I'm using the "user function" to define some volumetric heat generation and I've run into a question.

When source is applied, you can specify in W or in W/m^3 which I believe is more correct to use.

Over what volume does CFX use? My model has only one particular geometry that sees the power input, so I'm afraid that if I don't separate out that particular chamber, the W/m^3 will give me the W/model volume^3.

Anyone doing this sort of thing?

Sources (including heat sources) are applied to a subdomain, which has a perfectly well defined volume. The "subdomain" region is the chamber that you have to separate out from the rest of your geometry.

When you enter the "total" source then it is assumed that the shape of the source per unit volume is a constant. So, the local volumetric source is just your total source value divided by the subdomain volume.

There is no option to specify the shape distribution when you give the total source.

Thanks, I was thinking it would be around the subdomain, but that leaves my with some tricky steps on creating a sub-domain around the chamber.

I'll tell what I'm doing in the hopes someone can tell me some precautions to look for. I gave the shape distribution via the interpolation. For every x,y,z there is a W/m^3. The energy deposition program (MCNPX) output file was organized to output MeV/(cm-source particle) which can be turned into W/m^3. So the aim would be to have some volume with an energy deposition (turned to heat generation) shape that describes the volumetric heat generation. I've done this, and the distribution of heat looks like what we expect, but there was some question about the validity of the input data and what volumes were being considered. Based on the information you gave me, I need to wittle down the entire volume into the chamber where heat is being generated. Thanks if the info!

-M