# [ANSYS Meshing] Meshing for piston compression- deforming

 Register Blogs Members List Search Today's Posts Mark Forums Read

April 4, 2015, 11:04
Meshing for piston compression- deforming
#1
Member

sanjeet Limbu
Join Date: Mar 2015
Posts: 91
Rep Power: 2
Dear all,

I am trying to simulate a rapid compression machine- it simulate the compression stroke in a cylindrical - axi-symmetric chamber containing mixture which is compressed by a piston rigid body motion. I made a udf for the rigid body motion.
But the I am facing the error when i use the mesh shown in fluent." Divergence detected- AMG solver issues- Enthalpy".

Please suggest how to get it done.
Also I am unable to view the mesh motion as per the UDF function for piston - named element.

The idea is to make the piston face move and adjacent interior cell compress/ by layering method. The velocity of piston as rigid body accelerate to a peak and decelerate to zero : this according to compression ratio 16.5.
The following is the udf:
#include "udf.h"
DEFINE_CG_MOTION(oscillate, dt, vel, omega, time, dtime)
{
face_t f; /* define the variables */

t = DT_THREAD(dt); /* get the thread pointer for which the motion is defined */

/* if (!Data_Valid_P())
/* return; /* check if the values of the variables are accessible before you compute the function */

begin_f_loop(f, t) /* loop over each face in the zone to create an array of data */
{
if (time <= 0.010)
vel[0] = (600* time); /* define the velocity of the moving zone---*/
else if (0.01 < time < 0.019)
vel[0] = 6 +(888.89 * time);
else if (0.019 < time < 0.027)
vel[0] = 14;
else if (0.027 < time < 0.03)
vel[0] = 14 - (466.67*time);
else if (0.033 < time)
vel[0] = 0;
}
end_f_loop(f, t)
}

Pl suggest how to remove the negative volume issue[/QUOTE]
Attached Images
 Capture.jpg (46.9 KB, 15 views) Capture1.jpg (36.5 KB, 15 views) Capture2_dynamic mesh on.jpg (43.4 KB, 10 views) Capture3.jpg (40.8 KB, 10 views) Capture4.jpg (58.3 KB, 8 views)

Last edited by sanjeetlimbu; April 4, 2015 at 20:53.

 April 10, 2015, 08:33 #2 Senior Member   Paritosh Vasava Join Date: Oct 2012 Location: Lappeenranta, Finland Posts: 553 Rep Power: 13 Not that I am complaining but you are in the wrong part of the forum. You can post your question on the fluent forum and you might get help quickly. First of all if you plan to use dynamic mesh do consider switching to tetrahedral meshes. As you can see in one of the pictures the problem is with the mesh update. You can try to use a smaller time step. For example a time step small enough so that the maximum displacement of the moving boundary is smaller than the smallest mesh element. Also use appropriate dynamic mesh parameters. In the mesh method setting see the mesh scale info. Use these values in the setting. The AMG solver problem would be fixed once your time step is small enough and mesh quality is reasonable.

April 10, 2015, 09:03
#3
Member

sanjeet Limbu
Join Date: Mar 2015
Posts: 91
Rep Power: 2
ok i got the motion part done, i am getting the zone motion preview as needed.

but the mesh seems not good enough. that is now causing negative volume issue/error after few time steps(i tried lowering both time step and elemnet size but it may be due to the skewness)
Attached Images
 Orthogonal_0.9_near partition.jpg (38.9 KB, 8 views) skewness_0.355.jpg (49.2 KB, 6 views) neg_volume error.jpg (45.9 KB, 4 views) mesh_negative.jpg (52.8 KB, 6 views)

 April 11, 2015, 10:57 #4 Member   sanjeet Limbu Join Date: Mar 2015 Posts: 91 Rep Power: 2 Dear sir I am getting negative volume error even after doing the edge sizing and mapping the mesh zones. when i remove the crevice part and used the rectangle for cylinder shape only then it run, by reducing the time step to 4.16E-07s . But if i wanna use the crevice as shown in photos before the grid move for one cel length -first layering and instantly shows the error I think the moving rigid body mesh - crevice is not having the grid in line -parrallel to the grid line in the body -part. They are at angle due to the crevice shape. how to make it work- please suggest

 April 13, 2015, 03:22 #5 Senior Member   Paritosh Vasava Join Date: Oct 2012 Location: Lappeenranta, Finland Posts: 553 Rep Power: 13 You have disabled Smoothing and Remeshing. Do enable these options and in the smoothing part play around Spring Constant Factor. I still believe that 1e-3 sec. is a very big time step for that narrow region with 1 mm gap while the wall velocity is ranging from 6 m/s to 20 m/s. Try a drastically small time step lets say 1e-20 or 1e-10. Check the mesh motion and if everything seems alright increase the time step order gradually. Last edited by vasava; April 13, 2015 at 03:23. Reason: typo

 April 13, 2015, 09:03 #6 Member   sanjeet Limbu Join Date: Mar 2015 Posts: 91 Rep Power: 2 thanks I will try that ... but teh time will increase i i go very low.. I managed to run using e-6s I am getting some divergence error AMG solver- Enthalpy. and some warning that the maximum temperature exceeded the PDF table do you have any idea- i am using ignition delay model

 April 14, 2015, 01:43 #7 Senior Member   Paritosh Vasava Join Date: Oct 2012 Location: Lappeenranta, Finland Posts: 553 Rep Power: 13 I am not expert with ignition. But you can limit the temperature range and use smaller relaxation to keep solution going little further.

 Thread Tools Display Modes Linear Mode

 Posting Rules You may not post new threads You may not post replies You may not post attachments You may not edit your posts BB code is On Smilies are On [IMG] code is On HTML code is OffTrackbacks are On Pingbacks are On Refbacks are On Forum Rules

 Similar Threads Thread Thread Starter Forum Replies Last Post [ANSYS Meshing] Can I use ANSYS Meshing for manual meshing? Kirjain ANSYS Meshing & Geometry 6 November 30, 2015 09:11 FluentStarter FLUENT 2 April 19, 2015 18:48 KITetima FLUENT 3 April 19, 2015 14:22 Ken Main CFD Forum 0 September 4, 2003 11:09 ken FLUENT 0 September 4, 2003 11:08

All times are GMT -4. The time now is 12:28.