Permalink
Browse files

Merge pull request #5706 from rmcdermo/master

FDS User Guide: correct defaults for VN_MIN and VN_MAX
  • Loading branch information...
rmcdermo committed Nov 14, 2017
2 parents e6be949 + b7050b8 commit 288612c83a4e6e1f73328bc8b574675e66242300
Showing with 4 additions and 5 deletions.
  1. +3 −3 Manuals/FDS_User_Guide/FDS_User_Guide.tex
  2. +1 −2 Source/read.f90
@@ -1472,9 +1472,9 @@ \subsubsection{The Von Neumann Constraint}
The Von Neumann constraint is given by
\begin{equation}
\mbox{VN} \equiv \delta t \; \max \left[ \frac{\mu}{\rho},D_\alpha \right] \; \left( \frac{1}{\delta x^2}+\frac{1}{\delta y^2}+\frac{1}{\delta z^2} \right) < \frac{1}{2}
\mbox{VN} \equiv 2 \; \delta t \; \max \left[ \frac{\mu}{\rho},D_\alpha \right] \; \left( \frac{1}{\delta x^2}+\frac{1}{\delta y^2}+\frac{1}{\delta z^2} \right) < 1
\end{equation}
The Von Neumann stability check is invoked by setting {\ct CHECK\_VN=.TRUE.} on the {\ct MISC} line (for DNS, {\ct CHECK\_VN=.TRUE.} by default). The limits for VN may be adjusted using {\ct VN\_MIN} (default 0.4) and {\ct VN\_MAX} (default 0.5) on {\ct MISC}. We can understand this constraint in a couple of different ways. First, we could consider the model for the diffusion velocity of species $\alpha$ in direction $i$, $V_{\alpha,i} Y_\alpha = -D_\alpha \; \partial Y_\alpha/\partial x_i$, and we would then see that VN is simply a CFL constraint due to diffusive transport.
The limits for VN may be adjusted using {\ct VN\_MIN} (default 0.8 for LES, 0.4 for DNS) and {\ct VN\_MAX} (default 1.0 for LES, 0.5 for DNS) on {\ct MISC}. We can understand this constraint in a couple of different ways. First, we could consider the model for the diffusion velocity of species $\alpha$ in direction $i$, $V_{\alpha,i} Y_\alpha = -D_\alpha \; \partial Y_\alpha/\partial x_i$, and we would then see that VN is simply a CFL constraint due to diffusive transport.
We can also think of VN in terms of a total variation diminishing (TVD) constraint. That is, if we have variation (curvature) in the scalar field, we do not want to create spurious oscillations that can lead to an instability by overshooting the smoothing step. Consider the following explicit update of the heat equation for $u$ in 1-D. Here subscripts indicate grid indices and $\nu$ is the diffusivity.
\begin{equation}
@@ -9085,7 +9085,7 @@ \section{\texorpdfstring{{\tt MISC}}{MISC} (Miscellaneous Parameters)}
{\ct CFL\_VELOCITY\_NORM} & Integer & Section~\ref{info:CFL} & & 0 (LES), 1 (DNS) \\ \hline
{\ct CHECK\_HT} & Logical & Section~\ref{info:CFL} & & {\ct .FALSE.} \\ \hline
%{\ct CHECK\_REALIZABILITY} & Logical & Strict check of ZZ>0 \& SUM(ZZ)=1 & & {\ct .FALSE.} \\ \hline
{\ct CHECK\_VN} & Logical & Section~\ref{info:CFL} & & {\ct .FALSE.} \\ \hline
%{\ct CHECK\_VN} & Logical & Section~\ref{info:CFL} & & {\ct .FALSE.} \\ \hline
{\ct CLIP\_MASS\_FRACTION} & Logical & Section~\ref{info:CLIP} & & {\ct .FALSE.} \\ \hline
{\ct CNF\_CUTOFF} & Real & Section~\ref{info:particle_size} & & 0.005 \\ \hline
%{\ct COMPUTE_VISCOSITY_TWICE} & Logical & & & {\ct .TRUE.} \\ \hline
View
@@ -1605,7 +1605,7 @@ SUBROUTINE READ_MISC
GRAVITATIONAL_SETTLING,GVEC,DT_HVAC,H_F_REFERENCE_TEMPERATURE,&
HRRPUV_MAX_SMV,HUMIDITY,HVAC_MASS_TRANSPORT,&
IBLANK_SMV,IMMERSED_BOUNDARY_METHOD,INITIAL_UNMIXED_FRACTION,&
LES_FILTER_WIDTH,MAX_CHEMISTRY_ITERATIONS,&
MAX_CHEMISTRY_ITERATIONS,&
MAX_LEAK_PATHS,MAXIMUM_VISIBILITY,MPI_TIMEOUT,&
N_FIXED_CHEMISTRY_SUBSTEPS,N_INITIAL_PARTICLE_SUBSTEPS,NEAR_WALL_TURBULENCE_MODEL,&
NOISE,NOISE_VELOCITY,NO_EVACUATION,NO_RAMPS,&
@@ -1718,7 +1718,6 @@ SUBROUTINE READ_MISC
ELSE
FLUX_LIMITER = SUPERBEE_LIMITER
TURBULENCE_MODEL = 'DEARDORFF'
LES_FILTER_WIDTH = 'MEAN'
ENDIF
! FDS 5 options (diagnostic timing purposes)

0 comments on commit 288612c

Please sign in to comment.