CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   cooling a solid body (http://www.cfd-online.com/Forums/fluent/44443-cooling-solid-body.html)

Ralf Schmidt April 15, 2007 16:45

cooling a solid body
 
Hi!

I have a hot body, that is cooled by a passing fluid. It should be a coupled and transient simulation - the fluid temp. increaes, the body temp. decreases (and finaly, the fluid temp. will decrease as well).

Now the question: how to set the BC??? I have allready done a stady state simulation, the body-walls have the initial Temp., the body is meshed and reaches steady state temp.

Are that the next right steps?

Turn the thermal wall BC (also the "shadow"-walls) to "coupled" and leave the fields for "wall thikness" and "heat generation rate" empty

Leave the solid body BC as it was (no fixed values, no source therms)

Turn on the unsteady solver (with sec.-order implicit)

Start to iterate. Data sampling activated (I what to know the temp. vs time curve). The max it. per time step is 20 by default - the steady sim. took about 3000 iterations - should that value match?

Are there any other ideas or hints???

Any help is highly appreciated!

Ralf


Ralf Schmidt April 16, 2007 11:54

Re: cooling a solid body
 
Hi!

The way I discribed the simulation seems to work.

But: the simulation took a long time - and there are convergence problems: all residuals vary and the continouty residual dosn't fall below 10e-1

Also, there is the warning "temperatur limited to xxx in x cells on zone x in domain x"

The statinary sollution was very good! continuity residual went to about 10e-6 and no scatter at all in the residuals.

Any idea, how to increase quality of the sim??

Ralf


All times are GMT -4. The time now is 22:55.