[arsenis]

Hello,
I guess that my question could be in Main CFD forum, but I ask it here because it may have something to do with OpenFOAM.

I am trying to validate my Large Eddy Simulation settings recreating some experiments with very simple geometries (flow between two cubes).
The problem is that even though I use the recommended settings (orthogonal hex mesh, low yPlus, second order schemes, low Courant), some of my results tend to be a little off.
I use the pisoFoam solver and for most of my simulations the time step is steady in 1e-3 sec.

The fact with my data and my results is that they have a quite small velocity of the order of 0.15 m/sec (upper stream) to 0.01 m/sec (between the cubes) (they used a water channel, so they had high "enough" Reynolds in these velocities).
A typical output of my log files are something like this

Time = 87.631

Courant Number mean: 0.10769 max: 0.539192
smoothSolver: Solving for Ux, Initial residual = 0.000658275, Final residual = 1.4784e-06, No Iterations 2
smoothSolver: Solving for Uy, Initial residual = 0.00900013, Final residual = 1.36536e-06, No Iterations 3
smoothSolver: Solving for Uz, Initial residual = 0.00620374, Final residual = 9.20158e-07, No Iterations 3
GAMG: Solving for p, Initial residual = 0.0335739, Final residual = 0.000717153, No Iterations 2
time step continuity errors : sum local = 1.48048e-08, global = 1.7812e-19, cumulative = -5.59358e-13
GAMG: Solving for p, Initial residual = 0.00155629, Final residual = 9.97297e-07, No Iterations 21
time step continuity errors : sum local = 2.06779e-11, global = 5.51552e-19, cumulative = -5.59358e-13
ExecutionTime = 152202 s ClockTime = 153366 s

Today I realised, as you also can see, that my initial residual for the velocity is at best in the order of 1e-4, or in the worst in the order of 1e-3, when my results are in same order.
So my question is:
a) This initial residual seems that I already have lost my solution and that the final residual is just useless?
or
b) I shouldn't care, because that's why the method is iterative?

I am interested in you comments,
Arsenis