From nek5000-users at lists.mcs.anl.gov Tue May 10 14:57:12 2016 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Tue, 10 May 2016 19:57:12 +0000 Subject: [Nek5000-users] problem in running a case Message-ID: Hello, I am mansi patel and I am downloading nek5000 I was able to compile makenek eddy_uv but after that I am not able to run a case nekb eddy_uv Do I need to stay in eddy directory? even if I change directory I am getting same error command not found . what should I do? -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Tue May 10 16:56:51 2016 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Tue, 10 May 2016 21:56:51 +0000 Subject: [Nek5000-users] problem in running a case In-Reply-To: References: Message-ID: You should copy ~/nek5_svn/trunk/tools/scripts/* ~/bin This will put nekb in your path, assuming ~/bin is in your path. Paul ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Tuesday, May 10, 2016 2:57 PM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] problem in running a case Hello, I am mansi patel and I am downloading nek5000 I was able to compile makenek eddy_uv but after that I am not able to run a case nekb eddy_uv Do I need to stay in eddy directory? even if I change directory I am getting same error command not found . what should I do? -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Tue May 10 20:54:41 2016 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 11 May 2016 01:54:41 +0000 Subject: [Nek5000-users] problem in running a case In-Reply-To: References: Message-ID: Hi Mansi, Is nekb in your path? Try: $ which nekb If it is not, copy it from tools/scripts/nekb to your ~/bin . You should run the test from within the eddy directory. Hope this helps, Max On Tue, May 10, 2016 at 3:33 PM wrote: > Hello, > I am mansi patel and I am downloading nek5000 I was able to compile *makenek > eddy_uv* > but after that I am not able to run a case *nekb eddy_uv* Do I need to > stay in eddy directory? even if I change directory I am getting *same > error command not found *. > what should I do? > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Thu May 12 13:08:39 2016 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 12 May 2016 18:08:39 +0000 Subject: [Nek5000-users] Nek5000-users Digest, Vol 87, Issue 1 In-Reply-To: References: Message-ID: Thanks a lot Paul and Max the answers were very helpful since I am new to Linux and coding. ________________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov on behalf of nek5000-users-request at lists.mcs.anl.gov Sent: Wednesday, May 11, 2016 1:00 PM To: nek5000-users at lists.mcs.anl.gov Subject: Nek5000-users Digest, Vol 87, Issue 1 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. problem in running a case (nek5000-users at lists.mcs.anl.gov) 2. Re: problem in running a case (nek5000-users at lists.mcs.anl.gov) 3. Re: problem in running a case (nek5000-users at lists.mcs.anl.gov) ---------------------------------------------------------------------- Message: 1 Date: Tue, 10 May 2016 19:57:12 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: [Nek5000-users] problem in running a case Message-ID: Content-Type: text/plain; charset="iso-8859-1" Hello, I am mansi patel and I am downloading nek5000 I was able to compile makenek eddy_uv but after that I am not able to run a case nekb eddy_uv Do I need to stay in eddy directory? even if I change directory I am getting same error command not found . what should I do? -------------- next part -------------- An HTML attachment was scrubbed... URL: ------------------------------ Message: 2 Date: Tue, 10 May 2016 21:56:51 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: Re: [Nek5000-users] problem in running a case Message-ID: Content-Type: text/plain; charset="iso-8859-1" You should copy ~/nek5_svn/trunk/tools/scripts/* ~/bin This will put nekb in your path, assuming ~/bin is in your path. Paul ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Tuesday, May 10, 2016 2:57 PM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] problem in running a case Hello, I am mansi patel and I am downloading nek5000 I was able to compile makenek eddy_uv but after that I am not able to run a case nekb eddy_uv Do I need to stay in eddy directory? even if I change directory I am getting same error command not found . what should I do? -------------- next part -------------- An HTML attachment was scrubbed... URL: ------------------------------ Message: 3 Date: Wed, 11 May 2016 01:54:41 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: Re: [Nek5000-users] problem in running a case Message-ID: Content-Type: text/plain; charset="utf-8" Hi Mansi, Is nekb in your path? Try: $ which nekb If it is not, copy it from tools/scripts/nekb to your ~/bin . You should run the test from within the eddy