[Nek5000-users] dissipation

[nek5000-users at lists.mcs.anl.gov]

Dear neks,
I am trying to user comp_sij to compute dissipation in every point of the
domain.
The data storage is organised according to
      real sij(lx1*ly1*lz1,nij,lelv)
so the index of a component of tensor
precedes number of element, and
the is no continuous block of lx1*ly1*lz1,lelv points for an element of
tensor.
And so, the global functions like col3 of glsc can not be applied.
At least I don't understand how.

To bypass this, I copied comp_sij to my usr file and changed the order of
last indices in sij
      real sij(lx1*ly1*lz1,lelv,nij)


After that I use the followng code to compute viscous dissipation

       nij = 3
       if (if3d) nij=6

      call il_comp_sij(sij,nij,vx,vy,vz,ur,us,ut,vr,vs,vt,wr,ws,wt)

      call rzero(il_dissip,lv)

      do j=1,nij
      call col3(sij2,sij(1,1,1,1,j),sij(1,1,1,1,j),nxyz)
      il_dissip=il_dissip+sij2
       if (if3d .and. j.ge.4) then
      il_dissip=il_dissip+sij2
       end if
      if ( .not. if3d .and. j.eq.3 ) then
      il_dissip=il_dissip+sij2
      end if
      il_dissip=2*param(2)*il_dissip
      enddo

Is is a fair way to compute dissipation?

Thanks, Ilias
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/nek5000-users/attachments/20151108/ee2f97bd/attachment-0001.html>