[Nek5000-users] multiple runs and poisson for velocity

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

Dear Neks,
I have two issues that I would like to discuss with you.
Firstly, I would like to perform many steady state simulations with different values of the viscosity, each time using the previous solution as starting condition.
For this reason, I have built the following subroutine:

c-----------------------------------------------------------------------
      subroutine userchk
      include 'SIZE'  
      include 'TOTAL' 


      do 120 ire=1,10
      param(2) = -1*ire
      call setprop
      igeom=1
      call fluid(igeom)
120 continue    

      ifxyo = .true.
      call outpost(vx,vy,vz,pr,t,'ste')
      call exitt

      return
      end
c-----------------------------------------------------------------------

but this does not work as I expected, could somebody suggest me what I should do instead?

Secondly, I would like to solve (in postprocessing mode) a Poisson problem for the velocity components, in the form -lapl(vx) = fx and -lapl(vy)=fy, with fx and fy known.
Is it possible to do this within NEK? I think this could be done by adapting the Uzawa pressure solver in navier1.f, but I am not sure if this could work.
Best regards,
Giuseppe