README.md 2.04 KB
Newer Older
ylan's avatar
ylan committed
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
## 11145 (non-touching) pebbles in a cylinder (Last updated: Jun. 12, 2020)
cyl11145 N, Version 3.0, E3,575,076

This mesh is not very stable for Re10k. There is (at least) a known location having high distorted Cell.  
Run dt=5e-4 is fine for the first 14000 steps.   
After that dt=4e-4 will have CFL too high stop at 14000+5996th step (time=9.99800000e+00)   
We never manage to run over an convection unit  

This is a better mesh using improved techniques developed within last few months.  
Please note that the improvement has not finished.  

- New features:
  - A more robust and cleaner code with lots of user intreface improvement
  - Add Taubin smoothing before edge collapse
  - Switch to a passive tessellation if the first attempt didn't do well. 
  - Add local optimization for better performance.
  - dump co2 directly  
  - DEM results from project chrono on nurburg gpu
  - Fix some bugs as well  

- Geometry and BC:
ylan's avatar
ylan committed
22
  - Sphere distribution: free-free
ylan's avatar
ylan committed
23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
  - Sphere radius = 1.0 (Wall)
  - Cylinder radius = 40.672 (SYM), (convert SYM to Wall in usr file)
  - Zbot of the cylinder =  14.655 (Outflow)
  - Ztop of the cylinder = -12.627 (Inflow)
  - We use dimless scale to run Re10000 with vz=1.0

- Case files:
    - .re2, a curved (projected at lx1=3) hex20 re2 mesh, converted by a hex20 rea: xn3.rea+reatore2, R\_peb=1.0
      ```
        xyz min    -40.672      -40.672      -12.627    
        xyz max     40.672       40.672       14.655
      
        mesh metrics:
        GLL grid spacing min/max    : 3.67E-04 3.32E-01
        scaled Jacobian  min/max/avg: 1.52E-02 9.97E-01 3.65E-01
        aspect ratio     min/max/avg: 1.06E+00 6.06E+01 1.27E+01
      ```
    - Manage to run to time=7.0 with dt=5e-4 and the follow commit of NekRS
      ```
        commit c26b570af4a85e8ef8425b7e7057f0926f5aaace
        Author: Stefan Kerkemeier <stefanke@jlselogin2.ftm.alcf.anl.gov>
        Date:   Fri Aug 7 12:06:07 2020 +0000
        
            add missing file and fix gsh for scalars
      ```
    - peb.11k.dat, locations of the pebble centers under R\_peb=1.0