CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   conjugate heat transfer, plz help (http://www.cfd-online.com/Forums/main/104789-conjugate-heat-transfer-plz-help.html)

 sepidehkavousi July 16, 2012 07:47

conjugate heat transfer, plz help

hi my dear friends
I am modeling a micro channel heat sink by FV method...my thermal boundary conditions for heat sink is adiabatic except for bottom wall which is constant heat flux.
I used ghost cells and ta the interface and for ghost cells for solid I reached the equation using continuity of fluxes at interface and for gost cells of fluid I used continuity of temprature but when I solve temprature of solid and fluid does not change much by the heat flux that is meant...I don't know if I did some thing wrong...plz help

 Rami July 23, 2012 06:17

There is no need for ghost cells in weak formulation methods like the FEM and the FVM. In these methods - by application of the divergence theorem - the BCs are treated by the appropriate terms at the boundaries (e.g, the div(q) term in the FVM is given as integral over the cell surface of the normal flux, which is given as natural BC on the boundaries).

 Rami August 16, 2012 05:46

Quote:
 Originally Posted by sepidehkavousi (Post 371612) hi my dear friends I am modeling a micro channel heat sink by FV method...my thermal boundary conditions for heat sink is adiabatic except for bottom wall which is constant heat flux. I used ghost cells and ta the interface and for ghost cells for solid I reached the equation using continuity of fluxes at interface and for gost cells of fluid I used continuity of temprature but when I solve temprature of solid and fluid does not change much by the heat flux that is meant...I don't know if I did some thing wrong...plz help

Quote:
 Originally Posted by Rami (Post 372954) There is no need for ghost cells in weak formulation methods like the FEM and the FVM. In these methods - by application of the divergence theorem - the BCs are treated by the appropriate terms at the boundaries (e.g, the div(q) term in the FVM is given as integral over the cell surface of the normal flux, which is given as natural BC on the boundaries).

dear rami
I do not under stand what to do in BC. I should pay attention to cnvergance of fluxes at boundaries, but I do not know what to do. to cells near boundary what should I do?
I wuld be very happy if you help me
thanks my friend

Please post or drop me an e-mail with the geometry, the solved equations, the heat sink and the BCs you wish to solve, and I'll try to help you.

 All times are GMT -4. The time now is 16:13.