A question of understanding regarding the number of pressure steps about mplbm-ut

[je-santos]

That is a great question, Nick.

I don't have a straightforward answer. Many of those steps will converge instantly, since they won't do much. It's worth a shot. If you get some insights please send a PR so other people can benefit from this.

Some years ago I was thinking of implementing something better than linear, see #48
nevertheless, I ran out of time :(

[NckBlfldt]

Hello Javier,

Thanks for the quick feedback. I am trying this for a domain of the size 500x500x280 voxels with a convergence of 10^-3 for 100 pressure steps. The pressure steps have so far always converged after 1-2 iterations for the first 10 density differences.
Is it possible that if my steps in the density difference (e.g. 0.0005) are too small, with a tolerance of 10^-3 my simulation simply converges immediately and as a result the calculated values will be wrong?
Maybe you can help me to understand the convergence more closely, what exactly does 10^-3 mean in the Shan-Chen model?

Thanks for all the help!

With kind regards,

Nick

[je-santos]

It's not really related to the SC model is a way for my code to check if things aren't "moving as much". Check it out here:

https://github.com/je-santos/MPLBM-UT/tree/master/examples/young_laplace_validation