- # enable MPI (default true) #IFMPI="false" ... ----------------------------------------------------------------- Here is my command: ./makenek eddy_uv Here is an example what goes wrong: ------------------------------------------------------------------------- mpicc -c -O2 -DPTRSIZE8 -DMPI -DLONGINT8 -DUNDERSCORE -DGLOBAL_LONG_LONG -DPRE FIX=jl_ /Users/wenhaowu/Documents/testcases/nek5/trunk/nek/nek_comm.c -o obj/nek _comm.o In file included from /Users/wenhaowu/Documents/testcases/nek5/trunk/nek/nek_com m.c:17: /opt/local/include/mpich-mp/mpi.h:116:56: error: expected identifier static const MPI_Datatype mpich_mpi_char MPICH_ATTR_TYPE_TAG(char) = MPI_CHAR; ^ /opt/local/include/mpich-mp/mpi.h:24:106: note: expanded from macro 'MPICH_ATTR_TYPE_TAG' # define MPICH_ATTR_TYPE_TAG(type) __attribute__((type_tag_for_datatype(MPI,type))) ^ :2:13: note: expanded from here #define MPI 1 ------------------------------------------------------------------------- The problem is nek5000 has #defined MPI to be 1, but the mpich header file 'mpi.h' uses the symbol 'MPI' (for a completely different reason). How do other people use nek5000 with mpi on a Mac? Thanks. Sincerely, Wenhao Wu -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Thu May 7 20:17:28 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 7 May 2015 20:17:28 -0500 Subject: [Nek5000-users] Rayleigh example - no flow development Message-ID: Dear all neks, Recently I have tried to modify the rayleigh example to my case. However, I discovered that no flow field is developed after a long time. All velocities are zero. I guess I miss something important in the example. Anyone knows which part of the code could lead to this? Also, although I start with T = -z, I got asymmetric initial condition, something not as I specified as viewed in postnek. Did I miss any point in the .rea and .usr? The files are attached here. They are modified rea and usr from rayleigh to my case. Thank you very much. Regards, Simon -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: cyl.rea Type: application/octet-stream Size: 5837 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: cyl.usr Type: application/octet-stream Size: 5791 bytes Desc: not available URL: From nek5000-users at lists.mcs.anl.gov Thu May 7 21:25:25 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 8 May 2015 02:25:25 +0000 Subject: [Nek5000-users] Rayleigh example - no flow development In-Reply-To: References: Message-ID: Hi Simon, What is your Rayleigh number? Paul ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Thursday, May 07, 2015 8:17 PM To: nek5000-use. Subject: [Nek5000-users] Rayleigh example - no flow development Dear all neks, Recently I have tried to modify the rayleigh example to my case. However, I discovered that no flow field is developed after a long time. All velocities are zero. I guess I miss something important in the example. Anyone knows which part of the code could lead to this? Also, although I start with T = -z, I got asymmetric initial condition, something not as I specified as viewed in postnek. Did I miss any point in the .rea and .usr? The files are attached here. They are modified rea and usr from rayleigh to my case. Thank you very much. Regards, Simon -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Thu May 7 21:32:06 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 7 May 2015 21:32:06 -0500 Subject: [Nek5000-users] Rayleigh example - no flow development In-Reply-To: References: Message-ID: Hi Paul, In my modified version, Ra = 5e7, Pr = 5. For the example, I keep it as the original values Ra ~1700, Pr = 1. Regards, Simon 2015-05-07 21:25 GMT-05:00 : > > Hi Simon, > > What is your Rayleigh number? > > Paul > > ------------------------------ > *From:* nek5000-users-bounces at lists.mcs.anl.gov [ > nek5000-users-bounces at lists.mcs.anl.gov] on behalf of > nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] > *Sent:* Thursday, May 07, 2015 8:17 PM > *To:* nek5000-use. > *Subject:* [Nek5000-users] Rayleigh