From wangzhicheng at mail.etp.ac.cn Sun Sep 20 01:53:30 2009 From: wangzhicheng at mail.etp.ac.cn (=?gb2312?B?zfXWvrPJ?=) Date: Sun, 20 Sep 2009 14:53:30 +0800 Subject: [Nek5000-users] Need help for Cavity flow Reafile Message-ID: <200909201453299379706@mail.etp.ac.cn> Dear Developers: I have attached a tar docment that I use to run 2D cavity flow. The problem is that I just couldn't get any correct result even in such a simple case. I suppose I have wrongly set some parameters. But after altering some of these parameters in the reafile, I still couldn't get correct result. Please give me some hints for my problem. With my best wishes! Zhicheng Wang 2009-09-20 ??? -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: cavity01.tar Type: application/octet-stream Size: 20480 bytes Desc: not available URL: From wangzhicheng at mail.etp.ac.cn Mon Sep 21 06:07:52 2009 From: wangzhicheng at mail.etp.ac.cn (=?gb2312?B?d2FuZ3poaWNoZW5n?=) Date: Mon, 21 Sep 2009 19:07:52 +0800 (CST) Subject: [Nek5000-users] =?gb2312?b?aGVscCBmb3IgRWxlbWVudCBtZXNoIG1pc21h?= =?gb2312?b?dGNo?= Message-ID: <4AB75E88.000087.17289@app5> Hi developers The problems with cavity flow in previous email has been solved. However, I meet another trouble today. I attached a document that contain .rea file for the case of flow past a cylinder. The mesh is transformed from the format used in code Semtex to Nek5000, by tool sem2nek with some change. But when I try to run this case, nek5000 give me errors like this: **************************************************************************************************************************************************************** Number of processors: 1 REAL wdsize : 8 INTEGER wdsize : 4 Beginning session: /home/wzc/Code/nek5_svn/examples/2Dcylinder/cylinder01.rea timer accuracy: 1.0967255E-06 sec read .rea file nelgt/nelgv/lelt: 54 54 300 lx1 /lx2 /lx3 : 8 6 8 mapping elements to processors 0 54 54 54 54 NELV RANK 0 IEG 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 done :: mapping elements to processors 0 objects found readat time 3.292083740234375E-003 sec done :: read .rea file setup domain topology Right-handed check complete for 54 elements. OK. setvert2d: 8 8 1 setupds2d: 8 0 69 69 321 807 gs_setup: 0 unique labels shared gs_init time 3.2997E-04 seconds 8 54 3456 321 0 5 max multiplicity done :: setup domain topology call usrdat done :: usrdat generate geomerty data vol_t,vol_v: 134.648547351613 134.648547351613 done :: generate geomerty data call usrdat2 done :: usrdat2 regenerate geomerty data 1 vol_t,vol_v: 134.648547351613 134.648547351613 done :: regenerate geomerty data 1 verify mesh topology -5.00000000000000 10.0000000000000 Xrange -5.00000000000000 5.00000000000000 Yrange 0.00000000000000 0.00000000000000 Zrange WARNING1 Element mesh mismatch at: i,j,k,ie: 8 2 1 2 Near X = 0.43169461 0.47521817 0.0000000 , d: 0.0000000 -0.15500086E-02 2.1757104 call exitt: dying ... backtrace(): obtained 8 stack frames. ./nek5000 [0x4d040a] ./nek5000 [0x51f1a6] ./nek5000 [0x42af2d] ./nek5000 [0x403b86] ./nek5000 [0x40334c] ./nek5000 [0x61f8be] /lib/libc.so.6(__libc_start_main+0xf4) [0x7f8bddd421c4] ./nek5000 [0x403299] total elapsed time : 0.12290E-01 sec total solve time incl. I/O : 0.0000 sec time/timestep : 0.0000 sec ******************************************************************************************************************************************************************** With my best wishes Zhicheng -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: cylinder01.tar Type: application/x-tar Size: 51200 bytes Desc: not available URL: From schrader at mech.kth.se Mon Sep 21 09:01:58 2009 From: schrader at mech.kth.se (Lars-Uve Schrader) Date: Mon, 21 Sep 2009 16:01:58 +0200 Subject: [Nek5000-users] perturb.f Message-ID: <4AB78756.6040903@mech.kth.se> Dear Paul and Stefan, I realized that the parameter nfldt had been added to the argument list of outpost in one of the last revisions. I would like to point out that it is missing in one place, namely in perturb.f, line 810. I guess, it