What does Resolution mean?

[Massload]

Hi, everyone.
I’m a newbie of LBM and I am testing the analysis with the driveaer model. I would like to set a lower resolution for the lattice because the model is large but I don’t know what Resolution(N) in the settings section means.
I understand that a larger N will increase the resolution but what value should I input for the model?
Is it possible to check the generated lattice before the calculation?
Now it seems that the error “Discrete normal is not valid for interpolated pressure boundary!” is caused by the resolution being too low.

[Adrian]

The N value passed to the unit converter is the number of cells per physical reference length (which is also managed by the unit converter). The resulting physical discretization width is then passed to e.g. the cuboid geometry.

There is no general answer to which value one needs for an application. It really depends on what you are doing.

You can view the resulting geometry in the VTK output (also without calculating any timesteps).
There you will also see why you get the “Discrete normal is not valid for interpolated pressure boundary” error – likely the inflow (?) where you want to apply the velocity boundary is not resolved enough for the selected value of N.

In case you did not see it yet: We also have a user guide [PDF] where the basic usage of OpenLB is detailed.

[Massload]

Thanks for replying, it is important information that N is the number of pieces to the representative length.
On the other hand, as I increase N to 100, I seem to get the following error and cannot generate a lattice.
“Discrete Normal is not valid for interpolated pressure boundary!”
The number of N is quite large, so it should not be possible to create a lattice, but do you know what the problem is?
The representative length is 1.5m.

[mathias]

Difficult to say without knowing any details. An image (from e.g. paraview) of the materials associated with particular boundary would be nice.

[Author]

Posts