example - no flow development > > Dear all neks, > > Recently I have tried to modify the rayleigh example to my case. > However, I discovered that no flow field is developed after a long time. > All velocities are zero. I guess I miss something important in the example. > Anyone knows which part of the code could lead to this? > > Also, although I start with T = -z, I got asymmetric initial condition, > something not as I specified as viewed in postnek. Did I miss any point in > the .rea and .usr? The files are attached here. They are modified rea and > usr from rayleigh to my case. > > Thank you very much. > > Regards, > Simon > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Thu May 7 22:23:03 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 8 May 2015 03:23:03 +0000 Subject: [Nek5000-users] Rayleigh example - no flow development In-Reply-To: References: Message-ID: Hi Simon, It's usually a good idea to break symmetry in some way in order to get the flow to start moving. Usually that's done in useric by calling some type of random number generator to put small amplitude disturbances into the field. Paul ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Thursday, May 07, 2015 8:17 PM To: nek5000-use. Subject: [Nek5000-users] Rayleigh example - no flow development Dear all neks, Recently I have tried to modify the rayleigh example to my case. However, I discovered that no flow field is developed after a long time. All velocities are zero. I guess I miss something important in the example. Anyone knows which part of the code could lead to this? Also, although I start with T = -z, I got asymmetric initial condition, something not as I specified as viewed in postnek. Did I miss any point in the .rea and .usr? The files are attached here. They are modified rea and usr from rayleigh to my case. Thank you very much. Regards, Simon -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Fri May 8 02:55:42 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 8 May 2015 07:55:42 +0000 Subject: [Nek5000-users] Rayleigh example - no flow development In-Reply-To: References: Message-ID: Hi Simon, Breaking the symmetry is indeed a good idea in general. Although, at high Rayleigh numbers this is not entirely necessary in my experience. I typically don?t add any perturbations. I noticed a small typo in userf: ?ffz = bouy?, not ?ffz = buoy?. Also, you load a restart file in the .rea file. This will overload the initial conditions, T=-z, you have originally specified in useric. Regards, Gijs On 08 May 2015, at 05:23, nek5000-users at lists.mcs.anl.gov wrote: Hi Simon, It's usually a good idea to break symmetry in some way in order to get the flow to start moving. Usually that's done in useric by calling some type of random number generator to put small amplitude disturbances into the field. Paul ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Thursday, May 07, 2015 8:17 PM To: nek5000-use. Subject: [Nek5000-users] Rayleigh example - no flow development Dear all neks, Recently I have tried to modify the rayleigh example to my case. However, I discovered that no flow field is developed after a long time. All velocities are zero. I guess I miss something important in the example. Anyone knows which part of the code could lead to this? Also, although I start with T = -z, I got asymmetric initial condition, something not as I specified as viewed in postnek. Did I miss any point in the .rea and .usr? The files are attached here. They are modified rea and usr from rayleigh to my case. Thank you very much. Regards, Simon _______________________________________________ Nek5000-users mailing list Nek5000-users at lists.mcs.anl.gov https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Sun May 10 22:48:47 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Sun, 10 May 2015 22:48:47 -0500 Subject: [Nek5000-users] Rayleigh example - no flow development In-Reply-To: References: Message-ID: Dear all, I modified the rea and the buoy now works. However, I got all the velocity, temp diverge. As shown in my usr and rea, I set the initial and boundary condition both as T = -z, ux=uy=uz=0. But in my first fld output, I got a different look - My geometry starts from z = -0.5 to 0.5. There shouldn't be any asymmetry and T should not be less than -0.5. Is there anything else wrong with my setting? I believe it is the strange initial condition that leads to my problem. I set DT = 0.2E-2 in rea already. The divergence persists. It diverges right at the start of the simulation. Do you know the source of the problem? Thank you very much. Regards, Simon -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: initial condition.jpg Type: image/jpeg Size: 38661 bytes Desc: not available URL: From nek5000-users at lists.mcs.anl.gov Tue May 12 08:41:31 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Tue, 12 May 2015 15:41:31 +0200 Subject: [Nek5000-users] Newby's question: Inflow IC / BC from external file Message-ID: Hi Neks, I'm new to this amazing software, and I'd like to better understand it in order to run some cases. Right now I'm trying to impose a steady inflow velocity profile computed outside Nek as boundary / initial condition. I've found lots of threads about this, spread all over the forum, but i can't actually get how to "practically" do this. I know this is a dull question, but i can't really sort it out; any help is greatly appreciated! Thank you Francesco PS As a reference: https://lists.mcs.anl.gov/mailman/htdig/nek5000-users/2014-September/003058.html http://lists.mcs.anl.gov/pipermail/nek5000-users/2010-December/001151.html -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Tue May 12 10:30:30 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Tue, 12 May 2015 15:30:30 +0000 Subject: [Nek5000-users] Newby's question: Inflow IC / BC from external file In-Reply-To: References: Message-ID: Hi Francesco, Normally you would set the fluid BC in the .rea file to be "v " (v + 2 spaces) ... then, in userbc() in the .usr file you would do something like: ux=(1-y*y) .... Does this approach work for you? Paul ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Tuesday, May 12, 2015 8:41 AM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] Newby's question: Inflow IC / BC from external file Hi Neks, I'm new to this amazing software, and I'd like to better understand it in order to run some cases. Right now I'm trying to impose a steady inflow velocity profile computed outside Nek as boundary / initial condition. I've found lots of threads about this, spread all over the forum, but i can't actually get how to "practically" do this. I know this is a dull question, but i can't really sort it out; any help is greatly appreciated! Thank you Francesco PS As a reference: https://lists.mcs.anl.gov/mailman/htdig/nek5000-users/2014-September/003058.html http://lists.mcs.anl.gov/pipermail/nek5000-users/2010-December/001151.html -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Tue May 12 11:36:13 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Tue, 12 May 2015 18:36:13 +0200 Subject: [Nek5000-users] Newby's question: Inflow IC / BC from external file Message-ID: Hi, thank you for your answer. Actually, I'd need something more: I would like to use as inflow a streamwise corner flow boundary layer velocity field, of which there exist a self-similar solution. https://digital.library.adelaide.edu.au/dspace/bitstream/2440/56317/8/02chapters1-5.pdf At the moment I've tried to use an approximate analytical solution, of which I'm unsatisfied, and to modify the blasius.usr provided within the examples in order to compute my bc directly inside Nek, but I had some bad times with element numbering. Since I would like to have a more general tool, I would like to know how to read from an external file a stationary velocity field and to properly assign it as a boundary / initial condition. I've read that "doing that in serial, it's straightforward", but I really can't get how do to that. Thank you for your help