directory. Hope this helps, Max On Tue, May 10, 2016 at 3:33 PM wrote: > Hello, > I am mansi patel and I am downloading nek5000 I was able to compile *makenek > eddy_uv* > but after that I am not able to run a case *nekb eddy_uv* Do I need to > stay in eddy directory? even if I change directory I am getting *same > error command not found *. > what should I do? > _______________________________________________ > 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: ------------------------------ _______________________________________________ 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 87, Issue 1 ******************************************** From nek5000-users at lists.mcs.anl.gov Fri May 13 03:58:31 2016 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 13 May 2016 08:58:31 +0000 Subject: [Nek5000-users] Two-way coupling and mpi calls to userf() Message-ID: Hi Neks, I trust you are all well, I'm having some trouble with my two-way LPT code, in particular using information from the usrchk() subroutine (where all my LPTing is done) and userf() (where I want to apply a force to all meshpoints based on the number of particles in the surrounding cell). My issue is that the code doesn't count the number of particles in each cell properly, even though the xmc, xxc etc. arrays store the correct minimum and maximum values for the cell boundaries as in the usrchk() code I use these to calculate a full volume of the mesh and it is correct. I feel like my understanding on how the mpi works is lacking, so any advice on that would also be helpful - I understand that subroutine userf (ix,iy,iz,ieg) is called once per local element? So far my 2W code in the userf() looks like this: c !=== TWO WAY COUPLING ===! if(TWOWAYON .EQ. 1 .AND. istep .GT. 1) then real twowayforcex(lx1,ly1,lz1,lelv) real twowayforcey(lx1,ly1,lz1,lelv) real twowayforcez(lx1,ly1,lz1,lelv) real xmc(lx1,ly1,lz1,lelv) real xxc(lx1,ly1,lz1,lelv) real ymc(lx1,ly1,lz1,lelv) real yxc(lx1,ly1,lz1,lelv) real zmc(lx1,ly1,lz1,lelv) real zxc(lx1,ly1,lz1,lelv) real vol(lx1,ly1,lz1,lelv) real partpostw(ldim,lpart) real partftw(ldim,lpart) integer npartcell(lx1,ly1,lz1,lelv) integer TWOWAYON integer iel common /twowayr/ twowayforcex $ , twowayforcey $ , twowayforcez $ , xmc, xxc $ , ymc, yxc $ , zmc, zxc, vol $ , partpostw, partftw common /twowayi/ npartcell $ , TWOWAYON iel = gllel(ieg) if(istep .GT. 2) then do jp = 1,300000 if(partpostw(1,jp) .GE. xmc(ix,iy,iz,iel) .AND. $ partpostw(1,jp) .LE. xxc(ix,iy,iz,iel) .AND. $ partpostw(2,jp) .GE. ymc(ix,iy,iz,iel) .AND. $ partpostw(2,jp) .LE. yxc(ix,iy,iz,iel) .AND. $ partpostw(3,jp) .GE. zmc(ix,iy,iz,iel) .AND. $ partpostw(3,jp) .LE. zxc(ix,iy,iz,iel)) then npartcell(ix,iy,iz,iel) = npartcell(ix,iy,iz,iel) + 1 ffx = ffx + partftw(1,jp) ! Still need to average these ffy = ffy + partftw(2,jp) ffz = ffz + partftw(3,jp) endif enddo else ffx = 0.0 ffz = 0.0 ffy = 0.0 endif endif Thankyou for reading this, any help would be greatly appreciated! -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Thu May 19 05:27:39 2016 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 19 May 2016 10:27:39 +0000 Subject: [Nek5000-users] Temperature field values in mesh 1 only Message-ID: Hi Neks! I am running a Conjugate Heat Transfer simulation with a fluid box (velocity mesh) and two solid boxes (all boxes- temperature mesh). I am interested in computing the heat flux budget for this simulation. While doing this, on multiple occasions, I need to multiply the temperature or passive scalars throughout the fluid domain field with the velocity component values at the respective locations. I also need to calculate the gradients of these variables throughout the fluid domain. For this I use the sub-routine "gradm1" present in navier5.f . But I believe gradm1 calculates the gradients for field variables in the velocity mesh only. For all these functions, I think it's best if I could isolate an array with the values of the temperature and multiple passive scalars in the fluid domain (or velocity mesh) only, which should be a subset of the whole temperature array. Thus my question is, is there any way to get this? I am using multiple processors which complicates things. I want to create an array that contains the values of the temperature and the passive scalars at all the velocity mesh elements and this I have to do across all processors. In other words, from the standard NEK array t(lx2,ly2,lz2,lelt,ldimt) which contains the temperature and the passive scalars, I have to create an array like, say, tm1(lx1,ly1,lz1,lelv,ldimt) which would have these values. I am not sure if I have been able to explain my query clearly enough, but I would greatly appreciate if anybody could help me with this or if you could think of an easier way to calculate the gradient