Hi! I'm looking for references (papers, articles, etc.) about the fan model (a concentrated pressure variation associated with the normal local velocity and some empirical coefficient) in conjunction with the SIMPLE algorithm on collocated grids. The pressure loss in the radiator model is largely the same think. The fan model is present in the version 4.2 of Fluent as an additional momentum source term associated with some user defined control volumes. This approach is simple but not very realistic because the fan zone is then modeled as a cell zone. The model was strongly improved in the version 5.1 (or higher) of Fluent by associating the concentrated / semi-empiric pressure variation with the interface between two cells (edges or faces). The model is more realistic but the numerical treatment must be completely different. Unfortunately there are no reference given in Fluent User's Guide. I must now develop a fan model in an unstructured 2D SIPMLE algorithm and I cannot find any trace of some references. I would be very grateful for any help and piece of information. Thank's in advance!

Mihai ARGHIR

(1). Why don't you do two things: (a). measure the conditions in front of and behind the fan for some RPM's, to derive your own model. (b). Model the 3-D flow over the fan (just one blade sector flow field is enough) for some RPM's, and derive the model numerically. (2). The method-(a) is actually easier to do, but even with the method-(b), it is still fairly straight forward for your applications. (3). Anyway, for a model to be realistic, you will have to do some testings, or 3-D modeling first. For accurate measurement, a fan test cell would be ideal, where you can measure the RPM , mass flow and pressure conditions. Fan manufactures use the test cell results in their fan specifications, without the detailed flow field information.