Commit 61451fe2 authored by ylan's avatar ylan
Browse files

fix typos + set vb mode for 100 timesteps

parent db0e768b
......@@ -9,11 +9,11 @@ ann352625 N, Version 3.3, E98,782,067
- Fix edge insert, edge collapse, Eadj, and Qadj
- Notes
- DEM data from Yiqi (StarCCM++?)
- smaller edge collapse tol depends on geom + few local locations
- re2 is not in a regular Nek format. See userdat2 in user for proper boundary BC
- usr doesn't need dependencies (`my*.f`) if userchk is empty
- Geometries, pebble locations are stored at `ann350k.dat`
- DEM data from Yiqi (StarCCM++).
- smaller edge collapse tol depends few local locations
- re2 is NOT using a regular Nek format. See userdat2 in usr for proper boundary BC setups.
- usr doesn't need dependencies (`my*.f`) if userchk is empty.
- Geometries, pebble locations are stored in `ann350k.dat`
- Sphere Centers saved into `*txt` files
`pebble_centers_re2.txt`, sphere centers matched in re2
`pebble_centers_rad_1.txt`, after rescaled by `1.0/0.91945`
......@@ -30,8 +30,8 @@ ann352625 N, Version 3.3, E98,782,067
Zmax 60.3377
Rmin 24.1010
Rmax 82.3009
dtouch 1.997404 << min distance between any two spheres
dsep 1.999376 << "average" length to make each sphere touched by other 2.5 spheres
dtouch 1.997404 << min distance between any two spheres' centers
dsep 1.999376 << "average" length to seperate spheres, = kissing number at 2.5
```
(`modulde load python` on Summit)
(It takes some time to compute dtouch)
......@@ -63,12 +63,12 @@ ann352625 N, Version 3.3, E98,782,067
- Mesh file (`ann350k_t1r11.fld+X+time=0.0`)
- lx1=3 GLL points projected onto curved sides
- no issues for interpolation onto lx1=8 and mid-multigrid levels at lx1=4
- no issues for lx1=2 corase grid
- no issue for interpolation onto lx1=8 and mid-multigrid levels at lx1=4
- no issue for lx1=2 corase grid
- Production run
- Use re2 for boundary conditions, but use restart file for the curved mesh
- We use dimless scale to run Re5000 with vz=1.0 (Re10k seems ok, but no time/allocation left)
- We use dimless scale to run Re5000 with vz=1.0 (Re10k seems ok, but no time/allocationto test)
- NekRS version (after udf changes)
```
commit 82a68b6bbf550c84e3757ffcf06b783578a133d3
......
......@@ -31,7 +31,7 @@ void UDF_ExecuteStep(nrs_t *nrs, dfloat time, int tstep)
// from all solvers.
// After terminateVerboseOutput steps, allow verbose output
// only from solution projection.
const int terminateVerboseOutput = 20;
const int terminateVerboseOutput = 100;
if(tstep == (terminateVerboseOutput+1)){
nrs->pSolver->options.setArgs("VERBOSE", "FALSE");
if(nrs->uvwSolver){
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment