CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Temperature at subdomain (http://www.cfd-online.com/Forums/cfx/113925-temperature-subdomain.html)

 sunilpatil March 1, 2013 03:29

Temperature at subdomain

Hello sir,
How to specify Temperature boundary condition at subdomain in CFX
Thank you

 evcelica March 1, 2013 20:35

I don't think you can, but you can volumetric heat source in the subdomain that is a function of the temperature difference from what you are trying to achieve.

HeatSource = SomeConstant*(TemperatureDesired - Temperature)[W/m^3]

I've found that you can't just insert some huge number in that would make the temperature change by more than the difference in just one time step

 sunilpatil March 3, 2013 01:18

domain boundary condition

ok thank you sir
i will try to apply heat source as a temperature diff in subdomain.

 sunilpatil March 3, 2013 01:28

subdomain Boundary condition

Hello sir,
actually i wanted to simulate effect of temperature variation on solar panels.
i have experimental data of ambient temperature on hourly basis i want to apply these values varying temp with time(Transient case)

Thank you sir

 ghorrocks March 3, 2013 06:24

It is not clear what you want to do, Sunil.

If the temperature varies over time then simply use this as a boundary condition. This is basic stuff and very easily done. If a region has a known temperature then use the approach described by Erik. Or are you looking to implement something else?

 sunilpatil March 4, 2013 01:07

Domain Boundary Condition

Hello Sir
i attached image file of my project here if we apply varying temperature at opening then while solving temperature will take some time to reach solar plate because of that i wanted to apply varying temperature to whole domain so that the real atmospheric conditions can be accounted(1 km domain is to account radiation and natural convection effect)

http://s11.postimage.org/4jxihndmb/Untitled.png

Thank you sir

 ghorrocks March 4, 2013 05:51

I think you will find the main thing driving the temperature cycle from day to night is the change in intensity of solar radiation. You do not seem to have radiation in there at all. Also air currents (ie wind) move the heat around. And finally if your domain goes 1km up in the air you will probably need to account for the thermal gradient in the atmosphere, heat transfer to higher altitudes and radiation loss to space.

I think you need to have a think about what is actually driving the temperature in your system. The system you drew does not even look close (which is why the simulation results are miles off).

 sunilpatil March 4, 2013 06:39

domain boundary condition

Hello sir,
i am having experimental data of solar heat flux on solar panel(0 to 24 hours) i can apply those values directly on the panel surface as a heat flux(W m^-2).and the domain around the panel will account for convection losses from surface to air(if its possible to apply varying temperature at the domain).and we are neglecting thermal gradient in the atmosphere and emissivity of ε = 1.0 for opening boundary condition as black body

Thank you sir

 ghorrocks March 4, 2013 17:55

This approach will not work. You cannot simplify the solar heat flux to simply a heat load. The heat load the solar heat flux results in is a function of the temperature of the object, its surroundings and the incident radiation.

I would recommend you:
* Make the domain smaller so you are just modelling the local area around the thing. This means you do not have to worry about differences in atmospheric conditions over altitude and other complications
* Apply a gentle wind so you can "blow" air across with the temperature of the daily temperature cycle.
* Apply a radiation condition to apply the solar radiation heat load, again with the daily variations included.

 sunilpatil March 5, 2013 01:10

Domain Boundary condition

K sir Thank you for your suggestion

 All times are GMT -4. The time now is 02:38.