CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   convergence problem (http://www.cfd-online.com/Forums/fluent/44237-convergence-problem.html)

ammi reddy March 25, 2007 07:53

convergence problem
 
hi i am modeling a pulse tube refrigerator. In this project their is linear compressor which i modeled using dyanamic meshing and a UDF. but their is the problem of convergence. This problem is unstedy, axisymetric, segregated solver.it is closed domain so i cant take incompressible flow, under relaxation i has taken are vel=0.3

pr= 0.2

energy =0.8 please help if u have any sugestion.i am geting nearer to desire result but countinuty is not going below 1e-2 . plz help.

thanking u

ammi reddy


Seyed Farid Hosseinizadeh March 28, 2007 17:01

Re: convergence problem
 
Hey: Most of the time the residual for continuity around your numbe looks good. But if your results are suspended my advise as follow as below: 1- Chandge your number of mesh. Some time more, some times less. It depends. 2- Change your time step by calculating the dominat time scale in your physical system. You must know this number and choose a smaller value than this number(less than of half of it) 3-Change your underrelaxation factor to smaller one . Some times in the first itteration maybe you can choose 0.001. But you can change them when you get the better results in the process of solving. 4- Change your initial values to values that are reasonable nad make sense. 5- Check your non-dimensional Numbers in your system. Sometimes with choosing a lower number you can get the results and you can use this result as your initial guess for your exactly non-dimensional numbers.

Thanks/farid


All times are GMT -4. The time now is 07:58.