From nek5000-users at lists.mcs.anl.gov Sat Jul 9 18:10:29 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Sat, 9 Jul 2011 18:10:29 -0500 Subject: [Nek5000-users] variable density incompressible flow Message-ID: Hi I am a total newbie with regards to nek5000. I want to simulate a variable density incompressible flow where the variation in density is brought about by the motion of an interface separating two fluids. The interface is tracked via a level set method. To achieve this, an appropriate source term is added in the continuity equation, thus making the velocity field divergent. To do this using nek5000, I need to know how to input a user defined divergence of velocity. Could anyone throw some light on this or guide me to some documentation/example which would help me? Thanks, Regards. -Navin Fogla From nek5000-users at lists.mcs.anl.gov Sat Jul 9 21:03:04 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Sat, 9 Jul 2011 21:03:04 -0500 (CDT) Subject: [Nek5000-users] variable density incompressible flow In-Reply-To: References: Message-ID: Hi Navin, It's not clear that nek will be the right tool for this... Depends on what you're after. If your flow is incompressible, I don't see why you need a source term for the divergence, since one can have div.U=0 even if rho is variable. That being said, there is a usrdiv array that you can fill. See the peristaltic flow exampe in nek5_svn/examples/peris/peris.usr Look for "div" Paul On Sat, 9 Jul 2011, nek5000-users at lists.mcs.anl.gov wrote: > Hi > > I am a total newbie with regards to nek5000. I want to simulate a variable > density incompressible flow where the variation in density is brought about > by the motion of an interface separating two fluids. The interface is tracked > via a level set method. To achieve this, an appropriate source term is added > in the continuity equation, thus making the velocity field divergent. To do > this using nek5000, I need to know how to input a user defined divergence of > velocity. Could anyone throw some light on this or guide me to some > documentation/example which would help me? > > Thanks, > Regards. > > -Navin Fogla > > > > > > _______________________________________________ > 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 Wed Jul 13 16:55:35 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 13 Jul 2011 17:55:35 -0400 Subject: [Nek5000-users] Vorticity Message-ID: <20110713175535.1042430oct81vz2v@webmail.vt.edu> Hi, I need to calculate the vorticity vector and the rate of strain tensor at any arbitrary point in the domain. I wonder if these values are computed somewhere in the code or if I should interpolate them in the usr file. I appreciate any kind of help. Thanks in advance, Regards, Alireza Karimi From nek5000-users at lists.mcs.anl.gov Wed Jul 13 17:09:48 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 13 Jul 2011 17:09:48 -0500 (CDT) Subject: [Nek5000-users] Vorticity In-Reply-To: <20110713175535.1042430oct81vz2v@webmail.vt.edu> References: <20110713175535.1042430oct81vz2v@webmail.vt.edu> Message-ID: Hi Alireza, For vorticity computation, you can call call comp_vort3(vort,work1,work2,vx,vy,vz) from userchk with the appropriate definitions of vorticity and work arrays like common /scren/ vort (lx1,ly1,lz1,lelv,3) ! x y z components $ , work1(lx1,ly1,lz1,lelv) $ , work2(lx1,ly1,lz1,lelv) Let me check what we have for the calculation of strain tensor. Best, Aleks On Wed, 13 Jul 2011, nek5000-users at lists.mcs.anl.gov wrote: > Hi, > > I need to calculate the vorticity vector and the rate of strain tensor at any > arbitrary point in the domain. I wonder if these values are computed > somewhere in the code or if I should interpolate them in the usr file. I > appreciate any kind of help. > > Thanks in advance, > > > Regards, > Alireza Karimi > > _______________________________________________ > 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 Wed Jul 13 23:59:08 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 13 Jul 2011 23:59:08 -0500 (CDT) Subject: [Nek5000-users] Vorticity In-Reply-To: Message-ID: <600285192.94128.1310619548370.JavaMail.root@zimbra.anl.gov> Hi Alireza, For a strain stress computation you can call subroutine comp_sij call comp_sij(sij,nij,vx,vy,vz $ ,ur,us,ut,vr,vs,vt,wr,ws,wt) (navier5.f:2225) with nij=6 for 3D somewhere from userchk once you declare the stress component array sij and arrays with local coordinate r-s-t derivatives of velocity components common /csij/ $ sij(lx1*ly1*lz1,6,lelv) ! 6 components $ ,ur (lx1*ly1*lz1) ... $ ,wt (lx1*ly1*lz1) Best, Aleks ----- Original Message ----- From: nek5000-users at lists.mcs.anl.gov To: nek5000-users at lists.mcs.anl.gov Sent: Wednesday, July 13, 2011 5:09:48 PM Subject: Re: [Nek5000-users] Vorticity Hi Alireza, For vorticity computation, you can call call comp_vort3(vort,work1,work2,vx,vy,vz) from userchk with the appropriate definitions of vorticity and work arrays like common /scren/ vort (lx1,ly1,lz1,lelv,3) ! x y z components $ , work1(lx1,ly1,lz1,lelv) $ , work2(lx1,ly1,lz1,lelv) Let me check what we have for the calculation of strain tensor. Best, Aleks On Wed, 13 Jul 2011, nek5000-users at lists.mcs.anl.gov wrote: > Hi, > > I need to calculate the vorticity vector and the rate of strain tensor at any > arbitrary point in the domain. I wonder if these values are computed > somewhere in the code or if I should interpolate them in the usr file. I > appreciate any kind of help. > > Thanks in advance, > > > Regards, > Alireza Karimi > > _______________________________________________ > 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 From nek5000-users at lists.mcs.anl.gov Thu Jul 14 12:24:34 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 14 Jul 2011 13:24:34 -0400 Subject: [Nek5000-users] Vorticity In-Reply-To: References: Message-ID: <20110714132434.117602kzz1p9nuc2@webmail.vt.edu> Hi, Thanks for your quick reply. Just I wanna know if I can use the interpolated values of velocity at some arbitrary points in the domain (not necessarily the grid points) as the input arguments to theses subroutines. Regards, Alireza Quoting nek5000-users-request at lists.mcs.anl.gov: > Send Nek5000-users mailing list submissions to > nek5000-users at 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 at lists.mcs.anl.gov > > You can reach the person managing the list at > nek5000-users-owner at 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. Vorticity (nek5000-users at lists.mcs.anl.gov) > 2. Re: Vorticity (nek5000-users at lists.mcs.anl.gov) > 3. Re: Vorticity (nek5000-users at lists.mcs.anl.gov) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Wed, 13 Jul 2011 17:55:35 -0400 > From: nek5000-users at lists.mcs.anl.gov > Subject: [Nek5000-users] Vorticity > To: nek5000-users at lists.mcs.anl.gov > Message-ID: <20110713175535.1042430oct81vz2v at webmail.vt.edu> > Content-Type: text/plain; charset=ISO-8859-1; DelSp="Yes"; > format="flowed" > > Hi, > > I need to calculate the vorticity vector and the rate of strain tensor > at any arbitrary point in the domain. I wonder if these values are > computed somewhere in the code or if I should interpolate them in the > usr file. I appreciate any kind of help. > > Thanks in advance, > > > Regards, > Alireza Karimi > > > > ------------------------------ > > Message: 2 > Date: Wed, 13 Jul 2011 17:09:48 -0500 (CDT) > From: nek5000-users at lists.mcs.anl.gov > Subject: Re: [Nek5000-users] Vorticity > To: nek5000-users at lists.mcs.anl.gov > Message-ID: > Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed > > Hi Alireza, > > For vorticity computation, you can call > > call comp_vort3(vort,work1,work2,vx,vy,vz) > > from userchk with the appropriate definitions of vorticity and work arrays > like > > common /scren/ vort (lx1,ly1,lz1,lelv,3) ! x y z components > $ , work1(lx1,ly1,lz1,lelv) > $ , work2(lx1,ly1,lz1,lelv) > > Let me check what we have for the calculation of strain tensor. > Best, > Aleks > > > On Wed, 13 Jul 2011, nek5000-users at lists.mcs.anl.gov wrote: > >> Hi, >> >> I need to calculate the vorticity vector and the rate of strain >> tensor at any >> arbitrary point in the domain. I wonder if these values are computed >> somewhere in the code or if I should interpolate them in the usr file. I >> appreciate any kind of help. >> >> Thanks in advance, >> >> >> Regards, >> Alireza Karimi >> >> _______________________________________________ >> Nek5000-users mailing list >> Nek5000-users at lists.mcs.anl.gov >> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > > ------------------------------ > > Message: 3 > Date: Wed, 13 Jul 2011 23:59:08 -0500 (CDT) > From: nek5000-users at lists.mcs.anl.gov > Subject: Re: [Nek5000-users] Vorticity > To: nek5000-users at lists.mcs.anl.gov > Message-ID: > <600285192.94128.1310619548370.JavaMail.root at zimbra.anl.gov> > Content-Type: text/plain; charset=utf-8 > > Hi Alireza, > > > For a strain stress computation you can call subroutine comp_sij > > call comp_sij(sij,nij,vx,vy,vz > $ ,ur,us,ut,vr,vs,vt,wr,ws,wt) > > > (navier5.f:2225) with nij=6 for 3D somewhere from userchk once you > declare the stress component array sij and arrays with local > coordinate r-s-t derivatives of velocity components > > common /csij/ > $ sij(lx1*ly1*lz1,6,lelv) ! 