Hi Akash,

I promise you I will take a look to your problem. Now I'm quite busy. I hope I will answer to you next weekend. Sorry :confused:

Regards

Hello Enry,

Thank you very much. You can have a look at the problem at your convenience. Although, if you have limited time, I would request you to look into my actual model file first and then the box trial file.

I will be giving the final presentation for my part of work on 24th so would like to include the UDF part if it is solved.

Thanks a lot. Good luck for your work,

Akash

Hi Akas, sorry for my delay. I'm busy for my master thesis. Well, I'm just modified the "box" file, because it's easier for me to be understood. Here is the link; I hope you can upload the file, because it's the first time I use file hosting. However, Try to read journal file and I hope i translated well the journal from fluent6.3.26.

Now I'm taking a look to your problem.
Tell me if you have just solved it.

http://www.mediafire.com/?3rzgh3deaoxjv9e

I'm trying to read your file but fluent crash! However, I could see the grid. The box problem is solved in the way I posted you early; your problem also should be solved in the same way of box2 ones.

Set boundary condition in the same way, and tell me if your problem doesn't work. If it doesn't work, send me only the mesh file, and not the case file.

I hope you should solve your problem before your presentation.

Tell me know! ;)

Hello Enry,

Thank you for your reply. I appreciate that you are taking time out of your master thesis for my help.

I ran the box2 file which you sent in the fluent 12. This case is similar to my problem as I want to move the interface too (the differences being that mine is solid-liquid interface and transient simulation). When I ran the journal file sent by you in Fluent 12, although up and floor are defined stationary, they also get deformed. I have also tried the same settings previously in my problem and same thing happens. Even after stationary definition, the whole body deforms. I am attaching a picture of your box2 mesh movement for the reference.

Also, i am sending my UDF once again. Can you please check if i am missing anything important in the UDF. It is pretty basic one applied to interface and interface shadow. Rest of the zones - solid and fluid are set to deforming, and boundaries like top, bottom are stationary. (just like your mesh settings)

As you asked, i am attaching only the mesh file for you to try the attached UDF on it and see for yourself the problem. NOTE that when you load the mesh - you need to set the interface zone from 'interface to wall' which will create interface shadow zone. Then, you can use the UDF on it and dynamic zone settings. Here is the link to mesh file -

http://www.mediafire.com/?zd5f1n87pfmzb2e

My presentation is tomorrow. So,I believe by that time, I will not be able to solve this problem but I will continue to work on this after that too and hopefully get this solved by this month's end.

thank you for your help.

akash

Hi akash,
It's very strange.

Look at my picture:

http://www.mediafire.com/imgbnc.php/...28bf415d6g.jpg


I don't know why... try to check if FLUENT 6.3.26 journal I send is the same of FLUENT12 ones. I red both of them and they seem to be the same. Check if Fluent do what the journal say, looking at 6.3.26 journal file.

problems remeshing 3D Valve
 

Hello, I have a similar problem that I can not solve.
Actually I have tried everything without suceess.

My problem is to simulate a opening valve in a pipe in 3D. I have solved the problem in 2D but.
I can make the valve move but it always get a non-possitive volume because remeshing doesn´t work!
does is anything to do with the UDF Code that I am using? (it is attached herein) I don´t think so because the valve moves

does anyone know why remeshing doesn´t work?? Am I foggeting something when working with 3D cases?

I would appreciate if anyone can give me any instruction to go on

thanks so much

(this picture depicts the symetry wall of the 3D design)




this code works for both 2D and 3D

DEFINE_CG_MOTION(rotating_body, dt, vel, omega, time, dtime)
{
real ampl,disp,omg,count,grad,current_time;
static int xTime,xOmega; /* la variable static conserva el valor de forma independiente sin que le afecte el bucle.

/* reset velocities */
NV_S (vel, =, 0.0);
NV_S (omega, =, 0.0);
xOmega = 0;

/* motion */
grad = 90;
ampl = (grad * M_PI) / 180 ; /* amplitude of rotation in radians (0.5deg)
omega[0] = ampl * sin(2*time);
if (xOmega > omega[2])
omega[0] = 0;
xOmega = ampl * sin(2*time);

remeshing 3D
 

Hello, I have a similar problem that I can not solve.
Actually I have tried everything without suceess.

My problem is to simulate a opening valve in a pipe in 3D. I have solved the problem in 2D but.
I can make the valve move but it always get a non-possitive volume because remeshing doesn´t work!
does is anything to do with the UDF Code that I am using? (it is attached herein) I don´t think so because the valve moves

does anyone know why remeshing doesn´t work?? Am I foggeting something when working with 3D cases?

