Answered commented questions
This commit is contained in:
@ -139,6 +139,19 @@ void FixGranularMDR::setup_pre_force(int /*vflag*/)
|
|||||||
// ANSWER2: So users cannot define 2 MDR models with different moduli?
|
// ANSWER2: So users cannot define 2 MDR models with different moduli?
|
||||||
// Should an error be invoked if two MDR models are therefore defined?
|
// Should an error be invoked if two MDR models are therefore defined?
|
||||||
// Should there be an error if a user uses a non-MDR model with an MDR model?
|
// Should there be an error if a user uses a non-MDR model with an MDR model?
|
||||||
|
// ANSWER3: So users cannot define 2 MDR models with different moduli?
|
||||||
|
// -> Correct, as of now it only makes sense to have 1 MDR model with one set of material
|
||||||
|
// properties defined per simulation.
|
||||||
|
// Should an error be invoked if two MDR models are therefore defined?
|
||||||
|
// -> Yes, it would be good to have an error alerting the user that only 1 MDR model
|
||||||
|
// is allowed to be defined.
|
||||||
|
// Should there be an error if a user uses a non-MDR model with an MDR model?
|
||||||
|
// -> It depends, if it is a rolling or tangential model then its okay. We superimpose
|
||||||
|
// the MDR model with a rolling and tangential model for the tableting simulation.
|
||||||
|
// However, I don't think we should allow other normal models to be defined. I don't
|
||||||
|
// really know what it would mean physically if a hooke-particle contacted a mdr-particle.
|
||||||
|
// The gran/wall/region interaction should also be MDR so that the mean surface displacement
|
||||||
|
// is correctly calculated.
|
||||||
|
|
||||||
psi_b_coeff = norm_model->psi_b;
|
psi_b_coeff = norm_model->psi_b;
|
||||||
|
|
||||||
@ -165,9 +178,10 @@ void FixGranularMDR::pre_force(int)
|
|||||||
mean_surf_disp();
|
mean_surf_disp();
|
||||||
|
|
||||||
// QUESTION: What about fix wall/gran?
|
// QUESTION: What about fix wall/gran?
|
||||||
// Answer: We never considered interaction between the mdr contact model and fix wall/gran
|
// ANSWER: We never considered interaction between the mdr contact model and fix wall/gran
|
||||||
// ANSWER2: Ideally this would be added, but at least temporarily I added an
|
// ANSWER2: Ideally this would be added, but at least temporarily I added an
|
||||||
// error. This incompatibility should be noted in the docs
|
// error. This incompatibility should be noted in the docs
|
||||||
|
// ANSWER3: Sounds good.
|
||||||
|
|
||||||
auto fix_list = modify->get_fix_by_style("wall/gran");
|
auto fix_list = modify->get_fix_by_style("wall/gran");
|
||||||
for (int i = 0; i < fix_list.size(); i++) {
|
for (int i = 0; i < fix_list.size(); i++) {
|
||||||
|
|||||||
@ -627,6 +627,7 @@ double GranSubModNormalMDR::calculate_forces()
|
|||||||
// However, this is written in a bit of a strange way now that I am reviewing it, I will give it some more thought.
|
// However, this is written in a bit of a strange way now that I am reviewing it, I will give it some more thought.
|
||||||
// ANSWER2: What if the contact type is WALL, then the iteration over contactSide = 0 will skip
|
// ANSWER2: What if the contact type is WALL, then the iteration over contactSide = 0 will skip
|
||||||
// defining R0 and R1? Is that correct?
|
// defining R0 and R1? Is that correct?
|
||||||
|
// ANSWER3: Correct, but in that case they are not needed since there is no partitioning of the displacement.
|
||||||
|
|
||||||
double delta_geo, delta_geo_alt;
|
double delta_geo, delta_geo_alt;
|
||||||
double delta_geoOpt1 = deltamax * (deltamax - 2.0 * R1) / (2.0 * (deltamax - R0 - R1));
|
double delta_geoOpt1 = deltamax * (deltamax - 2.0 * R1) / (2.0 * (deltamax - R0 - R1));
|
||||||
|
|||||||
Reference in New Issue
Block a user