6 components > $ ,ur (lx1*ly1*lz1) > ... > $ ,wt (lx1*ly1*lz1) > > > Best, > Aleks > > > > > > > ----- Original Message ----- > From: nek5000-users at lists.mcs.anl.gov > To: nek5000-users at lists.mcs.anl.gov > Sent: Wednesday, July 13, 2011 5:09:48 PM > Subject: Re: [Nek5000-users] Vorticity > > Hi Alireza, > > For vorticity computation, you can call > > call comp_vort3(vort,work1,work2,vx,vy,vz) > > from userchk with the appropriate definitions of vorticity and work arrays > like > > common /scren/ vort (lx1,ly1,lz1,lelv,3) ! x y z components > $ , work1(lx1,ly1,lz1,lelv) > $ , work2(lx1,ly1,lz1,lelv) > > Let me check what we have for the calculation of strain tensor. > Best, > Aleks > > > On Wed, 13 Jul 2011, nek5000-users at lists.mcs.anl.gov wrote: > >> Hi, >> >> I need to calculate the vorticity vector and the rate of strain >> tensor at any >> arbitrary point in the domain. I wonder if these values are computed >> somewhere in the code or if I should interpolate them in the usr file. I >> appreciate any kind of help. >> >> Thanks in advance, >> >> >> Regards, >> Alireza Karimi >> >> _______________________________________________ >> 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 > > > ------------------------------ > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > > End of Nek5000-users Digest, Vol 29, Issue 2 > ******************************************** > -- Alireza Karimi PhD Candidate Department of Engineering Science and Mechanics Virginia Polytechnic Institute and State University Blacksburg, VA 24061 From nek5000-users at lists.mcs.anl.gov Thu Jul 14 12:43:09 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 14 Jul 2011 12:43:09 -0500 (CDT) Subject: [Nek5000-users] Vorticity In-Reply-To: <20110714132434.117602kzz1p9nuc2@webmail.vt.edu> References: <20110714132434.117602kzz1p9nuc2@webmail.vt.edu> Message-ID: Hi Alireza, The best option is to call the vorticity/s_ij routines, then interpolate the fields that they generate. Attached is a routine that should do the interpolation --- I just hacked into another routine that was available and haven't tested, but it should work. Note that if you want to change your point list subsequent to the initial call, you should pass a handle so indicating. Interpolation requires two steps (done by the attached code): 1. Find the points (x,y,z)_i in the (r,s,t,e) parameter space (done once, and expensive) 2. Evaulate quantities of interest at (r,s,t,e)_i (potentially done often, and less expensive) Please let me know if you have any questions on usage, etc. Paul On Thu, 14 Jul 2011, nek5000-users at lists.mcs.anl.gov wrote: > Hi, > > Thanks for your quick reply. Just I wanna know if I can use the interpolated > values of velocity at some arbitrary points in the domain (not necessarily > the grid points) as the input arguments to theses subroutines. > > Regards, > Alireza > > > > > > Quoting nek5000-users-request at lists.mcs.anl.gov: > >> Send Nek5000-users mailing list submissions to >> nek5000-users at 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 at lists.mcs.anl.gov >> >> You can reach the person managing the list at >> nek5000-users-owner at 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. Vorticity (nek5000-users at lists.mcs.anl.gov) >> 2. Re: Vorticity (nek5000-users at lists.mcs.anl.gov) >> 3. Re: Vorticity (nek5000-users at lists.mcs.anl.gov) >> >> >> ---------------------------------------------------------------------- >> >> Message: 1 >> Date: Wed, 13 Jul 2011 17:55:35 -0400 >> From: nek5000-users at lists.mcs.anl.gov >> Subject: [Nek5000-users] Vorticity >> To: nek5000-users at lists.mcs.anl.gov >> Message-ID: <20110713175535.1042430oct81vz2v at webmail.vt.edu> >> Content-Type: text/plain; charset=ISO-8859-1; DelSp="Yes"; >> format="flowed" >> >> Hi, >> >> I need to calculate the vorticity vector and the rate of strain tensor >> at any arbitrary point in the domain. I wonder if these values are >> computed somewhere in the code or if I should interpolate them in the >> usr file. I appreciate any kind of help. >> >> Thanks in advance, >> >> >> Regards, >> Alireza Karimi >> >> >> >> ------------------------------ >> >> Message: 2 >> Date: Wed, 13 Jul 2011 17:09:48 -0500 (CDT) >> From: nek5000-users at lists.mcs.anl.gov >> Subject: Re: [Nek5000-users] Vorticity >> To: nek5000-users at lists.mcs.anl.gov >> Message-ID: >> Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed >> >> Hi Alireza, >> >> For vorticity computation, you can call >> >> call comp_vort3(vort,work1,work2,vx,vy,vz) >> >> from userchk with the appropriate definitions of vorticity and work arrays >> like >> >> common /scren/ vort (lx1,ly1,lz1,lelv,3) ! x y z components >> $ , work1(lx1,ly1,lz1,lelv) >> $ , work2(lx1,ly1,lz1,lelv) >> >> Let me check what we have for the calculation of strain tensor. >> Best, >> Aleks >> >> >> On Wed, 13 Jul 2011, nek5000-users at lists.mcs.anl.gov wrote: >> >>> Hi, >>> >>> I need to calculate the vorticity vector and the rate of strain tensor at >>> any >>> arbitrary point in the domain. I wonder if these values are computed >>> somewhere in the code or if I should interpolate them in the usr file. I >>> appreciate any kind of help. >>> >>> Thanks in advance, >>> >>> >>> Regards, >>> Alireza Karimi >>> >>> _______________________________________________ >>> Nek5000-users mailing list >>> Nek5000-users at lists.mcs.anl.gov >>> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users >> >> >> ------------------------------ >> >> Message: 3 >> Date: Wed, 13 Jul 2011 23:59:08 -0500 (CDT) >> From: nek5000-users at lists.mcs.anl.gov >> Subject: Re: [Nek5000-users] Vorticity >> To: nek5000-users at lists.mcs.anl.gov >> Message-ID: >> <600285192.94128.1310619548370.JavaMail.root at zimbra.anl.gov> >> Content-Type: text/plain; charset=utf-8 >> >> Hi Alireza, >> >> >> For a strain stress computation you can call subroutine comp_sij >> >> call comp_sij(sij,nij,vx,vy,vz >> $ ,ur,us,ut,vr,vs,vt,wr,ws,wt) >> >> >> (navier5.f:2225) with nij=6 for 3D somewhere from userchk once you declare >> the stress component array sij and arrays with local coordinate r-s-t >> derivatives of velocity components >> >> common /csij/ >> $ sij(lx1*ly1*lz1,6,lelv) ! 6 components >> $ ,ur (lx1*ly1*lz1) >> ... >> $ ,wt (lx1*ly1*lz1) >> >> >> Best, >> Aleks >> >> >> >> >> >> >> ----- Original Message ----- >> From: nek5000-users at lists.mcs.anl.gov >> To: nek5000-users at lists.mcs.anl.gov >> Sent: Wednesday, July 13, 2011 5:09:48 PM >> Subject: Re: [Nek5000-users] Vorticity >> >> Hi Alireza, >> >> For vorticity computation, you can call >> >> call comp_vort3(vort,work1,work2,vx,vy,vz) >> >> from userchk with the appropriate definitions of vorticity and work arrays >> like >> >> common /scren/ vort (lx1,ly1,lz1,lelv,3) ! x y z components >> $ , work1(lx1,ly1,lz1,lelv) >> $ , work2(lx1,ly1,lz1,lelv) >> >> Let me check what we have for the calculation of strain tensor. >> Best, >> Aleks >> >> >> On Wed, 13 Jul 2011, nek5000-users at lists.mcs.anl.gov wrote: >> >>> Hi, >>> >>> I need to calculate the vorticity vector and the rate of strain tensor at >>> any >>> arbitrary point in the domain. I wonder if these values are computed >>> somewhere in the code or if I should interpolate them in the usr file. I >>> appreciate any kind of help. >>> >>> Thanks in advance, >>> >>> >>> Regards, >>> Alireza Karimi >>> >>> _______________________________________________ >>> 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 >> >> >> ------------------------------ >> >> _______________________________________________ >> Nek5000-users mailing list >> Nek5000-users at lists.mcs.anl.gov >> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users >> >> >> End of Nek5000-users Digest, Vol 29, Issue 2 >> ******************************************** >> > > > > -- > Alireza Karimi > PhD Candidate > Department of Engineering Science and Mechanics > Virginia Polytechnic Institute and State University > Blacksburg, VA 24061 > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users -------------- next part -------------- C----------------------------------------------------------------------- subroutine interp_v3(uvw,xyz,v1,v2,v3,n) c c evaluate velocity for list of points xyz c c Note: -- modify c intpts to get rid off " WARNING: point on boundary or ..." include 'SIZE' include 'TOTAL' real uvw(ldim,n),xyz(ldim,n),v1(1),v2(1),v3(1) logical ifjac,ifpts parameter(nmax=lpart,nfldmax=ldim) common /rv_intp/ pts(ldim*nmax) common /iv_intp/ ihandle common /outtmp/ wrk(lx1*ly1*lz1*lelt,nfldmax) integer icalld,e save icalld data icalld /0/ nxyz = nx1*ny1*nz1 ntot = nxyz*nelt if (n.gt.nmax) call exitti ('ABORT: interp_v() n > nmax!$',n) if (nelgt.ne.nelgv) call exitti $ ('ABORT: interp_v() nelgt.ne.nelgv not yet supported!