Merge JS's changes in

[rupertnash]

1. Corrected forcing term (d3q15.c)
2. Fixed inconsistecy between forces and force densities (d3q15.c)
3. Included forcing in initialization (Lattice.i)
4. Fixed indexing error in PBCs (bc_pbc.c)

[joakimstenhammar]

Hi Rupe,

Is this something I need to act on? I can't find any button for
accepting the pull request... Or is it just for information?

Cheers,

Joakim

On 2016-11-15 15:30, Rupert Nash wrote:

1. Corrected forcing term (d3q15.c)
2. Fixed inconsistecy between forces and force densities (d3q15.c)
3. Included forcing in initialization (Lattice.i)
4. Fixed indexing error in PBCs (bc_pbc.c)

---

    You can view, comment on, or merge this pull request online at:

#3

    Commit Summary

• Fixed some bugs:

  File Changes
  

• M libd3q15/Makefile
  https://github.com/rupertnash/subgrid/pull/3/files#diff-0 (4)

• M libd3q15/bc_pbc.c
  https://github.com/rupertnash/subgrid/pull/3/files#diff-1 (2)

• M libd3q15/d3q15.c
  https://github.com/rupertnash/subgrid/pull/3/files#diff-2 (55)

• M python/d3q15/Lattice.i
  https://github.com/rupertnash/subgrid/pull/3/files#diff-3 (6)

• M python/dqTools/Controller.py
  https://github.com/rupertnash/subgrid/pull/3/files#diff-4 (1)

  Patch Links:
  

• https://github.com/rupertnash/subgrid/pull/3.patch

• https://github.com/rupertnash/subgrid/pull/3.diff

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
#3, or mute the thread
https://github.com/notifications/unsubscribe-auth/AIjjniHXSQqOokR6CxaygLlGWO_yeiAYks5q-cIZgaJpZM4KyluG.

[rupertnash]

This is really just for a logical place to discuss this.

I've done a preliminary test and if you set up your lattice with rho = 1, u = 0 and a non-zero force, then call initFromHydroVars() and updateHydroVars() you do not get zero velocity where the force != 0. I'm going to look at this more

[joakimstenhammar]

But neither should we - right? Is there not a correction term of
f/(2*\Delta t) that should be added to the velocity where f != 0? The
problem in my test was that this correction seems to end up in the wrong
lattice point... /Joakim

On 2016-11-15 16:30, Rupert Nash wrote:

This is really just for a logical place to discuss this.

I've done a preliminary test and if you set up your lattice with rho =
1, u = 0 and a non-zero force, then call initFromHydroVars() and
updateHydroVars() you do not get zero velocity where the force != 0.
I'm going to look at this more

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
#3 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/AIjjnnVh8DS5x8066JoEJbcX6Xct0yS0ks5q-dAvgaJpZM4KyluG.

[rupertnash]

If you have rho and u, compute the equilibrium f, then compute rho and u from that, you MUST get back the same inputs up to machine precision.

[rupertnash]

I am reasonably confident that my calc_hydro_site is correct (given that the "force" array stores a force density)

[joakimstenhammar]

In that case, yes, but in that case all the other routines should be
changed accordingly (phi and f_0). The latter two are currently
identical to your PRE equations, where F is treated as a force rather
than a force density (although you say differently in the introduction)

• do you agree?

On 2016-11-15 17:20, Rupert Nash wrote:

I am reasonably confident that my calc_hydro_site is correct (given
that the "force" array stores a force density)

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
#3 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/AIjjnkD-XozNYaYUH-h3b6DairayZNmnks5q-dvAgaJpZM4KyluG.

[joakimstenhammar]

Did you find any further leads? /J

On 15/11/16 17:20, Rupert Nash wrote:

I am reasonably confident that my calc_hydro_site is correct (given
that the "force" array stores a force density)

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
#3 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/AIjjnkD-XozNYaYUH-h3b6DairayZNmnks5q-dvAgaJpZM4KyluG.