should read: call outpost(vxp(1,jpp),vyp(1,jpp),vzp(1,jpp) & ,prp(1,jpp),tp(1,1,jpp),1,s3) Greetings from KTH Stockholm, Lars-Uve. From stefanke at lav.mavt.ethz.ch Mon Sep 21 09:17:11 2009 From: stefanke at lav.mavt.ethz.ch (stefanke) Date: Mon, 21 Sep 2009 16:17:11 +0200 Subject: [Nek5000-users] perturb.f In-Reply-To: <4AB78756.6040903@mech.kth.se> References: <4AB78756.6040903@mech.kth.se> Message-ID: <8A2323AC-912C-4099-8D75-5E9FFDD3F1E8@lav.mavt.ethz.ch> Thx - I just fixed it :) Stefan On Sep 21, 2009, at 4:01 PM, Lars-Uve Schrader wrote: > Dear Paul and Stefan, > > I realized that the parameter nfldt had been added to the argument > list of outpost in one of the last revisions. I would like to point > out that it is missing in one place, namely in > > perturb.f, line 810. > > I guess, it should read: > > call outpost(vxp(1,jpp),vyp(1,jpp),vzp(1,jpp) > & ,prp(1,jpp),tp(1,1,jpp),1,s3) > > > > Greetings from KTH Stockholm, > > > Lars-Uve. > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users From wangzhicheng at mail.etp.ac.cn Tue Sep 22 05:52:40 2009 From: wangzhicheng at mail.etp.ac.cn (=?gb2312?B?d2FuZ3poaWNoZW5n?=) Date: Tue, 22 Sep 2009 18:52:40 +0800 (CST) Subject: [Nek5000-users] =?gb2312?b?RmxvdyBwYXN0IGN5bGluZGVy?= Message-ID: <4AB8AC78.00011A.15172@app5> Hi everyone I finally solved the problem of element mismatch that mentioned in the previous email. This problem is a result of curves-reading bugs in tool sem2nek. Semtex is a quadrilateral spectral element DNS code that has a mesh converter from Gambit mesh generator and a tool---sem2nek, which converter mesh format from semtex to nekton format. So maybe we could use these two converters to create mesh that can be read by nek5000. http://users.monash.edu.au/~bburn/semtex.html Best Regards Zhicheng Wang -------------- next part -------------- An HTML attachment was scrubbed... URL: From stefanke at lav.mavt.ethz.ch Tue Sep 22 07:31:41 2009 From: stefanke at lav.mavt.ethz.ch (stefanke) Date: Tue, 22 Sep 2009 14:31:41 +0200 Subject: [Nek5000-users] help for Element mesh mismatch In-Reply-To: <4AB75E88.000087.17289@app5> References: <4AB75E88.000087.17289@app5> Message-ID: <3010BB68-4C27-44C8-9DA5-0365189F2FB1@lav.mavt.ethz.ch> Hi, it looks like your mesh coordinates and connectivity do not match. The given elemental boundary point (see err msg) is a shared D.O.F (the same physical point is represented multiple times) but there seems to be a difference in coordinates which is larger than the tolerance (1e-3). hth, Stefan On Sep 21, 2009, at 1:07 PM, wangzhicheng wrote: > Hi developers > > The problems with cavity flow in previous email has been solved. > However, I meet another trouble today. > I attached a document that contain .rea file for the case of flow > past a cylinder. The mesh is transformed from the format used in > code Semtex to Nek5000, by tool sem2nek with some change. > But when I try to run this case, nek5000 give me errors like this: > > **************************************************************************************************************************************************************** > > Number of processors: 1 > REAL wdsize : 8 > INTEGER wdsize : 4 > > > Beginning session: > /home/wzc/Code/nek5_svn/examples/2Dcylinder/cylinder01.rea > > > timer accuracy: 1.0967255E-06 sec > > read .rea file > nelgt/nelgv/lelt: 54 54 300 > lx1 /lx2 /lx3 : 8 6 8 > > mapping elements to processors > 0 54 54 54 54 NELV > RANK 0 IEG 1 2 3 4 5 > 6 7 8 > 9 10 11 12 13 14 > 15 16 > 17 18 19 20 21 22 > 23 24 > 25 26 27 28 29 30 > 31 32 > 33 34 35 36 37 38 > 39 40 > 41 42 43 44 45 46 > 47 48 > 49 50 51 52 53 54 > done :: mapping elements to processors > > 0 objects found > readat time 3.292083740234375E-003 sec > done :: read .rea file > > setup domain topology > Right-handed check complete for 54 elements. OK. > setvert2d: 8 8 1 > setupds2d: 8 0 69 69 > 321 807 > gs_setup: 0 unique labels shared > gs_init time 3.2997E-04 seconds 8 54 3456 > 321 0 > 5 max multiplicity > done :: setup domain topology > > call usrdat > done :: usrdat > > generate geomerty data > vol_t,vol_v: 134.648547351613 134.648547351613 > done :: generate geomerty data > > call usrdat2 > done :: usrdat2 > > regenerate geomerty data 1 > vol_t,vol_v: 134.648547351613 134.648547351613 > done :: regenerate geomerty data 1 > > verify mesh topology > -5.00000000000000 10.0000000000000 Xrange > -5.00000000000000 5.00000000000000 Yrange > 0.00000000000000 0.00000000000000 Zrange > WARNING1 Element mesh mismatch at: > i,j,k,ie: 8 2 1 2 > Near X = 0.43169461 0.47521817 0.0000000 , d: > 0.0000000 -0.15500086E-02 2.1757104 > > call exitt: dying ... > > backtrace(): obtained 8 stack frames. > ./nek5000 [0x4d040a] > ./nek5000 [0x51f1a6] > ./nek5000 [0x42af2d] > ./nek5000 [0x403b86] > ./nek5000 [0x40334c] > ./nek5000 [0x61f8be] > /lib/libc.so.6(__libc_start_main+0xf4) [0x7f8bddd421c4] > ./nek5000 [0x403299] > > total elapsed time : 0.12290E-01 sec > total solve time incl. I/O : 0.0000 sec > time/timestep : 0.0000 sec > ******************************************************************************************************************************************************************** > > With my best wishes > > > Zhicheng > > > > > > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users From stefanke at lav.mavt.ethz.ch Tue Sep 22 07:53:52 2009 From: stefanke at lav.mavt.ethz.ch (stefanke) Date: Tue, 22 Sep 2009 14:53:52 +0200 Subject: [Nek5000-users] Flow past cylinder In-Reply-To: <4AB8AC78.00011A.15172@app5> References: <4AB8AC78.00011A.15172@app5> Message-ID: <01EFD903-E236-423F-A490-04DF7D9E0208@lav.mavt.ethz.ch> Hi, a good meshing tool is really missing in nek5000 but I am afraid there is _no_ good high-order quad/hex mesher. In the case of spectral elements you typically end up with big elements and you really need a high order geometry representation otherwise you're not really flexible if it comes to meshing. In addition to it you may run into accuracy issues if you're using a lower (e.g. 2nd-order) representation of your geometry. However high-order surface meshing etc. is very difficult for complex geometries especially for hex meshes. Most of the meshing tools (Cubit, Gambit, ..) are designed for lower-order methods (e.g. FEM,FVM) so they're not ideal at all. Stefan On Sep 22, 2009, at 12:52 PM, wangzhicheng wrote: > Hi everyone > > I finally solved the problem of element mismatch that mentioned in > the previous email. This problem is a result of curves-reading bugs > in tool sem2nek. > > Semtex is a quadrilateral spectral element DNS code that has a mesh > converter from Gambit mesh generator and a tool---sem2nek, which > converter mesh format from semtex to nekton format. So maybe we > could use these two converters to create mesh that can be read by > nek5000. > > http://users.monash.edu.au/~bburn/semtex.html > > > > Best Regards > > Zhicheng Wang > > > > > > > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users From rpacheco at math.asu.edu Fri Sep 25 21:01:02 2009 From: rpacheco at math.asu.edu (Rafael Pacheco) Date: Fri, 25 Sep 2009 19:01:02 -0700 Subject: [Nek5000-users] makenek Message-ID: <20090926015954.M51876@math.asu.edu> Hello everybody, I am running into problems at the outset. I wonder if someone could guide me as to how to compile makenek I get the following error when typing makenek 07 mexico:/home/rpacheco/nek5_svn/trunk/nek > makenek usage: makenek [.usr filename|clean] thank you! Rafael From johan at mech.kth.se Sat Sep 26 01:52:11 2009 From: johan at mech.kth.se (johan at mech.kth.se) Date: Sat, 26 Sep 2009 08:52:11 +0200 Subject: [Nek5000-users] makenek In-Reply-To: <20090926015954.M51876@math.asu.edu> References: <20090926015954.M51876@math.asu.edu> Message-ID: <20090926085211.110ssssv8gggo00w@www.mech.kth.se> Hi Rafael, Did you try: $ makenek pipeflow3d where 'pipeflow3d' is your current case. Johan Quoting Rafael Pacheco : > > Hello everybody, > > I am running into problems at the outset. I wonder if someone could guide me > as to how to compile makenek > > I get the following error when typing makenek > > 07 