CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Main CFD Forum (https://www.cfd-online.com/Forums/main/)
-   -   initlization problem (https://www.cfd-online.com/Forums/main/128027-initlization-problem.html)

santu December 31, 2013 11:02

initlization problem
 
hi to all

I have tried a model in 3d multiphase problem which cover +and - coordinate of y-axis.when I giving inlet velocity and initialize my problem, velocity convert to negative so flow direction change. I am not getting exact profile.

I have done my model in icem cfd.


Can anybody help me regarding my poblem.

Thanks in advance

laingmj January 8, 2014 08:38

I haven't ever had that problem before as the flow at the inlet should be specified as normal to boundary and therefore the flow should be in the correct direction.

If you are particularly concerned that the model straddles +ve and -ve coordinates then you could always translate the model in Fluent to make it all positive. I doubt however this will have any impact.

santu April 24, 2014 08:12

hello friend


I have a main pipe which is connected to two branch pipe. All three inlets i am giving velocity inlet or mass inlet boundary condition, but fluent taking one branch pipe as +ve velocity and other as -ve velocity or when initialize my problem, velocity convert to negative so flow direction change. I have tried the model in 3d multiphase problem which cover +and - coordinate of y-axis. I am not getting exact profile.

I think the velocity inlet i am giving to both the branch pipe is opposite in direction so the velocity what i am giving is fluent taking as opposite sign so the above problem coming. Is it correct then please give me a solution or suggestion what i will do.

Thanks Laingmj but if i am giving normal to boundary then also the sign problem continue.

Please help me regarding the above problem

rsslnt April 24, 2014 10:24

I am guessing that you are using velocity in normal to surface direction and if the normal of surface is not as you wish then the velocities would appear to be reversed. Why not play with the signs for the numerical velocity input, maybe it would align to what you need.


All times are GMT -4. The time now is 16:42.