Sincerely Francesco -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Wed May 13 14:02:45 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 13 May 2015 15:02:45 -0400 Subject: [Nek5000-users] Compiling nek5000 with MPI on Mac. In-Reply-To: References: Message-ID: Hi, Aleks, First, I have tried to set F77 = "mpif77" and IFMPI = "false". And the example works fine. The result is: ------------------------------------------------------ ... ... ############################################################# # Compilation successful! # ############################################################# __TEXT __DATA __OBJC others dec hex 2633728 78450688 0 4295122944 4376207360 104d7a000 --------------------------------------------------------- Then I set: -------------------------------- #!/bin/bash # Nek5000 build config file # (c) 2008,2009,2010 UCHICAGO ARGONNE, LLC # source path SOURCE_ROOT="/Users/wenhaowu/Documents/testcases/nek5/trunk/nek" # Fortran compiler F77="mpif77" # C compiler CC="mpicc" ... # OPTIONAL SETTINGS # ----------------- # enable MPI (default true) #IFMPI="false" ... ---------------------------------- When I try to run "makenek", it still turns out several errors like this: ----------------------------------- mpicc -c -O2 -DPTRSIZE8 -DMPI -DLONGINT8 -DUNDERSCORE -DGLOBAL_LONG_LONG -DPREFIX=jl_ /Users/wenhaowu/Documents/testcases/nek5/trunk/nek/nek_comm.c -o obj/nek_comm.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -DPTRSIZE8 -DMPI -DLONGINT8 -DUNDERSCORE -DGLOBAL_LONG_LONG -I/Users/wenhaowu/Documents/testcases/nek5/examples/eddy -I/Users/wenhaowu/Documents/testcases/nek5/trunk/nek -I./ /Users/wenhaowu/Documents/testcases/nek5/trunk/nek/init_plugin.f -o obj/init_plugin.o In file included from /Users/wenhaowu/Documents/testcases/nek5/trunk/nek/nek_comm.c:17: /opt/local/include/mpich-mp/mpi.h:116:56: error: expected identifier static const MPI_Datatype mpich_mpi_char MPICH_ATTR_TYPE_TAG(char) = MPI_CHAR; ^ /opt/local/include/mpich-mp/mpi.h:24:106: note: expanded from macro 'MPICH_ATTR_TYPE_TAG' # define MPICH_ATTR_TYPE_TAG(type) __attribute__((type_tag_for_datatype(MPI,type))) ^ :2:13: note: expanded from here #define MPI 1 ^ ... ------------------------ Sincerely, Wenhao Wu On Thu, May 7, 2015 at 5:28 PM, wrote: > Hi Nicolai, > > Try C compiler cc and > > IFMPI="false" > > first. > > (Otherwise you need to be consistent and use mpif77 installed for gfortran) > > Aleks > > > > > > ------------------------------ > *From:* nek5000-users-bounces at lists.mcs.anl.gov [ > nek5000-users-bounces at lists.mcs.anl.gov] on behalf of > nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] > *Sent:* Thursday, May 07, 2015 4:03 PM > *To:* nek5000-users at lists.mcs.anl.gov > *Subject:* [Nek5000-users] Compiling nek5000 with MPI on Mac. > > Hi, > I'm trying to use nek5000 with MPI on a Mac. OS is Yosemite. > > Fortran compiler = gfortran: > > nikolai:eddy wenhaowu$ gfortran --version > GNU Fortran (GCC) 5.1.0 > Copyright (C) 2014 Free Software Foundation, Inc. > > C compiler = mpicc from mpich: > > nikolai:eddy wenhaowu$ mpicc --version > Apple LLVM version 6.1.0 (clang-602.0.49) (based on LLVM 3.6.0svn) > Target: x86_64-apple-darwin14.3.0 > Thread model: posix > > > And I followed the instruction 'QuickStart': > > Here is the beginning of my 'makenek': > ----------------------------------------------------------------- > #!/bin/bash > > > # Nek5000 build config file > > > # (c) 2008,2009,2010 UCHICAGO ARGONNE, LLC > > > > # source path > > > SOURCE_ROOT="/Users/wenhaowu/Documents/testcases/nek5/trunk/nek" > > > # Fortran compiler > > > F77="gfortran" > > > # C compiler > > > CC="mpicc" > > > # pre-processor symbol list > > > # (set PPLIST=? to get a list of available symbols) > > > #PPLIST="?" > > > > # plug-in list > > > PLUGIN_LIST="" > > > > # OPTIONAL SETTINGS > > > # ----------------- > > > > # enable MPI (default true) > > > #IFMPI="false" > > ... > ----------------------------------------------------------------- > > Here is my command: > ./makenek eddy_uv > > Here is an example what goes wrong: > ------------------------------------------------------------------------- > mpicc -c -O2 -DPTRSIZE8 -DMPI -DLONGINT8 -DUNDERSCORE -DGLOBAL_LONG_LONG > -DPRE > > FIX=jl_ /Users/wenhaowu/Documents/testcases/nek5/trunk/nek/nek_comm.c -o > obj/nek > > _comm.o > > In file included from > /Users/wenhaowu/Documents/testcases/nek5/trunk/nek/nek_com > > m.c:17: > > /opt/local/include/mpich-mp/mpi.h:116:56: error: expected identifier > > static const MPI_Datatype mpich_mpi_char > MPICH_ATTR_TYPE_TAG(char) > > = MPI_CHAR; > > ^ > > /opt/local/include/mpich-mp/mpi.h:24:106: note: expanded from macro > 'MPICH_ATTR_TYPE_TAG' > > # define MPICH_ATTR_TYPE_TAG(type) > __attribute__((type_tag_for_datatype(MPI,type))) > > > ^ > > :2:13: note: expanded from here > > #define MPI 1 > ------------------------------------------------------------------------- > > The problem is nek5000 has #defined MPI to be 1, but the mpich header file > 'mpi.h' uses the symbol 'MPI' (for a completely different reason). > > How do other people use nek5000 with mpi on a Mac? > > Thanks. > > Sincerely, > > Wenhao Wu > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Thu May 14 17:23:19 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 14 May 2015 18:23:19 -0400 Subject: [Nek5000-users] unusual quantities Message-ID: Dear Nek5000 Users, I have some unusual quantities I am hoping to compute while solving the 3D Rayleigh-Benard problem. They are described in the attached pdf-file. I am hoping someone has a suggestion how to do this using Nek5000. Thanks, Mike -------------- next part -------------- A non-text attachment was scrubbed... Name: 3DnoWhy.pdf Type: application/octetstream Size: 86477 bytes Desc: not available URL: From nek5000-users at lists.mcs.anl.gov Sun May 17 22:19:56 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Sun, 17 May 2015 22:19:56 -0500 (CDT) Subject: [Nek5000-users] unusual quantities In-Reply-To: References: Message-ID: Hi Mike, I've taken a pass at setting something up to compute these. See attached. Other than testing that it compiles and runs, I've not tested to see if it actually computing what you want. If it's not, I would guess that it's close... Please have a look at the attached files. In principle, you can use a script like the following genbox << EOF rb3d.box EOF genmap << EOF rb3d 0.1 EOF makenek rb3d and be set to go with nek rb3d or nekmpi rb3d 8 say. You would adjust resolution by editing rb3d.box --- making certain that the SIZE file is compliant. In particular, to get the z-integrals that you want you need to be certain that lelx and lely are big enough so that they match the number of elements in your x-y plane. Ditto with nelx and nely, which are set in the .usr file. (The limits, lelx and lely, are set in the SIZE file.) Paul On Thu, 14 May 2015, nek5000-users at lists.mcs.anl.gov wrote: > Dear Nek5000 Users, > > I have some unusual quantities I am hoping to compute > while solving the 3D Rayleigh-Benard problem. They > are described in the attached pdf-file. I am hoping > someone has a suggestion how to do this using Nek5000. > > Thanks, > > Mike > > -------------- next part -------------- A non-text attachment was scrubbed... Name: t.tgz Type: application/x-gtar-compressed Size: 16207 bytes Desc: URL: From nek5000-users at lists.mcs.anl.gov Mon May 18 04:46:04 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Mon, 18 May 2015 11:46:04 +0200 Subject: [Nek5000-users] Nek optimal setting Message-ID: Hi Nek's I'm going to run Nek5000 on a Bull cluster. In order to submit the process, i have to set the number of nodes, of tasks per node and of threads per core. Setting the threads per core equals to one, how can I adjust the number of nodes and the number of cores per node keeping Nek5000 in an optimal arrangement? In my case, I have 33000 elements with a spectral order equal to 8. Alessandro -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Fri May 22 08:21:34 