mexico:/home/rpacheco/nek5_svn/trunk/nek > makenek > usage: makenek [.usr filename|clean] > > > thank you! > Rafael > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program. From stefanke at lav.mavt.ethz.ch Sat Sep 26 03:58:47 2009 From: stefanke at lav.mavt.ethz.ch (stefanke) Date: Sat, 26 Sep 2009 10:58:47 +0200 Subject: [Nek5000-users] makenek In-Reply-To: <20090926015954.M51876@math.asu.edu> References: <20090926015954.M51876@math.asu.edu> Message-ID: check this: https://nek5000.mcs.anl.gov/index.php/Guided_Tutorial hth, Stefan On Sep 26, 2009, at 4:01 AM, Rafael Pacheco wrote: > > Hello everybody, > > I am running into problems at the outset. I wonder if someone could > guide me > as to how to compile makenek > > I get the following error when typing makenek > > 07 mexico:/home/rpacheco/nek5_svn/trunk/nek > makenek > usage: makenek [.usr filename|clean] > > > thank you! > Rafael > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users From rpacheco at math.asu.edu Sat Sep 26 09:36:20 2009 From: rpacheco at math.asu.edu (Rafael Pacheco) Date: Sat, 26 Sep 2009 07:36:20 -0700 Subject: [Nek5000-users] makenek In-Reply-To: <20090926085211.110ssssv8gggo00w@www.mech.kth.se> References: <20090926015954.M51876@math.asu.edu> <20090926085211.110ssssv8gggo00w@www.mech.kth.se> Message-ID: <20090926143124.M52017@math.asu.edu> Hi Johan, obviously I am not knowledgeable on the use of this software, but thanks for your help. I have makenek in the /trunk/nek directory, so I would be grateful if you guide me step by step on how to run an example, or a document that shows how to do this. thanks! Rafael > Hi Rafael, > > Did you try: > > $ makenek pipeflow3d > > where 'pipeflow3d' is your current case. > > Johan > > Quoting Rafael Pacheco : > > > > > Hello everybody, > > > > I am running into problems at the outset. I wonder if someone could guide me > > as to how to compile makenek > > > > I get the following error when typing makenek > > > > 07 mexico:/home/rpacheco/nek5_svn/trunk/nek > makenek > > usage: makenek [.usr filename|clean] > > > > > > thank you! > > Rafael > > _______________________________________________ > > Nek5000-users mailing list > > Nek5000-users at lists.mcs.anl.gov > > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > > > ---------------------------------------------------------------- > This message was sent using IMP, the Internet Messaging Program. > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users -- Rafael -------------------------------------------------------------------- Department of Mathematics and Statistics Arizona State University P.O. Box 871804 Tempe, Arizona 85287-1804 Email: rpacheco at asu.edu http://math.la.asu.edu/~rpacheco -------------------------------------------------------------------- From rpacheco at math.asu.edu Sat Sep 26 09:37:32 2009 From: rpacheco at math.asu.edu (Rafael Pacheco) Date: Sat, 26 Sep 2009 07:37:32 -0700 Subject: [Nek5000-users] makenek In-Reply-To: References: <20090926015954.M51876@math.asu.edu> Message-ID: <20090926143722.M90057@math.asu.edu> Stefan, thanks! Rafael > check this: > https://nek5000.mcs.anl.gov/index.php/Guided_Tutorial > > hth, > Stefan > > On Sep 26, 2009, at 4:01 AM, Rafael Pacheco wrote: > > > > > Hello everybody, > > > > I am running into problems at the outset. I wonder if someone could > > guide me > > as to how to compile makenek > > > > I get the following error when typing makenek > > > > 07 mexico:/home/rpacheco/nek5_svn/trunk/nek > makenek > > usage: makenek [.usr filename|clean] > > > > > > thank you! > > Rafael > > _______________________________________________ > > Nek5000-users mailing list > > Nek5000-users at lists.mcs.anl.gov > > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users -- Rafael -------------------------------------------------------------------- Department of Mathematics and Statistics Arizona State University P.O. Box 871804 Tempe, Arizona 85287-1804 Email: rpacheco at asu.edu http://math.la.asu.edu/~rpacheco --------------------------------------------------------------------