Hello everybody,

I recently discovered OpenFoam in the Internet.

In my company we are developing passive, high-power microwave components. We are interested in the simulation of heat transport problems in both solids and fluids, incl. convection. The heat is generated due to ohmic losses by the "skin effect" on the surfaces of conductors. We know the values of the power density (W/m^2) using electromagnetic solvers.

I would like to know whether OpenFoam is suitable to simulate these kinds of problems, whether it allows such boundary conditions on internal surfaces. The main questions are:

* How can one generate the 3D-mesh most easily and affordable using .SAT or .STL files describing the geometry? We were trying the Harpoon mesher from Sharc...

* Are there any graphical tools to set the boundary conditions, like adiabatic walls or heat transfer coefficients on outer surfaces, and heat generation densities on inner surfaces?

* How can one import the heat generation densities (just a file with x/y/z position and power density for all surface mesh points) into OpenFOAM?

We are also interested in commercial support, if it is affordable...

Best regards,
Martin Lorenz

OpenFoam is suitable to simulate these kinds of problems, in fact it can do the electromagnetics as well.

Harpoon is not a bad choice of mesher, you might also want to consider ICEM-HEXA from ICEM or GridGen from Pointwise.

FoamX will allow you to set boundary conditions on a patch-by-patch basis provided you have regioned the boundary appropriately in the meshing tool. If this is difficult there are patching tools in OpenFOAM which will allow you to re-patch the boundary.

Importing the heat generation densities will be trivial if the order of the list of values corresponds to the order of the patch faces. If not some kind of mapping procedure will be required.

If you are interested in obtaining commercial support from OpenCFD please contact us directly at enquiries.

Just for info: Icem is now owned by Ansys.

Hrv

Henry, you said that it can do the electromagnetics as well: Please consider that the electromagnetic simulation runs at a timescale of nanoseconds whereas the heat propagation has a time constant of seconds to hours!
We would need a two-step simulation, first electromagnetics then thermodynamics...

Sure that's no problem. There are various ways in FOAM of coupling two equation systems and they need not have the same time-step as long as they are synchronised. Also while OpenFOAM includes an MHD solver it's basic and only includes simple boundary conditions but I should think it should be extendable to cover your problem. If you would like to discuss the options of getting OpenCFD involved in your project please contact us at enquiries.

I have been using OpenFOAM for a few weeks now and have noticed that the thermal energy does not seem to transport beyond the boundayr layer nodes attached to surfaces by wall functions. Not sure how to attach images, but the main issue is that the air temperature of flo going over a pipe parallel to the flow only heats up in the first layer of cells and does not 'diffuse' into the air beyond those cells. Please provide any insite. Thanks.

[gschaider]

Craig, please provide us with information which solver you are using (for instance I'd be very surprised if icoFoam transports any energy at all ...)

I have tried buoyantSimpleFoam and rhoSimpleFoam. The velocities look good, but the temperatures are basically the same no matter what inlet velocity I use and the buoyancy effects in buoyantSimpleFoam are not calculated properly either. Any ideas?

[eugene]

Check whether you are using "type wallBuoyantPressure" for your wall pressure boundary conditions and that gravity is set correctly in "constant/enviromentalProperties".

I have and I still am only getting the heat to transfer into the surrounding cells and not the cells that are several away from the surface of the pipe.

[amitshah]

I have used laplacianFoam and finding similar behavior. The temperatures die out at the boundaries and the internal temperature does not change.

[hjasak]

It works for me (of course!). You have either messed up the boundary conditions or specified zero conductivity.

Hrv

I guess I don't see where to specify conductivity. There is nowhere to find it. Any help would be appreciated.

[eugene]

constant/thermophysicalProperties

Not sure what all those numbers mean though.

[gschaider]

Me neither. But I found Chapter 8 of the UserGuide very helpful.

[oscar_j]

Hi everybody,

I wish to deal a typic thermodynamics problem with Open Foam; heat transfer by conduction, convection and radiation. I'm not sure if it's possible because would be necessary to implement a mesh for solid elements and other mesh for fluids, ¿That's right?. In addition, how definite various materials for different sections.

Thankfully,
Oscar

[hellorishi]

Hello All,

I would like to simulate a heat-transfer problem:
- one inlet with 298K air
- Pressure outlet condition: ref 1atm
- outer circumference wall with 573K
- some inner walls to split the inlet flow, like some kind of shell & tube heat exchanger.

I would like to know whats the difference between rhoSimpleFoam(RSF) and buoyantSimpleFoam(BSF).

My current understanding & questions is:
- BSF has buoyancy. If I give g=(0,0,0) is BSF same as RSF?
- BSF requires pd as a BC. which is the best BC instead for
-- adiabatic walls? fixedFluxBuoyantPressure ?
-- inlet and outlet?

Thanks for your time,
Rishi