Hi there.

We have been doing some simulations of a buoyant jet/plume discharged vertically upwards in CFX using the k-epsilon turbulence model. We have been using a turbulent Schmidt number of 0.7 (after Yimer et al. (2002); Can Aeronaut Space J 48(3), 195-200), and getting fairly reasonable results, but the tracer (buoyancy) spread was too low. We decided to try decreasing the Schmidt number to 0.4. However, when we analysed the results, all of the data we were plotting (velocity decay, concentration decay and the velocity and concentration spread rates) were exactly the same. Does anyone know why this would be the case? Are k-e simulations really not very sensitive to the turbulent Schmidt number?

Thanks!

Dear col16,

There are several Schmidt number throughout ANSYS CFX. Which Turbulent Schimdt number are you referring to? k? or epsilon? or the tracer's?

Opaque

The tracer's Schmidt number, sorry.

Hi,

The tracer's schmidt number just increases the dissipation. If the result does not change much as the Schmidt number changes then dissipation is not significant in your model, so other terms (probably convection) are dominating.

Don't forget that modelling jets is a challenging task for a turbulence model. There is lots of papers out there on modelling turbulent jets, I strongly recommend you have a look at some.

Glenn Horrocks

Thanks Glenn for your reply. Yes I realise jets are challenging to model. I have been looking at a number of papers, and I'm still continuing to make my way through them. As someone new to the CFD field, it's difficult to figure out which will have the most relevant information - the older papers with their original studies, or the newer papers with their latest information but much more narrow focus!

Over the last few days we've tried Schmidt numbers an order of magnitude bigger and an order of magnitude smaller, but the difference in output result has been very minor. The Schmidt number that we're wanting to change is the one in the transport equation. If you have CFX 11, searching for "the general form of the transport equation" (including quotation marks) brings up the page I'm referring to. Equation 136 has a variable Sct, which below is explained to be "the turbulence Schmidt number". We are quite sure that changing this value (e.g. by an order of magnitude) would have a substantial effect on the dispersion of the tracer. However, when I designate which of the two components in my simulation should be using the transport equation (under Domain - Fluid Models), the only input variable is kinematic diffusivity. Does anyone know where I can change the Sct for that transport equation? What I have been doing thus far is changing the turbulent Schmidt number under the buoyancy turbulence section, but from a good look at the help documentation, I'm pretty sure that's a different number.

Hope someone knows what I'm talking about! Thanks in advance for your help; it's much appreciated.