of the temperature and passive scalars throughout the velocity field. Let me know if you want more clarity with the question, I will try to explain better. Thanks in advance and cheers! Pradyumna -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Thu May 19 06:09:28 2016 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 19 May 2016 11:09:28 +0000 Subject: [Nek5000-users] Temperature field values in mesh 1 only In-Reply-To: References: Message-ID: Hi Padyumna, I just checked that gradm1() operates on all elements (1:nelt) If you want a field that is zero in the solid elements, here is one approach: real tm1(lx1,ly1,lz1,lelt,ldimt) [ note, use lelt here ] nv = nx1*ny1*nz1*nelv nt = nx1*ny1*nz1*nelt do i=1,ldimt call rzero(tm1(1,1,1,1,i),nt) call copy (tm1(1,1,1,1,i),t(1,1,1,1,i),nv) enddo Note that if you're trying to compute (say) grad V_x * T, you can do so via: call col2(vxt,vx,t,nt) call gradm1(vxtx,vxty,vxtz,vxt) Note that vxt will be zero in the solid elements because vx is zero there. Best, Paul ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Thursday, May 19, 2016 5:27 AM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] Temperature field values in mesh 1 only Hi Neks! I am running a Conjugate Heat Transfer simulation with a fluid box (velocity mesh) and two solid boxes (all boxes- temperature mesh). I am interested in computing the heat flux budget for this simulation. While doing this, on multiple occasions, I need to multiply the temperature or passive scalars throughout the fluid domain field with the velocity component values at the respective locations. I also need to calculate the gradients of these variables throughout the fluid domain. For this I use the sub-routine ?gradm1? present in navier5.f . But I believe gradm1 calculates the gradients for field variables in the velocity mesh only. For all these functions, I think it?s best if I could isolate an array with the values of the temperature and multiple passive scalars in the fluid domain (or velocity mesh) only, which should be a subset of the whole temperature array. Thus my question is, is there any way to get this? I am using multiple processors which complicates things. I want to create an array that contains the values of the temperature and the passive scalars at all the velocity mesh elements and this I have to do across all processors. In other words, from the standard NEK array t(lx2,ly2,lz2,lelt,ldimt) which contains the temperature and the passive scalars, I have to create an array like, say, tm1(lx1,ly1,lz1,lelv,ldimt) which would have these values. I am not sure if I have been able to explain my query clearly enough, but I would greatly appreciate if anybody could help me with this or if you could think of an easier way to calculate the gradient of the temperature and passive scalars throughout the velocity field. Let me know if you want more clarity with the question, I will try to explain better. Thanks in advance and cheers! Pradyumna -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Thu May 19 09:32:21 2016 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 19 May 2016 14:32:21 +0000 Subject: [Nek5000-users] Temperature field values in mesh 1 only In-Reply-To: References: Message-ID: Hi Paul, Thank you for your reply. Yes you are right, the gradm1 does operate on all the elements. I got mistaken by looking at the lx1, ly1 and other terms. Also I think you meant use the "call col3(vxt,vx,t,nt)" instead of the subroutine col2, since the latter takes only two array arguments. The suggestions that you gave me should help me a lot. Thanks! Regards, Pradyumna From: nek5000-users-bounces at lists.mcs.anl.gov [mailto:nek5000-users-bounces at lists.mcs.anl.gov] On Behalf Of nek5000-users at lists.mcs.anl.gov Sent: Thursday, May 19, 2016 1:12 PM To: nek5000-users at lists.mcs.anl.gov Subject: Re: [Nek5000-users] Temperature field values in mesh 1 only Hi Padyumna, I just checked that gradm1() operates on all elements (1:nelt) If you want a field that is zero in the solid elements, here is one approach: real tm1(lx1,ly1,lz1,lelt,ldimt) [ note, use lelt here ] nv = nx1*ny1*nz1*nelv nt = nx1*ny1*nz1*nelt do i=1,ldimt call rzero(tm1(1,1,1,1,i),nt) call copy (tm1(1,1,1,1,i),t(1,1,1,1,i),nv) enddo Note that if you're trying to compute (say) grad V_x * T, you can do so via: call col2(vxt,vx,t,nt) call gradm1(vxtx,vxty,vxtz,vxt) Note that vxt will be zero in the solid elements because vx is zero there. Best, Paul ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Thursday, May 19, 2016 5:27 AM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] Temperature field values in mesh 1 only Hi Neks! I am running a Conjugate Heat Transfer simulation with a fluid box (velocity mesh) and two solid boxes (all boxes- temperature mesh). I am interested