$',nelgv) do i=1,n ! ? not moving -> save? pts(i) = xyz(1,i) pts(i + n) = xyz(2,i) if (if3d) pts(i + n*2) = xyz(3,i) enddo if (icalld.eq.0) then ! interpolation setu !? intpts_done(ih_intp_v)? icalld = 1 tolin = 1.e-8 call intpts_setup(tolin,ihandle) endif nflds = ndim ! number of fields to interpolate ! pack working array call opcopy(wrk(1,1),wrk(1,2),wrk(1,3),v1,v2,v3) ! interpolate ifjac = .true. ! output transpose (of Jacobian) ifpts = .true. ! find points call intpts(wrk,nflds,pts,n,uvw,ifjac,ifpts,ihandle) ! copy array instead? return end c----------------------------------------------------------------------- From nek5000-users at lists.mcs.anl.gov Thu Jul 14 16:36:41 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 14 Jul 2011 17:36:41 -0400 Subject: [Nek5000-users] Vorticity In-Reply-To: References: Message-ID: <20110714173641.82945dfuhqfm4aqx@webmail.vt.edu> Hi, Thanks for the code. I just have a problem regarding s_ij. Should I use opcopy subroutine twice as follows? call opcopy(wrk(1,1),wrk(1,2),wrk(1,3),v1,v2,v3) call opcopy(wrk(1,4),wrk(1,5),wrk(1,6),v4,v5,v6) Or I should use another subroutine in this case in order to pack working array? Regards, Alireza Quoting nek5000-users-request at lists.mcs.anl.gov: > Send Nek5000-users mailing list submissions to > nek5000-users at 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 at lists.mcs.anl.gov > > You can reach the person managing the list at > nek5000-users-owner at 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. Vorticity (nek5000-users at lists.mcs.anl.gov) > 2. Re: Vorticity (nek5000-users at lists.mcs.anl.gov) > 3. Re: Vorticity (nek5000-users at lists.mcs.anl.gov) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Wed, 13 Jul 2011 17:55:35 -0400 > From: nek5000-users at lists.mcs.anl.gov > Subject: [Nek5000-users] Vorticity > To: nek5000-users at lists.mcs.anl.gov > Message-ID: <20110713175535.1042430oct81vz2v at webmail.vt.edu> > Content-Type: text/plain; charset=ISO-8859-1; DelSp="Yes"; > format="flowed" > > Hi, > > I need to calculate the vorticity vector and the rate of strain tensor > at any arbitrary point in the domain. I wonder if these values are > computed somewhere in the code or if I should interpolate them in the > usr file. I appreciate any kind of help. > > Thanks in advance, > > > Regards, > Alireza Karimi > > > > ------------------------------ > > Message: 2 > Date: Wed, 13 Jul 2011 17:09:48 -0500 (CDT) > From: nek5000-users at lists.mcs.anl.gov > Subject: Re: [Nek5000-users] Vorticity > To: nek5000-users at lists.mcs.anl.gov > Message-ID: > Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed > > Hi Alireza, > > For vorticity computation, you can call > > call comp_vort3(vort,work1,work2,vx,vy,vz) > > from userchk with the appropriate definitions of vorticity and work arrays > like > > common /scren/ vort (lx1,ly1,lz1,lelv,3) ! x y z components > $ , work1(lx1,ly1,lz1,lelv) > $ , work2(lx1,ly1,lz1,lelv) > > Let me check what we have for the calculation of strain tensor. > Best, > Aleks > > > On Wed, 13 Jul 2011, nek5000-users at lists.mcs.anl.gov wrote: > >> Hi, >> >> I need to calculate the vorticity vector and the rate of strain >> tensor at any >> arbitrary point in the domain. I wonder if these values are computed >> somewhere in the code or if I should interpolate them in the usr file. I >> appreciate any kind of help. >> >> Thanks in advance, >> >> >> Regards, >> Alireza Karimi >> >> _______________________________________________ >> Nek5000-users mailing list >> Nek5000-users at lists.mcs.anl.gov >> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > > ------------------------------ > > Message: 3 > Date: Wed, 13 Jul 2011 23:59:08 -0500 (CDT) > From: nek5000-users at lists.mcs.anl.gov > Subject: Re: [Nek5000-users] Vorticity > To: nek5000-users at lists.mcs.anl.gov > Message-ID: > <600285192.94128.1310619548370.JavaMail.root at zimbra.anl.gov> > Content-Type: text/plain; charset=utf-8 > > Hi Alireza, > > > For a strain stress computation you can call subroutine comp_sij > > call comp_sij(sij,nij,vx,vy,vz > $ ,ur,us,ut,vr,vs,vt,wr,ws,wt) > > > (navier5.f:2225) with nij=6 for 3D somewhere from userchk once you > declare the stress component array sij and arrays with local > coordinate r-s-t derivatives of velocity components > > common /csij/ > $ sij(lx1*ly1*lz1,6,lelv) ! 6 components > $ ,ur (lx1*ly1*lz1) > ... > $ ,wt (lx1*ly1*lz1) > > > Best, > Aleks > > > > > > > ----- Original Message ----- > From: nek5000-users at lists.mcs.anl.gov > To: nek5000-users at lists.mcs.anl.gov > Sent: Wednesday, July 13, 2011 5:09:48 PM > Subject: Re: [Nek5000-users] Vorticity > > Hi Alireza, > > For vorticity computation, you can call > > call comp_vort3(vort,work1,work2,vx,vy,vz) > > from userchk with the appropriate definitions of vorticity and work arrays > like > > common /scren/ vort (lx1,ly1,lz1,lelv,3) ! x y z components > $ , work1(lx1,ly1,lz1,lelv) > $ , work2(lx1,ly1,lz1,lelv) > > Let me check what we have for the calculation of strain tensor. > Best, > Aleks > > > On Wed, 13 Jul 2011, nek5000-users at lists.mcs.anl.gov wrote: > >> Hi, >> >> I need to calculate the vorticity vector and the rate of strain >> tensor at any >> arbitrary point in the domain. I wonder if these values are computed >> somewhere in the code or if I should interpolate them in the usr file. I >> appreciate any kind of help. >> >> Thanks in advance, >> >> >> Regards, >> Alireza Karimi >> >> _______________________________________________ >> 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 > > > ------------------------------ > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > > End of Nek5000-users Digest, Vol 29, Issue 2 > ******************************************** > -- Alireza Karimi PhD Candidate Department of Engineering Science and Mechanics Virginia Polytechnic Institute and State University Blacksburg, VA 24061 From nek5000-users at lists.mcs.anl.gov Thu Jul 14 16:47:33 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 14 Jul 2011 16:47:33 -0500 (CDT) Subject: [Nek5000-users] Vorticity In-Reply-To: <20110714173641.82945dfuhqfm4aqx@webmail.vt.edu> Message-ID: <1444764930.97000.1310680053573.JavaMail.root@zimbra.anl.gov> Hi Alizera, I would use copy routine (math.f:422) for each array separately w/ n=nx1*ny1*nz1*nelv Best, Aleks ----- Original Message ----- From: nek5000-users at lists.mcs.anl.gov To: nek5000-users at lists.mcs.anl.gov Sent: Thursday, July 14, 2011 4:36:41 PM Subject: Re: [Nek5000-users] Vorticity Hi, Thanks for the code. I just have a problem regarding s_ij. Should I use opcopy subroutine twice as follows? call opcopy(wrk(1,1),wrk(1,2),wrk(1,3),v1,v2,v3) call opcopy(wrk(1,4),wrk(1,5),wrk(1,6),v4,v5,v6) Or I should use another subroutine in this case in order to pack working array? Regards, Alireza Quoting nek5000-users-request at lists.mcs.anl.gov: > Send Nek5000-users mailing list submissions to > nek5000-users at 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 at lists.mcs.anl.gov > > You can reach the person managing the list at > nek5000-users-owner at 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. Vorticity (nek5000-users at lists.mcs.anl.gov) > 2. Re: Vorticity (nek5000-users at lists.mcs.anl.gov) > 3. Re: Vorticity (nek5000-users at lists.mcs.anl.gov) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Wed, 13 Jul 2011 17:55:35 -0400 > From: nek5000-users at lists.mcs.anl.gov > Subject: [Nek5000-users] Vorticity > To: nek5000-users at lists.mcs.anl.gov > Message-ID: <20110713175535.1042430oct81vz2v at webmail.vt.edu> > Content-Type: text/plain; charset=ISO-8859-1; DelSp="Yes"; > format="flowed" > > Hi, > > I need to calculate the vorticity vector and the rate of strain tensor > at any arbitrary point in the domain. I wonder if these values are > computed somewhere in the code or if I should interpolate them in the > usr file. I appreciate any kind of help. > > Thanks in advance, > > > Regards, > Alireza Karimi > > > > ------------------------------ > > Message: 2 > Date: Wed, 13 Jul 2011 17:09:48 -0500 (CDT) > From: nek5000-users at lists.mcs.anl.gov > Subject: Re: [Nek5000-users] Vorticity > To: nek5000-users at lists.mcs.anl.gov > Message-ID: > Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed > > Hi Alireza, > > For vorticity computation, you can call > > call comp_vort3(vort,work1,work2,vx,vy,vz) > > from userchk with the appropriate definitions of vorticity and work arrays > like > > common /scren/ vort (lx1,ly1,lz1,lelv,3) ! x y z components > $ , work1(lx1,ly1,lz1,lelv) > $ , work2(lx1,ly1,lz1,lelv) > > Let me check what we have for the calculation of strain tensor. > Best, > Aleks > > > On Wed, 13 Jul 2011, nek5000-users at lists.mcs.anl.gov wrote: > >> Hi, >> >> I need to calculate the vorticity vector and the rate of strain >> tensor at any >> arbitrary point in the domain. I wonder if these values are computed >> somewhere in the code or if I should interpolate them in the usr file. I >> appreciate any kind of help. >> >> Thanks in advance, >> >> >> Regards, >> Alireza Karimi >> >> _______________________________________________ >> Nek5000-users mailing list >> Nek5000-users at lists.mcs.anl.gov >> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > > ------------------------------ > > Message: 3 > Date: Wed, 13 Jul 2011 23:59:08 -0500 (CDT) > From: nek5000-users at lists.mcs.anl.gov > Subject: Re: [Nek5000-users] Vorticity > To: nek5000-users at lists.mcs.anl.gov > Message-ID: > <600285192.94128.1310619548370.JavaMail.root at zimbra.anl.gov> > Content-Type: text/plain; charset=utf-8 > > Hi Alireza, > > > For a strain stress computation you can call subroutine comp_sij > > call comp_sij(sij,nij,vx,vy,vz > $ ,ur,us,ut,vr,vs,vt,wr,ws,wt) > > > (navier5.f:2225) with nij=6 for 3D somewhere from userchk once you > declare the stress component array sij and arrays with local > coordinate r-s-t derivatives of velocity components > > common /csij/ > $ sij(lx1*ly1*lz1,6,lelv) ! 