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 22 May 2015 15:21:34 +0200 (CEST) Subject: [Nek5000-users] read temperature Message-ID: Hi Neks I'm trying to save some variables in this way: call outpost(u_bf,v_bf,w_bf,du_bfdy,vis_t,'BF_') after a simulation. then i would to read the aforementioned variables: ifto = .true. b_flow = 'BF_Channel0.f00001' call load_fld(b_flow) but it seems to read in the correct way just the first 4 variables: xyz min 0.0000 -1.0000 0.0000 uvwpt min 0.0000 0.96051E-20 0.96051E-20 -9.6269 0.96051E-20 PS min 0.0000 0.99000E+22 xyz max 12.560 1.0000 6.2800 uvwpt max 0.97894 0.80000E-19 0.80000E-19 9.6744 0.80000E-19 PS max 0.0000 -0.99000E+22 because the tmin e tmax are not zero in BF_Channel0.00001! Could someone help me? Thanks -- Mr. Mirko FaranoPh.D. student at Politecnico di BariTel.: +39-0883529289 / +39-0805963462Mob.: +39-3202342719Department of Mechanics Mathematics and Management - Fluid Machinery and Energy SystemsVia Re David, 200 - 70125 BARI (Italy)mirko.farano at poliba.itIn joint supervision with theEcole Nationale Sup?rieure d'Arts et M?tiers de Paris - ParistechTel.: +33-144246436Mob.: +33-789822134Dynfluid Laboratory151 Boulevard de l'H?pital - 75013 Paris (France)mirko.farano at ensam.eu -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Fri May 22 10:11:10 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 22 May 2015 15:11:10 +0000 Subject: [Nek5000-users] read temperature In-Reply-To: References: Message-ID: Hi Mirko, Did you check whether you temperature filed is dumped into the field file by checking whether there is a letter 'T' in the header of head BF_Channel0.f00001|strings|head -1 ? If not than you could either turn in the temperature dump by modifying turbChannel.rea's line F TEMPERATURE to T TEMPERATURE or inserting the line in userchk() before your call to outpost() ifto = .true. (which is the output specification) Aleks ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Friday, May 22, 2015 8:21 AM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] read temperature Hi Neks I'm trying to save some variables in this way: call outpost(u_bf,v_bf,w_bf,du_bfdy,vis_t,'BF_') after a simulation. then i would to read the aforementioned variables: ifto = .true. b_flow = 'BF_Channel0.f00001' call load_fld(b_flow) but it seems to read in the correct way just the first 4 variables: xyz min 0.0000 -1.0000 0.0000 uvwpt min 0.0000 0.96051E-20 0.96051E-20 -9.6269 0.96051E-20 PS min 0.0000 0.99000E+22 xyz max 12.560 1.0000 6.2800 uvwpt max 0.97894 0.80000E-19 0.80000E-19 9.6744 0.80000E-19 PS max 0.0000 -0.99000E+22 because the tmin e tmax are not zero in BF_Channel0.00001! Could someone help me? Thanks -- Mr. Mirko Farano Ph.D. student at Politecnico di Bari Tel.: +39-0883529289 / +39-0805963462 Mob.: +39-3202342719 Department of Mechanics Mathematics and Management - Fluid Machinery and Energy Systems Via Re David, 200 - 70125 BARI (Italy) mirko.farano at poliba.it In joint supervision with the Ecole Nationale Sup?rieure d'Arts et M?tiers de Paris - Paristech Tel.: +33-144246436 Mob.: +33-789822134 Dynfluid Laboratory 151 Boulevard de l'H?pital - 75013 Paris (France) mirko.farano at ensam.eu -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Fri May 22 12:15:18 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 22 May 2015 19:15:18 +0200 (CEST) Subject: [Nek5000-users] R: Re: read temperature Message-ID: Hi Aleks, this is the first line: #std 4 8 8 8 7680 7680 0.5487520544140E-02 0 0 1 XUPT so the temperature is written, moreover when i see the BF_Channel0.f00001 in visit, i see the right value of the temperature variable. -- Mr. Mirko FaranoPh.D. student at Politecnico di BariTel.: +39-0883529289 / +39-0805963462Mob.: +39-3202342719Department of Mechanics Mathematics and Management - Fluid Machinery and Energy SystemsVia Re David, 200 - 70125 BARI (Italy)mirko.farano at poliba.itIn joint supervision with theEcole Nationale Sup?rieure d'Arts et M?tiers de Paris - ParistechTel.: +33-144246436Mob.: +33-789822134Dynfluid Laboratory151 Boulevard de l'H?pital - 75013 Paris (France)mirko.farano at ensam.eu ----Messaggio originale---- Da: nek5000-users at lists.mcs.anl.gov Data: 22-mag-2015 17.11 A: "nek5000-users at lists.mcs.anl.gov" Ogg: Re: [Nek5000-users] read temperature --> Hi Mirko, Did you check whether you temperature filed is dumped into the field file by checking whether there is a letter 'T' in the header of head BF_Channel0.f00001|strings|head -1 ? If not than you could either turn in the temperature dump by modifying turbChannel.rea's line F TEMPERATURE to T TEMPERATURE or inserting the line in userchk() before your call to outpost() ifto = .true. (which is the output specification) Aleks From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Friday, May 22, 2015 8:21 AM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] read temperature Hi Neks I'm trying to save some variables in this way: call outpost(u_bf,v_bf,w_bf,du_bfdy,vis_t,'BF_') after a simulation. then i would to read the aforementioned variables: ifto = .true. b_flow = 'BF_Channel0.f00001' call load_fld(b_flow) but it seems to read in the correct way just the first 4 variables: xyz min 0.0000 -1.0000 0.0000 uvwpt min 0.0000 0.96051E-20 0.96051E-20 -9.6269 0.96051E-20 PS min 0.0000 0.99000E+22 xyz max 12.560 1.0000 6.2800 uvwpt max 0.97894 0.80000E-19 0.80000E-19 9.6744 0.80000E-19 PS max 0.0000 -0.99000E+22 because the tmin e tmax are not zero in BF_Channel0.00001! Could someone help me? Thanks -- Mr. Mirko Farano Ph.D. student at Politecnico di Bari Tel.: +39-0883529289 / +39-0805963462 Mob.: +39-3202342719 Department of Mechanics Mathematics and Management - Fluid Machinery and Energy Systems Via Re David, 200 - 70125 BARI (Italy) mirko.farano at poliba.it In joint supervision with the Ecole Nationale Sup?rieure d'Arts et M?tiers de Paris - Paristech Tel.: +33-144246436 Mob.: +33-789822134 Dynfluid Laboratory 151 Boulevard de l'H?pital - 75013 Paris (France) mirko.farano at ensam.eu -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Fri May 22 13:26:53 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 22 May 2015 18:26:53 +0000 Subject: [Nek5000-users] R: Re: read temperature In-Reply-To: References: Message-ID: Thanks, Mirko, A curious behavior... Could you contact me offline at obabko at mcs.anl.gov? Aleks ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Friday, May 22, 2015 12:15 PM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] R: Re: read temperature Hi Aleks, this is the first line: #std 4 8 8 8 7680 7680 0.5487520544140E-02 0 0 1 XUPT so the temperature is written, moreover when i see the BF_Channel0.f00001 in visit, i see the right value of the temperature variable. -- Mr. Mirko Farano Ph.D. student at Politecnico di Bari Tel.: +39-0883529289 / +39-0805963462 Mob.: +39-3202342719 Department of Mechanics Mathematics and Management - Fluid Machinery and Energy Systems Via Re David, 200 - 70125 BARI (Italy) mirko.farano at poliba.it In joint supervision with the Ecole Nationale Sup?rieure d'Arts et M?tiers de Paris - Paristech Tel.: +33-144246436 Mob.: +33-789822134 Dynfluid Laboratory 151 Boulevard de l'H?pital - 75013 Paris (France) mirko.farano at ensam.eu ----Messaggio originale---- Da: nek5000-users at lists.mcs.anl.gov Data: 22-mag-2015 17.11 A: "nek5000-users at lists.mcs.anl.gov" Ogg: Re: [Nek5000-users] read temperature Hi Mirko, Did you check whether you temperature filed is dumped into the field file by checking whether there is a letter 'T' in the header of head BF_Channel0.f00001|strings|head -1 ? If not than you could either turn in the temperature dump by modifying turbChannel.rea's line F TEMPERATURE to T TEMPERATURE or inserting the line in userchk() before your call to outpost() ifto = .true. (which is the output specification) Aleks ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Friday, May 22, 2015 8:21 AM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] read temperature Hi Neks I'm trying to save some