October 2016 - Nek5000-users - mailman3.cels.anl.gov

Re: [Nek5000-users] How to assembly different geometry
by nek5000-users＠lists.mcs.anl.gov 31 Oct '16

31 Oct '16

Hi Alessandro, Running genbox creates a box.rea file. You can copy box.rea to blah.rea where blah is your case name. Ketan

1 0

Re: [Nek5000-users] How to assembly different geometry
by nek5000-users＠lists.mcs.anl.gov 31 Oct '16

31 Oct '16

Hi Alessandro, I believe the problem with the box file that you pasted is that the boundary conditions are not of required length. Each boundary condition must be exactly 3 characters long, and the first boundary conditions that you specified for Box 4,5,6 and 7 is 4 characters long. Deleting one character from those empty boundary conditions makes them of the right length, and genbox runs fine after that. Ketan

1 1

How to assembly different geometry
by nek5000-users＠lists.mcs.anl.gov 30 Oct '16

30 Oct '16

Hi All, I would like to generate a mesh as Figure 2 in http://www.mathematics.uh.edu/docs/math/NASC-preprint-series/2015_2016/Prep… If I build three different boxes, how can I then assemble them? Could anyone please help me? Thanks, Alessandro -- Dr Alessandro Alla Florida State University Department of Scienfitic Computing 400 Dirac Science Library Tallahassee FL 32306-4120 Room: 462 Phone: +1 8506441010 E-Mail: aalla(a)fsu.edu Skype: alessandro.alla www.alessandroalla.com

1 6

Genbox error
by nek5000-users＠lists.mcs.anl.gov 27 Oct '16

27 Oct '16

Hi all, I have troubles with genbox. I have tried all the possible combination ./genbox turbChannel.box ./genbox turbChannel and I get this error. Do you have any idea? input file name: turbChannel forrtl: No such file or directory forrtl: severe (29): file not found, unit 7, file /galileo/home/userexternal/aalla000/exam_2710/turbChannel Image PC Routine Line Source libifcore.so.5 00007F69D5EE1947 Unknown Unknown Unknown libifcore.so.5 00007F69D5F00730 Unknown Unknown Unknown genbox 00000000004014F5 Unknown Unknown Unknown genbox 000000000040134E Unknown Unknown Unknown libc.so.6 00007F69D3FFEB15 Unknown Unknown Unknown genbox 0000000000401259 Unknown Unknown Unknown Thanks, Alessandro -- Dr Alessandro Alla Florida State University Department of Scienfitic Computing 400 Dirac Science Library Tallahassee FL 32306-4120 Room: 462 Phone: +1 8506441010 E-Mail: aalla(a)fsu.edu Skype: alessandro.alla www.alessandroalla.com

1 2

Pauls v2 Integration routine...
by nek5000-users＠lists.mcs.anl.gov 27 Oct '16

27 Oct '16

Paul, Hi... I'd like to use this also; however, was your intent for the argument of the facint_v() sub to be v2, and not v? Also, would you want to pass v2surf back in /mystuff/? if (cbc(f,e,1).eq.'v ') v2surf=v2surf+facint_v(v,area,f,e) Thanks, Murph ----------------------------------------------------------------------------------------------------------- common /mystuff/ v2(lx1*ly1*lz1*lelt) integer e,f n = nx1*ny1*nz1*nelt do i=1,n v2(i)=vx(i,1,1,1)**2+vy(i,1,1,1)**2+vz(i,1,1,1)**2 enddo v2surf = 0 do e=1,nelv ! Integrate over all surfaces with cbc='v ' do f=1,2*ndim if (cbc(f,e,1).eq.'v ') v2surf=v2surf+facint_v(v,area,f,e) enddo enddo v2surf = glsum(v2surf,1) ! Sum across all processors Note that "area" is passed in the TOTAL common block, which I assume you are including in the routine. On 10/26/2016 1:00 PM, nek5000-users-request(a)lists.mcs.anl.gov wrote: > Send Nek5000-users mailing list submissions to > nek5000-users(a)lists.mcs.anl.gov > > To subscribe or unsubscribe via the World Wide Web, visit > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > or, via email, send a message with subject or body 'help' to > nek5000-users-request(a)lists.mcs.anl.gov > > You can reach the person managing the list at > nek5000-users-owner(a)lists.mcs.anl.gov > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Nek5000-users digest..." > > > Today's Topics: > > 1. Order of appearance : USERBC and USERCHECK > (nek5000-users(a)lists.mcs.anl.gov) > 2. Re: Integrate velocity field along a line > (nek5000-users(a)lists.mcs.anl.gov) > > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users(a)lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users

1 1

Integrate velocity field along a line
by nek5000-users＠lists.mcs.anl.gov 26 Oct '16