6 components > $ ,ur (lx1*ly1*lz1) > ... > $ ,wt (lx1*ly1*lz1) > > > Best, > Aleks > > > > > > > ----- Original Message ----- > From: nek5000-users at lists.mcs.anl.gov > To: nek5000-users at lists.mcs.anl.gov > Sent: Wednesday, July 13, 2011 5:09:48 PM > Subject: Re: [Nek5000-users] Vorticity > > Hi Alireza, > > For vorticity computation, you can call > > call comp_vort3(vort,work1,work2,vx,vy,vz) > > from userchk with the appropriate definitions of vorticity and work arrays > like > > common /scren/ vort (lx1,ly1,lz1,lelv,3) ! x y z components > $ , work1(lx1,ly1,lz1,lelv) > $ , work2(lx1,ly1,lz1,lelv) > > Let me check what we have for the calculation of strain tensor. > Best, > Aleks > > > On Wed, 13 Jul 2011, nek5000-users at lists.mcs.anl.gov wrote: > >> Hi, >> >> I need to calculate the vorticity vector and the rate of strain >> tensor at any >> arbitrary point in the domain. I wonder if these values are computed >> somewhere in the code or if I should interpolate them in the usr file. I >> appreciate any kind of help. >> >> Thanks in advance, >> >> >> Regards, >> Alireza Karimi >> >> _______________________________________________ >> 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 > > > ------------------------------ > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > > End of Nek5000-users Digest, Vol 29, Issue 2 > ******************************************** > -- Alireza Karimi PhD Candidate Department of Engineering Science and Mechanics Virginia Polytechnic Institute and State University Blacksburg, VA 24061 _______________________________________________ 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 Jul 14 18:00:29 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 14 Jul 2011 18:00:29 -0500 (CDT) Subject: [Nek5000-users] Vorticity In-Reply-To: <20110714173641.82945dfuhqfm4aqx@webmail.vt.edu> References: <20110714173641.82945dfuhqfm4aqx@webmail.vt.edu> Message-ID: Hi Alireza, I'm not certain of the context of your question, but you are correct that you will have to repack sij() from a (lx1*ly1*lz1,6,lelt) layout to a set of 6 vectors of the form: (lx1*ly1*lz1,lelt) This will be best done with a loop of the form: nxyz=nx1*ny1*nz1 do e=1,nelv do i=1,n v1(i,1,1,e) = sij(i,1,1,1,e) v2(i,1,1,e) = sij(i,1,1,2,e) v3(i,1,1,e) = sij(i,1,1,3,e) v4(i,1,1,e) = sij(i,1,1,4,e) v5(i,1,1,e) = sij(i,1,1,5,e) v6(i,1,1,e) = sij(i,1,1,6,e) enddo enddo then call interp() with (v1,v2,v3...) and (v4,v5,v6). You should also check the sij code to understand which parts of the tensor are stored where and whether the computed sij includes the 0.5 factor. (I think in this case that it does not.) hth, Paul On Thu, 14 Jul 2011, nek5000-users at lists.mcs.anl.gov wrote: > Hi, > > Thanks for the code. I just have a problem regarding s_ij. Should I use > opcopy subroutine twice as follows? > > call opcopy(wrk(1,1),wrk(1,2),wrk(1,3),v1,v2,v3) > call opcopy(wrk(1,4),wrk(1,5),wrk(1,6),v4,v5,v6) > > Or I should use another subroutine in this case in order to pack working > array? > > Regards, > Alireza > > > > > > Quoting nek5000-users-request at lists.mcs.anl.gov: > >> Send Nek5000-users mailing list submissions to >> nek5000-users at 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 at lists.mcs.anl.gov >> >> You can reach the person managing the list at >> nek5000-users-owner at 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. Vorticity (nek5000-users at lists.mcs.anl.gov) >> 2. Re: Vorticity (nek5000-users at lists.mcs.anl.gov) >> 3. Re: Vorticity (nek5000-users at lists.mcs.anl.gov) >> >> >> ---------------------------------------------------------------------- >> >> Message: 1 >> Date: Wed, 13 Jul 2011 17:55:35 -0400 >> From: nek5000-users at lists.mcs.anl.gov >> Subject: [Nek5000-users] Vorticity >> To: nek5000-users at lists.mcs.anl.gov >> Message-ID: <20110713175535.1042430oct81vz2v at webmail.vt.edu> >> Content-Type: text/plain; charset=ISO-8859-1; DelSp="Yes"; >> format="flowed" >> >> Hi, >> >> I need to calculate the vorticity vector and the rate of strain tensor >> at any arbitrary point in the domain. I wonder if these values are >> computed somewhere in the code or if I should interpolate them in the >> usr file. I appreciate any kind of help. >> >> Thanks in advance, >> >> >> Regards, >> Alireza Karimi >> >> >> >> ------------------------------ >> >> Message: 2 >> Date: Wed, 13 Jul 2011 17:09:48 -0500 (CDT) >> From: nek5000-users at lists.mcs.anl.gov >> Subject: Re: [Nek5000-users] Vorticity >> To: nek5000-users at lists.mcs.anl.gov >> Message-ID: >> Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed >> >> Hi Alireza, >> >> For vorticity computation, you can call >> >> call comp_vort3(vort,work1,work2,vx,vy,vz) >> >> from userchk with the appropriate definitions of vorticity and work arrays >> like >> >> common /scren/ vort (lx1,ly1,lz1,lelv,3) ! x y z components >> $ , work1(lx1,ly1,lz1,lelv) >> $ , work2(lx1,ly1,lz1,lelv) >> >> Let me check what we have for the calculation of strain tensor. >> Best, >> Aleks >> >> >> On Wed, 13 Jul 2011, nek5000-users at lists.mcs.anl.gov wrote: >> >>> Hi, >>> >>> I need to calculate the vorticity vector and the rate of strain tensor at >>> any >>> arbitrary point in the domain. I wonder if these values are computed >>> somewhere in the code or if I should interpolate them in the usr file. I >>> appreciate any kind of help. >>> >>> Thanks in advance, >>> >>> >>> Regards, >>> Alireza Karimi >>> >>> _______________________________________________ >>> Nek5000-users mailing list >>> Nek5000-users at lists.mcs.anl.gov >>> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users >> >> >> ------------------------------ >> >> Message: 3 >> Date: Wed, 13 Jul 2011 23:59:08 -0500 (CDT) >> From: nek5000-users at lists.mcs.anl.gov >> Subject: Re: [Nek5000-users] Vorticity >> To: nek5000-users at lists.mcs.anl.gov >> Message-ID: >> <600285192.94128.1310619548370.JavaMail.root at zimbra.anl.gov> >> Content-Type: text/plain; charset=utf-8 >> >> Hi Alireza, >> >> >> For a strain stress computation you can call subroutine comp_sij >> >> call comp_sij(sij,nij,vx,vy,vz >> $ ,ur,us,ut,vr,vs,vt,wr,ws,wt) >> >> >> (navier5.f:2225) with nij=6 for 3D somewhere from userchk once you declare >> the stress component array sij and arrays with local coordinate r-s-t >> derivatives of velocity components >> >> common /csij/ >> $ sij(lx1*ly1*lz1,6,lelv) ! 6 components >> $ ,ur (lx1*ly1*lz1) >> ... >> $ ,wt (lx1*ly1*lz1) >> >> >> Best, >> Aleks >> >> >> >> >> >> >> ----- Original Message ----- >> From: nek5000-users at lists.mcs.anl.gov >> To: nek5000-users at lists.mcs.anl.gov >> Sent: Wednesday, July 13, 2011 5:09:48 PM >> Subject: Re: [Nek5000-users] Vorticity >> >> Hi Alireza, >> >> For vorticity computation, you can call >> >> call comp_vort3(vort,work1,work2,vx,vy,vz) >> >> from userchk with the appropriate definitions of vorticity and work arrays >> like >> >> common /scren/ vort (lx1,ly1,lz1,lelv,3) ! x y z components >> $ , work1(lx1,ly1,lz1,lelv) >> $ , work2(lx1,ly1,lz1,lelv) >> >> Let me check what we have for the calculation of strain tensor. >> Best, >> Aleks >> >> >> On Wed, 13 Jul 2011, nek5000-users at lists.mcs.anl.gov wrote: >> >>> Hi, >>> >>> I need to calculate the vorticity vector and the rate of strain tensor at >>> any >>> arbitrary point in the domain. I wonder if these values are computed >>> somewhere in the code or if I should interpolate them in the usr file. I >>> appreciate any kind of help. >>> >>> Thanks in advance, >>> >>> >>> Regards, >>> Alireza Karimi >>> >>> _______________________________________________ >>> 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 >> >> >> ------------------------------ >> >> _______________________________________________ >> Nek5000-users mailing list >> Nek5000-users at lists.mcs.anl.gov >> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users >> >> >> End of Nek5000-users Digest, Vol 29, Issue 2 >> ******************************************** >> > > > > -- > Alireza Karimi > PhD Candidate > Department of Engineering Science and Mechanics > Virginia Polytechnic Institute and State University > Blacksburg, VA 24061 > _______________________________________________ > 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 Fri Jul 15 15:05:39 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 15 Jul 2011 15:05:39 -0500 Subject: [Nek5000-users] MPI strange error Message-ID: <4E209D93.5040105@exosent.com> Howdy Neks! Anybody seen this kind of failure before? MPI_Recv(187).......................: MPI_Recv(buf=0x7fffa1440cf8, count=4, MPI_BYTE, src=MPI_ANY_SOURCE, tag=1, comm=0x84000004, status=0x7fffa1440640) failed MPIDI_CH3U_Request_unpack_uebuf(600): Message truncated; 32 bytes received but buffer size is 4 I've run many other cases on the exact same computer with no issue. Ran the bad case on a different computer, and got the same error. I assume the issue is specific to that case, but don't quite understand how. Any ideas? Thanks, Andrew From nek5000-users at lists.mcs.anl.gov Sat Jul 16 02:23:56 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Sat, 16 Jul 2011 09:23:56 +0200 Subject: [Nek5000-users] MPI strange error In-Reply-To: <4E209D93.5040105@exosent.com> References: <4E209D93.5040105@exosent.com> Message-ID: Hi Andrew Please post your .usr file. -Stefan On 7/15/11, nek5000-users at lists.mcs.anl.gov wrote: > Howdy Neks! > > Anybody seen this kind of failure before? > > MPI_Recv(187).......................: MPI_Recv(buf=0x7fffa1440cf8, > count=4, MPI_BYTE, src=MPI_ANY_SOURCE, tag=1, comm=0x84000004, > status=0x7fffa1440640) failed > MPIDI_CH3U_Request_unpack_uebuf(600): Message truncated; 32 bytes > received but buffer size is 4 > > I've run many other cases on the exact same computer with no issue. Ran > the bad case on a different computer, and got the same error. I assume > the issue is specific to that case, but don't quite understand how. Any > ideas? > > Thanks, > Andrew > > _______________________________________________ > 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 Sat Jul 16 03:07:23 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Sat, 16 Jul 2011 13:37:23 +0530 Subject: [Nek5000-users] 3D MHD example Message-ID: Dear Nek developers, Kindly provide an update on 3D MHD example for Nek5000? I read in a year old thread that Aleks was working on that. Eagerly waiting for the example. Cheers, Rakesh -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Sat Jul 16 09:17:14 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Sat, 16 Jul 2011 09:17:14 -0500 (CDT) Subject: [Nek5000-users] 3D MHD example In-Reply-To: Message-ID: <345252173.100684.1310825834876.JavaMail.root@zimbra.anl.gov> Hi Rakesh, For the 3D MHD example please look for the case nek5_svn/examples/mhd/gpf.* and check the comments in gpf.usr for the details. An upgrade of the code which handles magnetic boundary conditions that are different from velocity BCs has been successfully tested for a magnetoconvection case and I will try to finish the corresponding example with the write-up in one month. In the meantime for your setup you can use the boundary condition string 'ddn' in .box or in .rea file with mesh (no .re2 i.e. NDIM>0 in .rea) for the case of the idealized isolator in Bz field: Bx = By = 0 = d Bz / dz Please, test this for your setup first and let me know if you encounter any problems. Best, Aleks ----- Original Message ----- From: nek5000-users at lists.mcs.anl.gov To: nek5000-users at lists.mcs.anl.gov Sent: Saturday, July 16, 2011 3:07:23 AM Subject: [Nek5000-users] 3D MHD example Dear Nek developers, Kindly provide an update on 3D MHD example for Nek5000? I read in a year old thread that Aleks was working on that. Eagerly waiting for the example. Cheers, Rakesh _______________________________________________ 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 Sat Jul 16 09:43:38 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Sat, 16 Jul 2011 20:13:38 +0530 Subject: [Nek5000-users] 3D MHD example In-Reply-To: <345252173.100684.1310825834876.JavaMail.root@zimbra.anl.gov> References: <345252173.100684.1310825834876.JavaMail.root@zimbra.anl.gov> Message-ID: Thank you very much, Alex, for such a fast reply! I will certainly try what you suggested. I'll keep you posted. Best, Rakesh On Sat, Jul 16, 2011 at 7:47 PM, wrote: > Hi Rakesh, > > > For the 3D MHD example please look for the case > > nek5_svn/examples/mhd/gpf.* > > and check the comments in gpf.usr for the details. > > > > An upgrade of the code which handles magnetic boundary conditions that are > different from velocity BCs has been successfully tested for a > magnetoconvection case and I will try to finish the corresponding example > with the write-up in one month. > > > In the meantime for your setup you can use the boundary condition string > > 'ddn' > > in .box or in .rea file with mesh (no .re2 i.e. NDIM>0 in .rea) for the > case of the idealized isolator in Bz field: > > Bx = By = 0 = d Bz / dz > > Please, test this for your setup first and let me know if you encounter any > problems. > Best, > Aleks > > > > > > > ----- Original Message ----- > From: nek5000-users at lists.mcs.anl.gov > To: nek5000-users at lists.mcs.anl.gov > Sent: Saturday, July 16, 2011 3:07:23 AM > Subject: [Nek5000-users] 3D MHD example > > > Dear Nek developers, > > > Kindly provide an update on 3D MHD example for Nek5000? I read in a year > old thread that Aleks was working on that. > > > Eagerly waiting for the example. > > > Cheers, > Rakesh > _______________________________________________ > 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 > -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Wed Jul 27 17:31:19 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 27 Jul 2011 16:31:19 -0600 Subject: [Nek5000-users] Averaging in the y-z plane Message-ID: Hi Neks, I am trying to write a routine that averages over the y-z plane, yielding an array that varies in the x dimension. I have seen the planar_average_z routine in navier5.f that averages over x-y, and I have a planar_average_y routine that Paul sent me. So, I am trying to write planar_average_x. I have been able to proceed by analogy, except for the two lines in which the weighting coefficients are computed. I planar_average_z, yy = (1.-zgm1(j,2))/2. ! = 1 for j=1, = 0 for j=ny1 aa = yy*area(i,1,1,e) + (1-yy)*area(i,1,3,e) ! wgtd jacobian, fc 1&3 In planar_aveage_y, zz = (1.-zgm1(k,3))/2. ! = 1 for k=1, = 0 for k=nz1 aa = zz*area(i,1,5,e) + (1-zz)*area(i,1,6,e) ! wgtd jacobian Can someone tell me what the analogous lines would be for planar_average_x, as well as what the area tensor is? Thanks very much, David -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Wed Jul 27 18:38:29 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 27 Jul 2011 18:38:29 -0500 (CDT) Subject: [Nek5000-users] Averaging in the y-z plane In-Reply-To: References: Message-ID: Hi David, Without digging into the context (for verification), symmetry would lead me to write: xx = (1.-zgm1(k,1))/2. ! = 1 for k=1, = 0 for k=nx1 aa = xx*area(i,1,2,e) + (1-xx)*area(i,1,4,e) ! wgtd jacobian I'm not certain of the "k" (one would normally use "i"... but it is being used for the area() counter). Does this get you started? I can try to look in more detail if you need. Best regards, Paul On Wed, 27 Jul 2011, nek5000-users at lists.mcs.anl.gov wrote: > Hi Neks, > > I am trying to write a routine that averages over the y-z plane, yielding an > array that varies in the x dimension. I have seen the planar_average_z > routine in navier5.f that averages over x-y, and I have a planar_average_y > routine that Paul sent me. So, I am trying to write planar_average_x. I > have been able to proceed by analogy, except for the two lines in which the > weighting coefficients are computed. > > I planar_average_z, > > yy = (1.-zgm1(j,2))/2. ! = 1 for j=1, = 0 for j=ny1 > aa = yy*area(i,1,1,e) + (1-yy)*area(i,1,3,e) ! wgtd jacobian, fc 1&3 > > In planar_aveage_y, > > zz = (1.-zgm1(k,3))/2. ! = 1 for k=1, = 0 for k=nz1 > aa = zz*area(i,1,5,e) + (1-zz)*area(i,1,6,e) ! wgtd jacobian > > Can someone tell me what the analogous lines would be for planar_average_x, > as well as what the area tensor is? > > Thanks very much, > > David > From nek5000-users at lists.mcs.anl.gov Thu Jul 28 15:57:28 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 28 Jul 2011 14:57:28 -0600 Subject: [Nek5000-users] Primer for reacting flows In-Reply-To: References: Message-ID: <3F8B179C-815A-4AEE-BD19-7C05B90AFE8A@nrel.gov> All, I'm interested in using Nek for simulation of reacting flows. Can anyone send me, or point me towards, a simple example that could get me going? I guess I'm most interested to see the details of storage/operation with passive scalars. Thanks. --Mike From nek5000-users at lists.mcs.anl.gov Thu Jul 28 16:33:54 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 28 Jul 2011 15:33:54 -0600 Subject: [Nek5000-users] Primer for reacting flows In-Reply-To: <3F8B179C-815A-4AEE-BD19-7C05B90AFE8A@nrel.gov> References: <3F8B179C-815A-4AEE-BD19-7C05B90AFE8A@nrel.gov> Message-ID: Hi Mike, Here are the .rea and .usr files of a case I did with 2 reacting scalars. The files are modified versions of the rayleigh case from repo that simulates the Boussinesq equations, but I turned off the buoyancy term and added a second scalar with reaction. In the userq routine in the .usr file, the qvol term specifies scalar production due to reaction. The first scalar is called temp. The next ones are ps(1), ps(2), etc. The conditional statement (ifield.eq.2) specifies production of temp, the (ifield.eq.3) statement specifies production of ps(1). The README of the shear4 example from repo, which has 4 passive scalars, describes how you can change the parameters relating to the multiple scalars. hth, David On Thu, Jul 28, 2011 at 2:57 PM, wrote: > All, > > I'm interested in using Nek for simulation of reacting flows. Can anyone > send me, or point me towards, a simple example that could get me going? I > guess I'm most interested to see the details of storage/operation with > passive scalars. > > Thanks. > --Mike > > _______________________________________________ > 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: -------------- next part -------------- A non-text attachment was scrubbed... Name: dd.rea Type: application/octet-stream Size: 4507 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: dd.usr Type: application/octet-stream Size: 17269 bytes Desc: not available URL: From nek5000-users at lists.mcs.anl.gov Fri Jul 29 09:42:57 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 29 Jul 2011 16:42:57 +0200 Subject: [Nek5000-users] Primer for reacting flows In-Reply-To: References: <3F8B179C-815A-4AEE-BD19-7C05B90AFE8A@nrel.gov> Message-ID: Hi Mike, The base version of Nek is not ready to simulate chemically reacting flows. What are the fits and gaps? Fits: - Low Mach number formulation to deal with density varying flows - Accurate splitting formulation to segregate the hydrodynamic sub-system (momentum+continuity eqn) and the thermochemical sub-system (energy and species eqn) - Interface to CVODE (a high order integrator to solve the stiff thermochemical sub system efficiently) Gaps: - Routine to compute the RHS (needed input for CVODE) - The routine to compute the thermal divergence does not include the species eqns - Routines to compute the thermophysical and kinetic properties I developed a plug-in to do reactive DNS using Nek when I was a PhD student at ETH Zurich. It performed really well and we did some pretty impressive simulations at the time (http://nek5000.mcs.anl.gov/index.php/Visualization_Gallery). I am afraid the plug-in is not available for the public :( Cheers, Stefan On 7/28/11, nek5000-users at lists.mcs.anl.gov wrote: > Hi Mike, > > Here are the .rea and .usr files of a case I did with 2 reacting scalars. > The files are modified versions of the rayleigh case from repo that > simulates the Boussinesq equations, but I turned off the buoyancy term and > added a second scalar with reaction. > > In the userq routine in the .usr file, the qvol term specifies scalar > production due to reaction. The first scalar is called temp. The next ones > are ps(1), ps(2), etc. The conditional statement (ifield.eq.2) specifies > production of temp, the (ifield.eq.3) statement specifies production of > ps(1). > > The README of the shear4 example from repo, which has 4 passive scalars, > describes how you can change the parameters relating to the multiple > scalars. > > hth, > > David > > > On Thu, Jul 28, 2011 at 2:57 PM, wrote: > >> All, >> >> I'm interested in using Nek for simulation of reacting flows. Can anyone >> send me, or point me towards, a simple example that could get me going? I >> guess I'm most interested to see the details of storage/operation with >> passive scalars. >> >> 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 Fri Jul 29 09:53:04 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 29 Jul 2011 09:53:04 -0500 Subject: [Nek5000-users] Primer for reacting flows In-Reply-To: References: <3F8B179C-815A-4AEE-BD19-7C05B90AFE8A@nrel.gov> Message-ID: Hi Stefan, Reactive flows are something my group is also wanting to look at possibly in the near future. I know you said the plug-in is not available, but was a paper produced from the work? I thought it might be interesting (and informative) to read the paper if it exists. Thanks, Josh On Fri, Jul 29, 2011 at 9:42 AM, wrote: > Hi Mike, > > The base version of Nek is not ready to simulate chemically reacting > flows. What are the fits and gaps? > > Fits: > - Low Mach number formulation to deal with density varying flows > - Accurate splitting formulation to segregate the hydrodynamic > sub-system (momentum+continuity eqn) and the thermochemical sub-system > (energy and species eqn) > - Interface to CVODE (a high order integrator to solve the stiff > thermochemical sub system efficiently) > > Gaps: > - Routine to compute the RHS (needed input for CVODE) > - The routine to compute the thermal divergence does not include the > species eqns > - Routines to compute the thermophysical and kinetic properties > > I developed a plug-in to do reactive DNS using Nek ?when I was a PhD > student at ETH Zurich. It performed really well and we did some pretty > impressive simulations at the time > (http://nek5000.mcs.anl.gov/index.php/Visualization_Gallery). I am > afraid the plug-in is not available for the public :( > > Cheers, > Stefan > > On 7/28/11, nek5000-users at lists.mcs.anl.gov > wrote: >> Hi Mike, >> >> Here are the .rea and .usr files of a case I did with 2 reacting scalars. >> The files are modified versions of the rayleigh case from repo that >> simulates the Boussinesq equations, but I turned off the buoyancy term and >> added a second scalar with reaction. >> >> In the userq routine in the .usr file, the qvol term specifies scalar >> production due to reaction. ?The first scalar is called temp. ?The next ones >> are ps(1), ps(2), etc. ?The conditional statement (ifield.eq.2) specifies >> production of temp, the (ifield.eq.3) statement specifies production of >> ps(1). >> >> The README of the shear4 example from repo, which has 4 passive scalars, >> describes how you can change the parameters relating to the multiple >> scalars. >> >> hth, >> >> David >> >> >> On Thu, Jul 28, 2011 at 2:57 PM, wrote: >> >>> All, >>> >>> I'm interested in using Nek for simulation of reacting flows. ?Can anyone >>> send me, or point me towards, a simple example that could get me going? ?I >>> guess I'm most interested to see the details of storage/operation with >>> passive scalars. >>> >>> Thanks. >>> --Mike >>> >>> _______________________________________________ >>> 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 > -- Josh Camp "All that is necessary for the triumph of evil is that good men do nothing" -- Edmund Burke From nek5000-users at lists.mcs.anl.gov Fri Jul 29 10:55:59 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 29 Jul 2011 17:55:59 +0200 Subject: [Nek5000-users] Primer for reacting flows In-Reply-To: References: <3F8B179C-815A-4AEE-BD19-7C05B90AFE8A@nrel.gov> Message-ID: My thesis is probably a good point to start :) http://e-collection.library.ethz.ch/view/eth:1899 -Stefan On 7/29/11, nek5000-users at lists.mcs.anl.gov wrote: > Hi Stefan, > > Reactive flows are something my group is also wanting to look at > possibly in the near future. I know you said the plug-in is not > available, but was a paper produced from the work? I thought it might > be interesting (and informative) to read the paper if it exists. > > Thanks, > > Josh > > On Fri, Jul 29, 2011 at 9:42 AM, wrote: >> Hi Mike, >> >> The base version of Nek is not ready to simulate chemically reacting >> flows. What are the fits and gaps? >> >> Fits: >> - Low Mach number formulation to deal with density varying flows >> - Accurate splitting formulation to segregate the hydrodynamic >> sub-system (momentum+continuity eqn) and the thermochemical sub-system >> (energy and species eqn) >> - Interface to CVODE (a high order integrator to solve the stiff >> thermochemical sub system efficiently) >> >> Gaps: >> - Routine to compute the RHS (needed input for CVODE) >> - The routine to compute the thermal divergence does not include the >> species eqns >> - Routines to compute the thermophysical and kinetic properties >> >> I developed a plug-in to do reactive DNS using Nek ?when I was a PhD >> student at ETH Zurich. It performed really well and we did some pretty >> impressive simulations at the time >> (http://nek5000.mcs.anl.gov/index.php/Visualization_Gallery). I am >> afraid the plug-in is not available for the public :( >> >> Cheers, >> Stefan >> >> On 7/28/11, nek5000-users at lists.mcs.anl.gov >> wrote: >>> Hi Mike, >>> >>> Here are the .rea and .usr files of a case I did with 2 reacting scalars. >>> The files are modified versions of the rayleigh case from repo that >>> simulates the Boussinesq equations, but I turned off the buoyancy term >>> and >>> added a second scalar with reaction. >>> >>> In the userq routine in the .usr file, the qvol term specifies scalar >>> production due to reaction. ?The first scalar is called temp. ?The next >>> ones >>> are ps(1), ps(2), etc. ?The conditional statement (ifield.eq.2) specifies >>> production of temp, the (ifield.eq.3) statement specifies production of >>> ps(1). >>> >>> The README of the shear4 example from repo, which has 4 passive scalars, >>> describes how you can change the parameters relating to the multiple >>> scalars. >>> >>> hth, >>> >>> David >>> >>> >>> On Thu, Jul 28, 2011 at 2:57 PM, wrote: >>> >>>> All, >>>> >>>> I'm interested in using Nek for simulation of reacting flows. ?Can >>>> anyone >>>> send me, or point me towards, a simple example that could get me going? >>>> ?I >>>> guess I'm most interested to see the details of storage/operation with >>>> passive scalars. >>>> >>>> Thanks. >>>> --Mike >>>> >>>> _______________________________________________ >>>> 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 >> > > > > -- > Josh Camp > > "All that is necessary for the triumph of evil is that good men do > nothing" -- Edmund Burke > _______________________________________________ > 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 Fri Jul 29 11:42:29 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 29 Jul 2011 10:42:29 -0600 Subject: [Nek5000-users] Primer for reacting flows In-Reply-To: References: <3F8B179C-815A-4AEE-BD19-7C05B90AFE8A@nrel.gov> Message-ID: <2AAB56B8-CCB7-41F7-B30C-AD573A96EAEC@nrel.gov> Stefan, Josh, Dave, thanks for the responses. Dave -- pointing me to the shear4 example and sending the files was helpful. Thanks! Stefan, I might be able to pull of what I need to do now with the current-version passive-scalar features. That said, coupling nek with CVODE is quite appealing. Are there any plans for your reactive-flow plug-in to be available publicly? --Mike On Jul 29, 2011, at 9:55 AM, wrote: > My thesis is probably a good point to start :) > http://e-collection.library.ethz.ch/view/eth:1899 > > -Stefan > > On 7/29/11, nek5000-users at lists.mcs.anl.gov > wrote: >> Hi Stefan, >> >> Reactive flows are something my group is also wanting to look at >> possibly in the near future. I know you said the plug-in is not >> available, but was a paper produced from the work? I thought it might >> be interesting (and informative) to read the paper if it exists. >> >> Thanks, >> >> Josh >> >> On Fri, Jul 29, 2011 at 9:42 AM, wrote: >>> Hi Mike, >>> >>> The base version of Nek is not ready to simulate chemically reacting >>> flows. What are the fits and gaps? >>> >>> Fits: >>> - Low Mach number formulation to deal with density varying flows >>> - Accurate splitting formulation to segregate the hydrodynamic >>> sub-system (momentum+continuity eqn) and the thermochemical sub-system >>> (energy and species eqn) >>> - Interface to CVODE (a high order integrator to solve the stiff >>> thermochemical sub system efficiently) >>> >>> Gaps: >>> - Routine to compute the RHS (needed input for CVODE) >>> - The routine to compute the thermal divergence does not include the >>> species eqns >>> - Routines to compute the thermophysical and kinetic properties >>> >>> I developed a plug-in to do reactive DNS using Nek when I was a PhD >>> student at ETH Zurich. It performed really well and we did some pretty >>> impressive simulations at the time >>> (http://nek5000.mcs.anl.gov/index.php/Visualization_Gallery). I am >>> afraid the plug-in is not available for the public :( >>> >>> Cheers, >>> Stefan >>> >>> On 7/28/11, nek5000-users at lists.mcs.anl.gov >>> wrote: >>>> Hi Mike, >>>> >>>> Here are the .rea and .usr files of a case I did with 2 reacting scalars. >>>> The files are modified versions of the rayleigh case from repo that >>>> simulates the