in computing the heat flux budget for this simulation. While doing this, on multiple occasions, I need to multiply the temperature or passive scalars throughout the fluid domain field with the velocity component values at the respective locations. I also need to calculate the gradients of these variables throughout the fluid domain. For this I use the sub-routine "gradm1" present in navier5.f . But I believe gradm1 calculates the gradients for field variables in the velocity mesh only. For all these functions, I think it's best if I could isolate an array with the values of the temperature and multiple passive scalars in the fluid domain (or velocity mesh) only, which should be a subset of the whole temperature array. Thus my question is, is there any way to get this? I am using multiple processors which complicates things. I want to create an array that contains the values of the temperature and the passive scalars at all the velocity mesh elements and this I have to do across all processors. In other words, from the standard NEK array t(lx2,ly2,lz2,lelt,ldimt) which contains the temperature and the passive scalars, I have to create an array like, say, tm1(lx1,ly1,lz1,lelv,ldimt) which would have these values. I am not sure if I have been able to explain my query clearly enough, but I would greatly appreciate if anybody could help me with this or if you could think of an easier way to calculate the gradient of the temperature and passive scalars throughout the velocity field. Let me know if you want more clarity with the question, I will try to explain better. Thanks in advance and cheers! Pradyumna -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Mon May 23 08:22:29 2016 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Mon, 23 May 2016 15:22:29 +0200 Subject: [Nek5000-users] Spectral coefficients for velocity Message-ID: Hi Neks, Priviously, the parameters p101 and p103 have been discussed: "In each direction, on each element, the solution is represented as a polynomial of degree N in the reference element (i.e., on the interval [-1,1]). For any Nth-order polynomial there is a modal expansion of the form: u(x) = \sum_k=0^N uh_k phi_k(x) where phi_0 = 1, phi_1 = x, and phi_k(x)=L_k(x) - L_{k-1} (x) for k > 1, where L_k(x)=kth Legendre polynomial." How can I obtain these coefficients "uh_k" on the several elements? Thanks in advance for any suggestion, Andrey -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Wed May 25 10:06:16 2016 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 25 May 2016 17:06:16 +0200 Subject: [Nek5000-users] Spectral coefficients for velocity In-Reply-To: References: Message-ID: Hi Andrey, I don't know if you got an answer yet but here are my suggestions. If you are asking about the theory behind spectral transforms, I suggest that you look at appendix B in the book "High-Order Methods for Incompressible Fluid Flow" by Fischer, Deville and Mund. In particular, Section B.3.1 gives the formula for the spectral coefficients uh_k uh_k = 1/gamma_k \int_-1^1 w(x) u(x) p_k(x) dx where gamma_k=||p_k||^2 is the discrete norm and w(x) is the weight function, both of them associated to polynomial p_k. If you want to compute these coefficients in Nek5000, then you can look at the routine "local_err_est" in the file navier5.f. There, one goes to the Legendre space by using call tensr3(uh,nx,u,nx,Lj,Ljt,Ljt,w) where uh is the array of spectral elements and the operators Lj and Ljt are computed in the routine "build_legend_transform" (also in navier5.f) from the GLL points. I hope that helps. Best regards, Nicolas Offermans Quoting nek5000-users at lists.mcs.anl.gov: > Hi Neks, > > Priviously, the parameters p101 and p103 have been discussed: > > "In each direction, on each element, the solution is represented as a > polynomial of degree N in the reference element (i.e., on the interval > [-1,1]). > > For any Nth-order polynomial there is a modal expansion of the form: > > u(x) = \sum_k=0^N uh_k phi_k(x) > > where phi_0 = 1, phi_1 = x, and phi_k(x)=L_k(x) - L_{k-1} (x) for k > 1, > where L_k(x)=kth Legendre polynomial." > > How can I obtain these coefficients "uh_k" on the several elements? > > Thanks in advance for any suggestion, > > Andrey From nek5000-users at lists.mcs.anl.gov Tue May 31 12:30:43 2016 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Tue, 31 May 2016 19:30:43 +0200 (CEST) Subject: [Nek5000-users] laplacian in axisymmetric mode (IFAXIS=T) Message-ID: Hi neks, I have a question about how to compute the laplacian of a scalar quantity (e. g. temperature), if IFAXIS is set to true, such as ( \nabla^2(TEMP) ). Is there an easy way to do it, or someone has already faced this problem? Thanks Mirko Farano -------------- next part -------------- An HTML attachment was scrubbed... URL: