CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   How to solve turbulent viscosity limited to (http://www.cfd-online.com/Forums/fluent/38082-how-solve-turbulent-viscosity-limited.html)

cxysohu October 3, 2005 02:45

How to solve turbulent viscosity limited to
 
turbulent viscosity limited to viscosity ratio of 1.000000e+005 in 99117 cells how to slover the problem

thanks

zxaar October 3, 2005 03:01

Re: How to solve turbulent viscosity limited to
 
search the forum with name zxaar and viscosity ratio, long ago i wrote a lot on it. might be help.

Another thing is do u get this problem as soon as you initilise or from the first iteration. If so, then initialise the k and e as k = 0.0001 and e = 0.001,

this shall not give you ratio warning if every thing is okey.


cxysohu October 5, 2005 04:58

Re: How to solve turbulent viscosity limited to
 
thanks your answer´╝ü some one have told me goto SOLVE....CONTROLS.....LIMITS .. and change the the maximum turbulent viscosity ratio to 1.0e10+07. do you think it fine the problem?

thanks


zxaar October 5, 2005 05:17

Re: How to solve turbulent viscosity limited to
 
well its not a good idea, because you see, as the viscosity ratio increases the effective viscosity increases, for two equation models the max ratio is mostly around 5000-10000, it could be more but usually it is around this. (effective viscosity affects your convergence very much). second issue is how fluent restrict the ratio, it does this by modifying turbulent viscosity and dissipation, that is k will be same, only epsilon will be adjusted for k-e model, now increase in limit means, dissipation max value is increased. Higher the dissipation (increasing to 10^7 etc) lower your chance of getting convergence with this model.

best bet is to restrict the ratio to maximum possible (realistically ) values. So if i aspect 5000 as max ratio, i would restrict it to 7000 max), its been observed that in some cases this approach result in covergence.



All times are GMT -4. The time now is 04:24.