Boussinesq equations, but I turned off the buoyancy term >>>> and >>>> added a second scalar with reaction. >>>> >>>> In the userq routine in the .usr file, the qvol term specifies scalar >>>> production due to reaction. The first scalar is called temp. The next >>>> ones >>>> are ps(1), ps(2), etc. The conditional statement (ifield.eq.2) specifies >>>> production of temp, the (ifield.eq.3) statement specifies production of >>>> ps(1). >>>> >>>> The README of the shear4 example from repo, which has 4 passive scalars, >>>> describes how you can change the parameters relating to the multiple >>>> scalars. >>>> >>>> hth, >>>> >>>> David >>>> >>>> >>>> On Thu, Jul 28, 2011 at 2:57 PM, wrote: >>>> >>>>> All, >>>>> >>>>> I'm interested in using Nek for simulation of reacting flows. Can >>>>> anyone >>>>> send me, or point me towards, a simple example that could get me going? >>>>> I >>>>> guess I'm most interested to see the details of storage/operation with >>>>> passive scalars. >>>>> >>>>> Thanks. >>>>> --Mike >>>>> >>>>> _______________________________________________ >>>>> 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 >>> >> >> >> >> -- >> Josh Camp >> >> "All that is necessary for the triumph of evil is that good men do >> nothing" -- Edmund Burke >> _______________________________________________ >> 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 From nek5000-users at lists.mcs.anl.gov Fri Jul 29 12:08:45 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 29 Jul 2011 19:08:45 +0200 Subject: [Nek5000-users] Primer for reacting flows In-Reply-To: <2AAB56B8-CCB7-41F7-B30C-AD573A96EAEC@nrel.gov> References: <3F8B179C-815A-4AEE-BD19-7C05B90AFE8A@nrel.gov> <2AAB56B8-CCB7-41F7-B30C-AD573A96EAEC@nrel.gov> Message-ID: Mike, I don't know. Contact me off list to discuss the details. -Stefan On 7/29/11, nek5000-users at lists.mcs.anl.gov wrote: > Stefan, Josh, Dave, thanks for the responses. > > Dave -- pointing me to the shear4 example and sending the files was helpful. > Thanks! > > Stefan, I might be able to pull of what I need to do now with the > current-version passive-scalar features. That said, coupling nek with CVODE > is quite appealing. Are there any plans for your reactive-flow plug-in to > be available publicly? > > --Mike > > On Jul 29, 2011, at 9:55 AM, wrote: > >> My thesis is probably a good point to start :) >> http://e-collection.library.ethz.ch/view/eth:1899 >> >> -Stefan >> >> On 7/29/11, nek5000-users at lists.mcs.anl.gov >> wrote: >>> Hi Stefan, >>> >>> Reactive flows are something my group is also wanting to look at >>> possibly in the near future. I know you said the plug-in is not >>> available, but was a paper produced from the work? I thought it might >>> be interesting (and informative) to read the paper if it exists. >>> >>> Thanks, >>> >>> Josh >>> >>> On Fri, Jul 29, 2011 at 9:42 AM, >>> wrote: >>>> Hi Mike, >>>> >>>> The base version of Nek is not ready to simulate chemically reacting >>>> flows. What are the fits and gaps? >>>> >>>> Fits: >>>> - Low Mach number formulation to deal with density varying flows >>>> - Accurate splitting formulation to segregate the hydrodynamic >>>> sub-system (momentum+continuity eqn) and the thermochemical sub-system >>>> (energy and species eqn) >>>> - Interface to CVODE (a high order integrator to solve the stiff >>>> thermochemical sub system efficiently) >>>> >>>> Gaps: >>>> - Routine to compute the RHS (needed input for CVODE) >>>> - The routine to compute the thermal divergence does not include the >>>> species eqns >>>> - Routines to compute the thermophysical and kinetic properties >>>> >>>> I developed a plug-in to do reactive DNS using Nek when I was a PhD >>>> student at ETH Zurich. It performed really well and we did some pretty >>>> impressive simulations at the time >>>> (http://nek5000.mcs.anl.gov/index.php/Visualization_Gallery). I am >>>> afraid the plug-in is not available for the public :( >>>> >>>> Cheers, >>>> Stefan >>>> >>>> On 7/28/11, nek5000-users at lists.mcs.anl.gov >>>> wrote: >>>>> Hi Mike, >>>>> >>>>> Here are the .rea and .usr files of a case I did with 2 reacting >>>>> scalars. >>>>> The files are modified versions of the rayleigh case from repo that >>>>> simulates the Boussinesq equations, but I turned off the buoyancy term >>>>> and >>>>> added a second scalar with reaction. >>>>> >>>>> In the userq routine in the .usr file, the qvol term specifies scalar >>>>> production due to reaction. The first scalar is called temp. The next >>>>> ones >>>>> are ps(1), ps(2), etc. The conditional statement (ifield.eq.2) >>>>> specifies >>>>> production of temp, the (ifield.eq.3) statement specifies production of >>>>> ps(1). >>>>> >>>>> The README of the shear4 example from repo, which has 4 passive >>>>> scalars, >>>>> describes how you can change the parameters relating to the multiple >>>>> scalars. >>>>> >>>>> hth, >>>>> >>>>> David >>>>> >>>>> >>>>> On Thu, Jul 28, 2011 at 2:57 PM, >>>>> wrote: >>>>> >>>>>> All, >>>>>> >>>>>> I'm interested in using Nek for simulation of reacting flows. Can >>>>>> anyone >>>>>> send me, or point me towards, a simple example that could get me >>>>>> going? >>>>>> I >>>>>> guess I'm most interested to see the details of storage/operation with >>>>>> passive scalars. >>>>>> >>>>>> Thanks. >>>>>> --Mike >>>>>> >>>>>> _______________________________________________ >>>>>> 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 >>>> >>> >>> >>> >>> -- >>> Josh Camp >>> >>> "All that is necessary for the triumph of evil is that good men do >>> nothing" -- Edmund Burke >>> _______________________________________________ >>> 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 > > _______________________________________________ > 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 Fri Jul 29 12:43:30 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 29 Jul 2011 19:43:30 +0200 Subject: [Nek5000-users] Primer for reacting flows In-Reply-To: References: <3F8B179C-815A-4AEE-BD19-7C05B90AFE8A@nrel.gov> <2AAB56B8-CCB7-41F7-B30C-AD573A96EAEC@nrel.gov> Message-ID: Let me make a couple of comments in addition to what I said: Most of the plug-in features were moved into Nek some time ago. The infrastructure is there to simulate reactive flows but the little (but still time consuming ) details are missing. All you have to do is: Code some routines to compute thermophysical and thermodynamic properties. In addition you need a routine to compute the species production term. For detailed chemistry and transport use a package like Chemkin. In case you want to use a mixture averaged transport model code up a routine to compute the terms related to the correction velocitiy and add them to the energy and species equations. That's it. Everything else is there! I guess the difficult part is to get familiar with the internals of Nek. Except from that it's pretty straightforward and shouldn't take that long to implement. Hth, Stefan On 7/29/11, S K wrote: > Mike, I don't know. Contact me off list to discuss the details. > -Stefan > > On 7/29/11, nek5000-users at lists.mcs.anl.gov > wrote: >> Stefan, Josh, Dave, thanks for the responses. >> >> Dave -- pointing me to the shear4 example and sending the files was >> helpful. >> Thanks! >> >> Stefan, I might be able to pull of what I need to do now with the >> current-version passive-scalar features. That said, coupling nek with >> CVODE >> is quite appealing. Are there any plans for your reactive-flow plug-in >> to >> be available publicly? >> >> --Mike >> >> On Jul 29, 2011, at 9:55 AM, wrote: >> >>> My thesis is probably a good point to start :) >>> http://e-collection.library.ethz.ch/view/eth:1899 >>> >>> -Stefan >>> >>> On 7/29/11, nek5000-users at lists.mcs.anl.gov >>> wrote: >>>> Hi Stefan, >>>> >>>> Reactive flows are something my group is also wanting to look at >>>> possibly in the near future. I know you said the plug-in is not >>>> available, but was a paper produced from the work? I thought it might >>>> be interesting (and informative) to read the paper if it exists. >>>> >>>> Thanks, >>>> >>>> Josh >>>> >>>> On Fri, Jul 29, 2011 at 9:42 AM, >>>> wrote: >>>>> Hi Mike, >>>>> >>>>> The base version of Nek is not ready to simulate chemically reacting >>>>> flows. What are the fits and gaps? >>>>> >>>>> Fits: >>>>> - Low Mach number formulation to deal with density varying flows >>>>> - Accurate splitting formulation to segregate the hydrodynamic >>>>> sub-system (momentum+continuity eqn) and the thermochemical sub-system >>>>> (energy and species eqn) >>>>> - Interface to CVODE (a high order integrator to solve the stiff >>>>> thermochemical sub system efficiently) >>>>> >>>>> Gaps: >>>>> - Routine to compute the RHS (needed input for CVODE) >>>>> - The routine to compute the thermal divergence does not include the >>>>> species eqns >>>>> - Routines to compute the thermophysical and kinetic properties >>>>> >>>>> I developed a plug-in to do reactive DNS using Nek when I was a PhD >>>>> student at ETH Zurich. It performed really well and we did some pretty >>>>> impressive simulations at the time >>>>> (http://nek5000.mcs.anl.gov/index.php/Visualization_Gallery). I am >>>>> afraid the plug-in is not available for the public :( >>>>> >>>>> Cheers, >>>>> Stefan >>>>> >>>>> On 7/28/11, nek5000-users at lists.mcs.anl.gov >>>>> wrote: >>>>>> Hi Mike, >>>>>> >>>>>> Here are the .rea and .usr files of a case I did with 2 reacting >>>>>> scalars. >>>>>> The files are modified versions of the rayleigh case from repo that >>>>>> simulates the Boussinesq equations, but I turned off the buoyancy >>>>>> term >>>>>> and >>>>>> added a second scalar with