I would appreciate if anyone can give me any instruction to go on

thanks so much

(this picture depicts the symetry wall of the 3D design)



DEFINE_CG_MOTION(rotating_body, dt, vel, omega, time, dtime)
{
real ampl,disp,omg,count,grad,current_time;
static int xTime,xOmega; /* la variable static conserva el valor de forma independiente sin que le afecte el bucle.

/* reset velocities */
NV_S (vel, =, 0.0);
NV_S (omega, =, 0.0);
xOmega = 0;

/* motion */
grad = 90;
ampl = (grad * M_PI) / 180 ; /* amplitude of rotation in radians (0.5deg)
omega[0] = ampl * sin(2*time);
if (xOmega > omega[2])
omega[0] = 0;
xOmega = ampl * sin(2*time);

remeshing 3d doesn´t work??
 

I have already read that "remeshing doesn't work for structured 3D"
what does it mean??
I am trying to simulate a opening valve in a 3D pipe but as you say "remeshing doesn´t work and it always get a non-positive volume...
so that, how would I do to move the valve??

I am using the follow UDF and it works for both 2D. Actually it works also for 3d but get negative volumes because of the mesh...

can you please help me out?
I would appreciate if you can give me any instruction
thanks so much

(I also send you a picture that depicts a valve in the pipe (it is a 3d but I send you just a symetry)

DEFINE_CG_MOTION(rotating_body, dt, vel, omega, time, dtime)
{
real ampl,disp,omg,count,grad,current_time;
static int xTime,xOmega; /* la variable static conserva el valor de forma independiente sin que le afecte el bucle.

/* reset velocities */
NV_S (vel, =, 0.0);
NV_S (omega, =, 0.0);
xOmega = 0;

/* motion */
grad = 90;
ampl = (grad * M_PI) / 180 ; /* amplitude of rotation in radians (0.5deg)
omega[0] = ampl * sin(2*time);
if (xOmega > omega[2])
omega[0] = 0;
xOmega = ampl * sin(2*time);

moving 3d valve
 

sorry about those messages, I got some problem with my connetion and I didn´t see that I have already sent it. Sorry about that.

I believe my 3d problem is nothing to do with UDF code because the valve moves.
Instead the problem shoud be related with remeshing because it doesn´t work. I have tried everything and now I ran out of ideas of how to proceed...
Please can you help me to figure it out??


thanks in advance

what type of mesh do you have? because I think remeshing doesn't work with structured mesh.

thanks for trying to help me,

I have tetra mesh so an unstructured mesh, right?

I am meshing with Ansys Workbench Mesh. And using Advance Functions "curvature and proximity"

I will be here for further information you´d need

opening valve
 

Hi again!
I send you a picture that depicts what is the problem
As you can see after some mm seconds I get negative volumes because
remeshing doesn´t work at all.
This is a valve that opens into a pipe. It is a 3D problem (but you just see a symetry wall)

Please I would appreciate any suggestion to get it done.
thanks in advance

have you tried reducing the time step size and changing the remeshing parameters? for example the size remeshing interval?

yes, I have tried reducing time step size further than 0.0001 and size remeshing interval from 10 to 1 and I didn´t find anything...
could you give me any address to send you a file so that you can see?

thanks

another approach...
 

Hi people,
I have managed to get the problem solved by using Sliding Mesh
but I still wonder how would I get same solution through Dynamic Mesh...
please, I need some help to make "remeshing" works...

any support??

thanks

Hi Andres,

I have this problem (negative volumes) even if I work with sliding interfaced mesh (circle inner quad-mesh oscillating, and outer tri-mesh steady to simulate 2d airfoil sinusoidal pitching), so without the "smoothing" that gave you problems like you mentioned above.

Please could you tell me how did you manage to get this problem solved by using sliding mesh?

Any idea about why I get negative volumes (when I run a preview mesh motion on Fluent) and how to solve it, please?

Many thanks.
Kind regards,

Alberto

Hi Alberto,
your problem can be faced using the mesh morphing tool RBF Morph (www.rbf-morph.com).
It's very fast and it can be used in transient analysis to prescribe the desired time history.

http://rbf-morph.com/images/rnd_image/sails-200.gif

Have a look to the last ANSYS webinar.

Regards.

Hi Marco,

thanks, sounds really interesting...but I am under University licence, and we have only Ansys package. I assume that RBF-Morph is not free even if you already have an Ansys licence, am I right? I don't see any software's download possibility on its website... (I have already created an account).

Regards,
Alberto

RBF Morph is a separate product, contact info@rbf-morph.com and ask!

Marco