How low is the reynolds number in turbulence to use the low reynolds number source terms? 100 maybe 50 maybe 1000? what tests has phoenics done for this case to compare with experiments? which model is best for jets and plumes?

Low-Reynolds-number turbulence models are designed to account for the effects of molecular viscosity on turbulence near walls. Therefore such models are valid throughout the viscous, transition and turbulent regimes in near-wall turbulence.

As an example consider the Lam-Bremhorst k-e model to see how wthe model coefficients depend on turbulent Reynolds number. The model differs from the standard high-Reynolds-number form in that the empirical coefficients CMUCD, C1E and C2E are multiplied respectively by the functions:

FMU = [ 1.-EXP(-0.0165*REYN) ]**2 (1.+20.5/REYT)

F1 = 1.+(0.05/FMU)**3

F2 = 1.-EXP(-REYT**2)

where

REYN = SQRT(KE)*YN/ENUL

REYT = KE**2/EP/ENUL

where YN is the distance to the nearest wall. For high-turbulence Reynolds numbers REYN or REYT, the functions FMU, F1 and F2 multiplying the three constants tend to unity.

This model and other k-e variants plus the Wilcox-Kolmogorov model have been tested extensively against near-wall data. Many example Q1 input files can be found in the advanced-turbulence-model and core libraries of a PHOENICS Installation. Some example applications are reported on CHAM's website at:

http://www.cham.co.uk/phoenics/d_pol...flows/cava.htm

http://www.cham.co.uk/phoenics/d_pol...ws/impinga.htm

Note that such low-Re models are not formally valid for free flows at low turbulent Reynolds numbers, such as the far-field axisymmetric wake where the turbulent Reynolds number decays with distance from the origin.

Which model is best for jets and plumes? It depends on the precise configuration. For example, even the standard k-e model is useless for a simple axisymmetric free jet unless it is modified. The standard model also fails to predict the plane wall jet correctly. Thus, universality cannot be expected of a turbulence model. Plumes usually require buoyancy corrections to the standard k-e model, but this area has been well researched and successful publications abound in the literature using PHOENICS and other software.

Thanks you.

(1). The answer is excellant. "very high quality " answer. Thank you very much. (2). This answer can be used to "measure" the user's understanding of the turbulence modeling. (3). I would like to add one more item: the book by David Wilcox " turbulence modeling for CFD " is an excellant book for anyone in CFD field.