reaction. >>>>>> >>>>>> In the userq routine in the .usr file, the qvol term specifies scalar >>>>>> production due to reaction. The first scalar is called temp. The >>>>>> next >>>>>> ones >>>>>> are ps(1), ps(2), etc. The conditional statement (ifield.eq.2) >>>>>> specifies >>>>>> production of temp, the (ifield.eq.3) statement specifies production >>>>>> of >>>>>> ps(1). >>>>>> >>>>>> The README of the shear4 example from repo, which has 4 passive >>>>>> scalars, >>>>>> describes how you can change the parameters relating to the multiple >>>>>> scalars. >>>>>> >>>>>> hth, >>>>>> >>>>>> David >>>>>> >>>>>> >>>>>> On Thu, Jul 28, 2011 at 2:57 PM, >>>>>> wrote: >>>>>> >>>>>>> All, >>>>>>> >>>>>>> I'm interested in using Nek for simulation of reacting flows. Can >>>>>>> anyone >>>>>>> send me, or point me towards, a simple example that could get me >>>>>>> going? >>>>>>> I >>>>>>> guess I'm most interested to see the details of storage/operation >>>>>>> with >>>>>>> passive scalars. >>>>>>> >>>>>>> Thanks. >>>>>>> --Mike >>>>>>> >>>>>>> _______________________________________________ >>>>>>> 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 >>>>> >>>> >>>> >>>> >>>> -- >>>> Josh Camp >>>> >>>> "All that is necessary for the triumph of evil is that good men do >>>> nothing" -- Edmund Burke >>>> _______________________________________________ >>>> 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 >> >> _______________________________________________ >> 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 Fri Jul 29 14:49:43 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 29 Jul 2011 14:49:43 -0500 Subject: [Nek5000-users] explstrs routine Message-ID: Hello Neks I came across the routine explstrs (and its subroutines) while exploring the code (and with the recommendation from Paul). I understand that it is essentially performing something like D^T * B * D * [u] I had a question about the use of the Jacobian in the routine expl_strs_e_3d (or _2d). Following the routine, I can see where it 1) Computes the gradients w.r.t. r, s, t (gradl_rst) 2) Multiplies these gradients by appropriate coordinate derivatives (rxm1, tzm1, etc) 3) Multiplies these results by the quadrature weights. (from w3m1) 4) Multiplies these results again by the coordinate transformations, and then apply the transpose operation (gradl_rst_t) When I look in coef.f, I see that when dr/dx, ds/dx, etc. are computed from dx/dr, dy/ds, etc., the inverse of the Jacobian is not applied yet. So, I can see why jacmi is multiplied along with w3m1 when multiplying by the quadrature weights. My question is, would we not need to multiply again by jacmi when the transpose derivative operator is applied (because of the use of rxm1, sxm1, etc)? Is this something that is applied later on, or am I missing something and it isn't necessary to apply it again? Thank you, Josh -- Josh Camp "All that is necessary for the triumph of evil is that good men do nothing" -- Edmund Burke From nek5000-users at lists.mcs.anl.gov Fri Jul 29 22:32:37 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 29 Jul 2011 22:32:37 -0500 (CDT) Subject: [Nek5000-users] explstrs routine In-Reply-To: References: Message-ID: Hi Josh, You need the jacobian at least once, since normally: B = w3m1*Jac Does this make sense? Paul On Fri, 29 Jul 2011, nek5000-users at lists.mcs.anl.gov wrote: > Hello Neks > > I came across the routine explstrs (and its subroutines) while > exploring the code (and with the recommendation from Paul). I > understand that it is essentially performing something like > > D^T * B * D * [u] > > I had a question about the use of the Jacobian in the routine > expl_strs_e_3d (or _2d). Following the routine, I can see where it > > 1) Computes the gradients w.r.t. r, s, t (gradl_rst) > > 2) Multiplies these gradients by appropriate coordinate derivatives > (rxm1, tzm1, etc) > > 3) Multiplies these results by the quadrature weights. (from w3m1) > > 4) Multiplies these results again by the coordinate transformations, > and then apply the transpose operation (gradl_rst_t) > > When I look in coef.f, I see that when dr/dx, ds/dx, etc. are computed > from dx/dr, dy/ds, etc., the inverse of the Jacobian is not applied > yet. So, I can see why jacmi is multiplied along with w3m1 when > multiplying by the quadrature weights. > > My question is, would we not need to multiply again by jacmi when the > transpose derivative operator is applied (because of the use of rxm1, > sxm1, etc)? Is this something that is applied later on, or am I > missing something and it isn't necessary to apply it again? > > Thank you, > > Josh > > -- > Josh Camp > > "All that is necessary for the triumph of evil is that good men do > nothing" -- Edmund Burke > _______________________________________________ > 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 Fri Jul 29 23:04:54 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 29 Jul 2011 23:04:54 -0500 Subject: [Nek5000-users] explstrs routine In-Reply-To: References: Message-ID: Paul, Ah, yes, I see it now. Basically, you WOULD need two 1/J 's in the computation, but yes, I was forgetting that w3m1 is just the weights without the Jacobian added in. So J will cancel out one of the 1/J's. Thanks! On Fri, Jul 29, 2011 at 10:32 PM, wrote: > > Hi Josh, > > You need the jacobian at least once, since normally: > > ? B = w3m1*Jac > > Does this make sense? > > Paul > > > On Fri, 29 Jul 2011, nek5000-users at lists.mcs.anl.gov wrote: > >> Hello Neks >> >> I came across the routine explstrs (and its subroutines) while >> exploring the code (and with the recommendation from Paul). ?I >> understand that it is essentially performing something like >> >> ?D^T * B * D * [u] >> >> I had a question about the use of the Jacobian in the routine >> expl_strs_e_3d (or _2d). ?Following the routine, I can see where it >> >> 1) Computes the gradients w.r.t. r, s, t ? (gradl_rst) >> >> 2) Multiplies these gradients by appropriate coordinate derivatives >> (rxm1, tzm1, etc) >> >> 3) Multiplies these results by the quadrature weights. (from w3m1) >> >> 4) Multiplies these results again by the coordinate transformations, >> and then apply the transpose operation (gradl_rst_t) >> >> When I look in coef.f, I see that when dr/dx, ds/dx, etc. are computed >> from dx/dr, dy/ds, etc., the inverse of the Jacobian is not applied >> yet. ?So, I can see why jacmi is multiplied along with w3m1 when >> multiplying by the quadrature weights. >> >> My question is, would we not need to multiply again by jacmi when the >> transpose derivative operator is applied (because of the use of rxm1, >> sxm1, etc)? ?Is this something that is applied later on, or am I >> missing something and it isn't necessary to apply it again? >> >> Thank you, >> >> Josh >> >> -- >> Josh Camp >> >> "All that is necessary for the triumph of evil is that good men do >> nothing" -- Edmund Burke >> _______________________________________________ >> 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 > -- Josh Camp "All that is necessary for the triumph of evil is that good men do nothing" -- Edmund Burke From nek5000-users at lists.mcs.anl.gov Sat Jul 30 05:19:46 2011 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Sat, 30 Jul 2011 05:19:46 -0500 (CDT) Subject: [Nek5000-users] explstrs routine In-Reply-To: References: Message-ID: On Fri, 29 Jul 2011, nek5000-users at lists.mcs.anl.gov wrote: > Paul, > > Ah, yes, I see it now. Basically, you WOULD need two 1/J 's in the > computation, but yes, I was forgetting that w3m1 is just the weights > without the Jacobian added in. So J will cancel out one of the 1/J's. > > Thanks! > > On Fri, Jul 29, 2011 at 10:32 PM, wrote: >> >> Hi Josh, >> >> You need the jacobian at least once, since normally: >> >> ? B = w3m1*Jac >> >> Does this make sense? >> >> Paul >> >> >> On Fri, 29 Jul 2011, nek5000-users at lists.mcs.anl.gov wrote: >> >>> Hello Neks >>> >>> I came across the routine explstrs (and its subroutines) while >>> exploring the code (and with the recommendation from Paul). ?I >>> understand that it is essentially performing something like >>> >>> ?D^T * B * D * [u] >>> >>> I had a question about the use of the Jacobian in the routine >>> expl_strs_e_3d (or _2d). ?Following the routine, I can see where it >>> >>> 1) Computes the gradients w.r.t. r, s, t ? (gradl_rst) >>> >>> 2) Multiplies these gradients by appropriate coordinate derivatives >>> (rxm1, tzm1, etc) >>> >>> 3) Multiplies these results by the quadrature weights. (from w3m1) >>> >>> 4) Multiplies these results again by the coordinate transformations, >>> and then apply the transpose operation (gradl_rst_t) >>> >>> When I look in coef.f, I see that when dr/dx, ds/dx, etc. are computed >>> from dx/dr, dy/ds, etc., the inverse of the Jacobian is not applied >>> yet. ?So, I can see why jacmi is multiplied along with w3m1 when >>> multiplying by the quadrature weights. >>> >>> My question is, would we not need to multiply again by jacmi when the >>> transpose derivative operator is applied (because of the use of rxm1, >>> sxm1, etc)? ?Is this something that is applied later on, or am I >>> missing something and it isn't necessary to apply it again? >>> >>> Thank you, >>> >>> Josh >>> >>> -- >>> Josh Camp >>> >>> "All that is necessary for the triumph of evil is that good men do >>> nothing" -- Edmund Burke >>> _______________________________________________ >>> 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 >> > > > > -- > Josh Camp > > "All that is necessary for the triumph of evil is that good men do > nothing" -- Edmund Burke > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users >