[almostafa67]

hi dear all...
i have same problem as u mayur, when i check mesh in fluent, fluent does not show any negative volume ,but when i start iteration this error shows up,i changed time step from 0.0001 to 0.00001 and cell height is 0.002 and my case is Dynamic mesh (smoothing and remeshing) could u help me out plz?
thanks a lot in advance for any help provided...

[-mAx-]

maybe is your body's displacement to big.
Try to reduce the time-step again

[almostafa67]

i changed time step from 0.001 to 10e-7 even smaller but again error message showed up!i think we should try something else and change another parameter such that fluent remeshes the pipe by itself. i think when the small cylinder moves fluent can not remesh that zone so this error message shoes up!any idea???
tnx in advance dear max....

[-mAx-]

you need to localize where is the problem. (which cells has negative volume)

[almostafa67]

how can i do that?

[-mAx-]

display/contour/
there is someting about mesh or cell (then volume)

[almostafa67]

hi dear MAX...
i reran fluent and this i used following command text
define/models/dynamic-mesh-controls/spring-on-all-shapes?...> yes

fluent did not show negative volume message this time and iteration progressed 35 time step
but another error message this timeivergence detected in AMG solver: x-momentum
i decreased under relaxation factor for momentum from 0.7 to 0.1 but no luck
could u plz guid me and help me out
tnx

[-mAx-]

prior to calculations did you check mesh motion (solve/mesh motion...)
it will help you debugging your case (if the divergence comes from your mesh or not)

[almostafa67]

i tried ur suggestion and this is the result

Error: eval: unbound variable
Error Object: motion
Invalid time
time step size (s) [9.9999997e-05]

i dont know what does it mean!by the way every time that i run fluent and change some parameters different error messages shows up,negative cell volume,Divergence detected in AMG solver: x-momentum,etc...sometimes it shows up soon and sometime late,what's the matter??

how can i solve this two annoying problem???
dear max it's very nice of you to point me in the right direction

[-mAx-]

the divergence should come from negative cells or high skewed cells.
But you are not suppose to get such warning in mesh motion since no physics equation is solved.
Are you giving a motion, or is the motion computed through an UDF.
Do you have the possibilty to use laering method instead of smoothing/remeshing?

[almostafa67]

i have to move center of a small circle on the edge of a bigger circle and for this i wrote an udf;small circle rotates at 4 rad/sec;i changed different parameters and iteration proceed 3700 iteration,but everytime error messages that i mentioned showed up!
you think for this motion i should set spring factor=1.0 ?

[kamal_g65]

hi
i have same problem, plz help me.
i read abovebut not useful for me.

[-mAx-]

Originally Posted by almostafa67

i have to move center of a small circle on the edge of a bigger circle and for this i wrote an udf;small circle rotates at 4 rad/sec;i changed different parameters and iteration proceed 3700 iteration,but everytime error messages that i mentioned showed up!
you think for this motion i should set spring factor=1.0 ?

I am not very experienced wit smoothing/remeshing, but more with layering.
you can try changing parameters...
Since I don't have Fluent, I cannot help you more than I did.
But once again: before starting to iterate, you need to be sure that your motion don't give you bad mesh.

[vmlxb6]

Hello,

I am facing the exactly same problem of negative mesh volume. My case is moving of a cyinder in the y direction only.I created the mesh in gambit. Initially before running the mesh in CFX, when I checked it in ICEM all the vol were positive but after about 5 iterations the calculations is terminated and error says negative vol detected.

I would appreciate if you could tell me how to go about. Can I repair the mesh or smooth it in ICEM.

Please help is needed. I would really appreciate any type of help.

[Subramanian]

Originally Posted by Jason
;134389

Is it axisymmetric? If so, then you're problem is that you're below the X-Axis. In Fluent 2Daxi model, you must use the X-Axis as your axis of rotation, and your geometry must all lie in the +Y direction (above the X Axis).

Hope this helps, and good luck, Jason

I had the same doubt. Your advice helped. Thanks a ton