Includes topological point generation and optional removal of all points (and
triangles) that do not originate from mesh edges. This generates
polygons rather than triangles providing smoother surfaces and significantly
small files (~ factor of 3-4 smaller).
Contributed by Mattijs Janssens
This was not being used and had not been kept up to date. It also didn't
actually do dimension checking, so the name was misleading, and the
functionality added on top of SubField was minimal.
coneInjection has been extended to include the functionality of
coneNozzleInjection, and the latter has been removed.
Some parameters have changed names. The "positionAxis" entry from
coneInjection has been removed in preferance of coneNozzleInjection's
single "position" and "direction" entries. This means that only one
injection site is possible per model (dictionary substitutions mean that
only minimal additions are required to add further injection sites with
the same parameters). The name of the velocity magnitude has been
standardised as "Umag" and "innerDiameter" and "outerDiamater" have been
renamed "dInner" and "dOuter" for consistency with the inner and outer
spray angles.
Velocity magnitude and diameters are no longer read when they are not
required.
The randomisation has been altered so that the injections generate a
uniform distribution on an cross section normal to the direction of
injection. Previously there was an unexplained bias towards the
centreline.
An example specification with a full list of parameters is shown below.
injectionModels
{
model1
{
type coneInjection;
// Times
SOI 0;
duration 1;
// Quantities
massTotal 0; // <-- not used with these settings
parcelBasisType fixed;
parcelsPerSecond 1000000;
flowRateProfile constant 1;
nParticle 1;
// Sizes
sizeDistribution
{
type fixedValue;
fixedValueDistribution
{
value 0.0025;
}
}
// Geometry
positions (-0.15 -0.1 0);
directions (1 0 0);
thetaInner 0;
thetaOuter 45;
// - Inject at a point
injectionMethod point;
// - Or, inject over a disc:
/*
injectionMethod disc;
dInner 0;
dOuter 0.05;
*/
// Velocity
// - Inject with constant velocity
flowType constantVelocity;
Umag 1;
// - Or, inject with flow rate and discharge coefficient
// This also requires massTotal, dInner and dOuter
/*
flowType flowRateAndDischarge;
Cd 0.9;
*/
// - Or, inject at a pressure
/*
flowType pressureDrivenVelocity;
Pinj 10e5;
*/
}
model2
{
// The same as model1, but at a different position
$model1;
position (-0.15 0.1 0);
}
}
By default the prefix is no longer added to the field names but the previous
behaviour can be reproduced by setting the optional "prefix" entry to "on" or
"yes".
to avoid the need to evaluate departure functions and simplify evaluation of the
temperature. In general it makes more sense to use and e/Cv based
thermodynamics when solving for internal energy rather than h/Cp and have
convert between the energy forms.
All related tutorials and test cases have also been updated.
Changed liquid thermo from sensibleEnthalpy to sensibleInternalEnergy in
tutorials. It is generally more convergent and stable to solve for internal
energy if the fluid is incompressible or weakly compressible.
Zeroing a dimensioned field can now be achieved by assignment to the
zero type. This prevents the clutter associated with constructing an
appropriate dimensioned type, or having to use multiply-equals-zero as a
workaround.
This interpolation method was previously removed by commit fbf00209.
The intention of this method is to provide a slip-like wall boundary
condition for the velocity when interpolated to the location of a
Lagrangian element. This is difficult because any velocity which points
through the wall can cause a drag model and a rebound wall interaction
to "fight"; i.e., the drag pushes the particle to the wall, the wall
bounces it back. This can result in the program hanging.
This method extrapolates a vector field to the wall points and then
modifies the result so that it does not point through the wall. It does
this by rotating the vectors towards the (reversed) point normal. The
result is also scaled so that is reduced to zero if the necessary
rotation exceeds 90 degrees.
This provides an alternate resolution to bug report
https://bugs.openfoam.org/view.php?id=2826
Within this structure the BirdCarreau, Casson, CrossPowerLaw, HerschelBulkley,
powerLaw and strainRateFunction strain-dependent viscosity based non-Newtonian
fluid models may be selected for incompressible or compressible flow.
In the case of compressible flow the strain-dependent viscosity functions are
applied to the temperature dependent viscosity so that if the WLF viscosity
model is chosen in conjunction with the CrossPowerLaw the effective model is
Cross-WLF which is commonly used for polymer flow.
These models are selected in the constant/turbulenceProperties file, e.g.
simulationType laminar;
laminar
{
laminarModel generalizedNewtonian;
viscosityModel CrossPowerLaw;
nuInf 10;
m 0.4;
n 3;
}
This new implementation is more general and flexible than the previous
incompressible only non-Newtonian viscosity models, which were selected in the
constant/transportProperties file. This implementation is now deprecated and
will be phased-out.
Description
Transport package using the Williams-Landel-Ferry model.
Templated into a given thermodynamics package (needed for thermal
conductivity).
Dynamic viscosity [kg/m.s]
\f[
\mu = \mu_0 \exp \left(\frac{-C_1 ( T - T_r )}{C_2 + T - T_r}\right)
\f]
References:
\verbatim
Williams, M. L., Landel, R. F., & Ferry, J. D. (1955).
The temperature dependence of relaxation mechanisms
in amorphous polymers and other glass-forming liquids.
Journal of the American Chemical society, 77(14), 3701-3707.
\endverbatim
This fix changes how the intersections loop ignores previously
intersected faces. It now marks them by their index so that subsequent
iterations ignore them.
Before this change, after an intersection was found the start point was
advanced by a small amount to move the past the intersection. The
problem with this was if multiple boundary faces or the end point were
in close proximity to the intersection then the move forward might span
them. This could lead to intersections being missed or counted multiple
times, in some cases indefinitely.
Based on a patch contributed by Mattijs Janssens
Resolves bug report https://bugs.openfoam.org/view.php?id=1147