26 Oct '16

Good morning Neks! I was wondering if it possible to integrate a function over the boundary in the userchk. Let me develop : The domain is a box, I want to integrate the boundary velocity field (prescribed in userbc) at the wall (bottom boundary) : In latex it would be int_{bottom boundary } u_{bound}^2 + v_{bound}^2 + w_{bound}^2v dx_{bound} Is there a function like glsc3() that would allow me to do so? Thank you for your time, Best wishes, Arnold

1 2

Order of appearance : USERBC and USERCHECK
by nek5000-users＠lists.mcs.anl.gov 26 Oct '16

26 Oct '16

Hi all, I was wondering what was being called first between USERBC and USERCHECK when we solve for fluid with nek. Does anyone know about this? I still could not tell even after checking the sources. Thank you for your help, Best wishes, Arnold

1 0

Arbitrary mesh deformation
by nek5000-users＠lists.mcs.anl.gov 25 Oct '16

25 Oct '16

Hello, I would like to deform a mesh, given an arbitrary normal wall deformation. Ideally I would like to solve a laplace problem to find the "volumetric" deformation function, like in the ALE approach. Note that I need to deform the mesh only in the preprocessing, while for the simulation the mesh is kept fixed. Any hints? Thanks, Andrea

1 0

Mesh consistency check failed for merged 2D meshes
by nek5000-users＠lists.mcs.anl.gov 25 Oct '16

25 Oct '16

Hi, I have problems with merging two 2D meshes. I want to merge two meshes generated with genbox: 1. walls.box This mesh is an "Y"-type mesh. It goes from beeing a square of side 10 into a square of side 0.5 via two circles (of radii 1.1 and 1.0). 2. center.box This is an "Box"-type mesh. It is a square of size %%%%%%%%%%%%%%%%%%%%%%%% This is walls.box: reafile.rea 2 1 # Y 4 -8 0. 0. bbccb 0.5 0.6 1. 1.1 10. 0 1 1 W ,W ,E ,E , %%%%%%%%%%%%%%%%% And this is center.box: reafile.rea 2 1 # Box -2 -2 -0.5 0.5 1. -0.5 0.5 1. W ,W ,W ,W , %%%%%%% I can plot the two meshes on top of each other, and the elements seem to align perfectly. I use nekmerge2 (from the example case "expansion") to generate a merged .rea file: nekmerge2 center walls bottom I now obtain a bottom.rea file. I generate bottom.map using genmap (tolerance 0.2). I compile (using a bottom.usr file and a SIZE file) using makenek. I run serially using "nek". Now I get several warnings of the type: WARNING2 Element mesh mismatch at: I,J,K,IE: 1 2 1 1 Near Y = -0.50000000 -0.41269471 0.0000000 , d: 0.0000000 -0.36629354E-01 2.0000000 And then "Mesh consistency check failed. EXITING in VRDSMSH." No output is generated. What is the problem here? Have I missed some critical aspect of merging meshes? Best, Johan

1 1

Dirichlet BC for scalars
by nek5000-users＠lists.mcs.anl.gov 25 Oct '16

25 Oct '16

I am a new user of Nek, and am learning to possibilities of the program. I have successfully implemented a lid-driven cavity. I now want to, in addidtion, solve the problem for a scalar with certain boundary conditions (I did choose the temperature for this). I use the .box file: reafile.rea 2 2 # Box 1 -20 -20 0.0 1.0 1.0 0.0 1.0 1.0 W ,W ,W ,v , t ,t ,O ,I , Thus: the domains is a 2D 1x1 square where: * The y = 1 boundary is the "lid" whose velocity can be set in the .usr file (Dirichlet BC) * The vertical walls also have Dirichlet BC * The bottom wall is an "open boundary" for the temperature * The top wall is a n "insulator" for the temperature Given the .rea file generated with this .box file, I set the initial conditions in useric: subroutine useric (ix,iy,iz,ieg) include 'SIZE' include 'TOTAL' include 'NEKUSE' ux=0.0 uy=0.0 uz=0.0 temp=0 return end I set the Dirichlet BC in userbc: subroutine userbc (ix,iy,iz,iside,ieg) c NOTE ::: This subroutine MAY NOT be called by every process include 'SIZE' include 'TOTAL' include 'NEKUSE' ux=1 uy=0.0 uz=0.0 temp=x return end I expected this to initialize the velocity and temperature to zero in the entire domain, except for: * Top boundary has velocity 1 in the x-direction * The right boundary has temperature 1 But when I visualize the results of the simulation, I find that the temperature is set to "x" in the entire domain, not only at the boundarys that "should be set". However, the velocity boundary condition works as I expected, and is set only for the "v" boundaries. What am I doing wrong here? How may I set up dirichlet BC for the temperature?

1 1