I'm simulating droplet impact onto a moving particle (axisymmetric VOF+6DOF in dynamic mesh). I tried to increase the particle mass in UDF by a factor of 1000 (for the same particle size). But, as you can see in the attached picture, apparently Fluent doesn't realize this huge change in mass.
I guess the reason my results are quite different compared to what I get in experiments is that Fluent doesn't have a sense of the particle mass (and so kinetic energy). How should I define the mass for Fluent?
Here is UDF that I currently use (in SI units):
#include "udf.h"
DEFINE_SDOF_PROPERTIES(test_box, prop, dt, time, dtime)
{
prop[SDOF_MASS] = 4.4e-6;
prop[SDOF_IXX] = 1.76e-12;
prop[SDOF_IYY] = 1.76e-12;
prop[SDOF_IZZ] = 1.76e-12;
prop[SDOF_ZERO_TRANS_X] = FALSE;
prop[SDOF_ZERO_TRANS_Y] = TRUE;
prop[SDOF_ZERO_TRANS_Z] = TRUE;
prop[SDOF_ZERO_ROT_X] = TRUE;
prop[SDOF_ZERO_ROT_Y] = TRUE;
prop[SDOF_ZERO_ROT_Z] = TRUE;
printf ("\n2d_test_box: Updated 6DOF properties");
}
Is the velocity of the moving particle kept same as before? Please !
No. I enter the particle velocity in 'Dynamic Mesh Zone' window in the 'Velocity of Centre of Gravity' (attached picture). This velocity starts changing as soon as I run the simulation (due to air drag, and when particle collides onto the droplet, the velocity drops suddenly).
The particle in this case should be deforming! isn't it?
Why there is another particle which is particle shadow. What is the general definition of it?
I used 'rigid body' for particle. What is the reason for defining that as deforming?
When I define my particle as a 'solid zone', this particle shadow is created and I can define the contact angle on that. Otherwise, if I don't have the solid zone, there is no particle shadow and I should define the contact angle for the particle as wall (in boundary conditions). Do you know why Fluent does not recognize the mass?
Yes! That is true.
Fluent does recognize the mass which has been verified in many cases under gravity conditions.
Dear Sayed,
Your problem is very interesting again. The shadow should be able to be linked with a density for consider mass "information". I revised the fluent equations for your problem and the density is right there.
We could consider more two options in this case:
- not create a shadow, but get into with a real solid with solid-fluid interface specified: in this case, you can specify the material and so the density;
- or use relative density materials: this could correct the density, but you must also correct the viscosity, since both properties represent the phenomena in your problem.
I will try something this week for helping you and I will keep you in touch.
Best regards
Dear Guilherme,
Thank you so much for replying.
Regarding your first alternative, if I need to assign a material to the particle, I need to define it as a separate zone. Therefore, I need to mesh inside the particle as well and so the "particle shadow" will be created anyway.
I've tried both and I think having or not having shadow doesn't change the results much.
I also tried to activate Discrete Phase model and checked the box "Interaction with Continuous Phase". But it didn't make any change in the results presented in the pictures above.
I'm not sure if I get your 2nd point regarding how to do use relative density materials.
Thanks again,
Sayed
PS.
1) I know Fluent recognizes the mass as we can see in this tutorial which I feel is very similar to my case and I used the tutorial configuration in my problem as much as possible.
2) In contrast with tutorial, I'm using laminar model as the maximum cell Reynolds number is 500 in all my solutions so far (I've tried turbulent as well, but laminar results are better!)
- Badges
- Science topic