variables in this way: call outpost(u_bf,v_bf,w_bf,du_bfdy,vis_t,'BF_') after a simulation. then i would to read the aforementioned variables: ifto = .true. b_flow = 'BF_Channel0.f00001' call load_fld(b_flow) but it seems to read in the correct way just the first 4 variables: xyz min 0.0000 -1.0000 0.0000 uvwpt min 0.0000 0.96051E-20 0.96051E-20 -9.6269 0.96051E-20 PS min 0.0000 0.99000E+22 xyz max 12.560 1.0000 6.2800 uvwpt max 0.97894 0.80000E-19 0.80000E-19 9.6744 0.80000E-19 PS max 0.0000 -0.99000E+22 because the tmin e tmax are not zero in BF_Channel0.00001! Could someone help me? Thanks -- Mr. Mirko Farano Ph.D. student at Politecnico di Bari Tel.: +39-0883529289 / +39-0805963462 Mob.: +39-3202342719 Department of Mechanics Mathematics and Management - Fluid Machinery and Energy Systems Via Re David, 200 - 70125 BARI (Italy) mirko.farano at poliba.it In joint supervision with the Ecole Nationale Sup?rieure d'Arts et M?tiers de Paris - Paristech Tel.: +33-144246436 Mob.: +33-789822134 Dynfluid Laboratory 151 Boulevard de l'H?pital - 75013 Paris (France) mirko.farano at ensam.eu -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Wed May 27 07:29:57 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 27 May 2015 14:29:57 +0200 (CEST) Subject: [Nek5000-users] variable viscosity: perturbative and adjoint Message-ID: Hi, Nek's Could I compute the perturbative equations, and so the adjoint one, for a variable viscosity flow? If yes, do I need to do like in the DNS simulation? Thanks. -- Mr. Mirko FaranoPh.D. student at Politecnico di BariTel.: +39-0883529289 / +39-0805963462Mob.: +39-3202342719Department of Mechanics Mathematics and Management - Fluid Machinery and Energy SystemsVia Re David, 200 - 70125 BARI (Italy)mirko.farano at poliba.itIn joint supervision with theEcole Nationale Sup?rieure d'Arts et M?tiers de Paris - ParistechTel.: +33-144246436Mob.: +33-789822134Dynfluid Laboratory151 Boulevard de l'H?pital - 75013 Paris (France)mirko.farano at ensam.eu -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Thu May 28 13:41:32 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 28 May 2015 18:41:32 +0000 Subject: [Nek5000-users] Detected non-right-handed element Message-ID: Dear Nek users, I am simulating a jet-type problem and created my own grid using Prex. I tried using that grid in the turbulent jet example so I could visualize it. However, after a few seconds, the simulation ends and gives me this error: WARNINGb: Detected non-right-handed element. Number 1 V1-8: 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 call exitt: dying ... After perusing the nek users archives, I've found that a similar problem occurred in 2010 but had been fixed in the latest repo at the time. It seemed that it only had that problem for certain compilers (gnu, ifort, etc). The compiler I am using is mpif77. Any input would be appreciated. Regards, Goran Marjanovic -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Fri May 29 03:01:49 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 29 May 2015 10:01:49 +0200 Subject: [Nek5000-users] Using results from a run as boundary condition for another one Message-ID: Goodmoring Neks, I'm running some calculation on a laminar channel flow that contains a geometrical disturbance, similar to a "bump". It has been proved experimentally that the flow upwind is not affected by the presence of this impingement. Thus, in order to reduce the computational cost, I would like to *split* the case, say compute the "upwind" laminar channel flow separately, and then use the results as boundary conditions for a successive run, in my case containing the bump itself. My question is: is it possible to extract the computed fields from a given plane, and then use them as an "inflow" boundary condition on a successive run? I think I'll need to use the same mesh topology, in order to avoid any interpolation. Thank you for your attention Sincerely Ciccio -------------- next part -------------- An HTML attachment was scrubbed... URL: