[eishinsnsayshin]

Question:
I'm dealing with external flow simulations of objects such as airfoils/automobiles/etc moving through fluid that is static. I'm dealing with Reynolds numbers from say 500,000 to 4 million. I've heard the Transition SST model in FLUENT could be good for these simulations as it's pretty good at predicting transition. So for my velocity inlet and pressure outlet boundary conditions, I was wondering how I know what the intermittency factor and viscosity ratio should be.

What I understand so far is that the turbulence intensity for a simulation of an object moving through a stagnant fluid is very low (much less than 1%, lets say 0.05 or 0.1 %). I also understand that the turbulent viscosity ratio for external flows should be between 1 and 10. Any idea how I should select the "pick a number 1 to 10" number? So basically lets say I have a 1 meter chord airfoil going through fluid at 70 m/s (about 5 million Re#). What's the viscosity ratio that I should enter for the inlet and outlets, and how do you estimate this number? Is it from experimental data?

I also understand the concept of the intermittency factor. From what I understand, it's the probability that a certain point in your domain lies within turbulent flow. So 1 would be that everything is turbulent, and 0 would be that none of it is. For transitional flows (like the flow over the airfoil), upstream of the transition points have an intermittency of 0, and downstream of the transition have an intermittency of 1. So how do I estimate the intermittency factor? Do I have to know where exactly transition occurs? Isn't that the point of running a transition sst model?

Side question: Can the Transition SST model be used for compressible flows? If so, do the turbulence intensity, viscosity ratio, and intermittency factors change for the compressible case? How do I know how much they change by?

I'm really pretty new to the model and don't understand it very well so any (even partial) explanations of these concepts would be greatly appreciated!

[Far]

Side question: Can the Transition SST model be used for compressible flows? If so, do the turbulence intensity, viscosity ratio, and intermittency factors change for the compressible case? How do I know how much they change by?

Yes it can be. In fact the treatment of compressible flows and incompressible flows is treated on turbulence model level. The valid segregation for transition model would be bypass transition, natural transition, shock induced transition and wake induced transition and SST gamma theta model is good at all modes of transition.

I'm dealing with external flow simulations of objects such as airfoils/automobiles/etc moving through fluid that is static. I'm dealing with Reynolds numbers from say 500,000 to 4 million. I've heard the Transition SST model in FLUENT could be good for these simulations as it's pretty good at predicting transition. So for my velocity inlet and pressure outlet boundary conditions, I was wondering how I know what the intermittency factor and viscosity ratio should be.

There are very good research papers published recently in AIAA journal on external aerodynamics transition prediction (contact Josh on cfd-online for the copy) and results are compared to DNS. You should only be worried about the turbulence intensity, which has the dominant effect on transition and transition model.

I also understand the concept of the intermittency factor. From what I understand, it's the probability that a certain point in your domain lies within turbulent flow. So 1 would be that everything is turbulent, and 0 would be that none of it is. For transitional flows (like the flow over the airfoil), upstream of the transition points have an intermittency of 0, and downstream of the transition have an intermittency of 1. So how do I estimate the intermittency factor?

First of all, all transitional models deals with the boundary layer flows where transition occurs. In the rest of the domain flow is fully turbulent.

Do I have to know where exactly transition occurs? Isn't that the point of running a transition sst model?

No. You don't need to know the transition location a priori. This is the basic function of transition model the predict the exit transition location. There are certain methods developed by differnt researches to know the exact location of transiton from the data.
Search for the papers by Ralph J. Volino
http://asmedl.org/getabs/servlet/Get...ifs=yes&ref=no
http://asmedl.org/vsearch/servlet/Ve...TSEBV&aqs=true


http://www.google.com.pk/url?sa=t&rc...L_Kkzq8U-VtwUQ
http://www-g.eng.cam.ac.uk/whittle/p...T-LPReview.pdf
http://www.pa.uky.edu/~rplebeau/cfd/lp.html

[eishinsnsayshin]

Hey I really appreciate the reply! Thanks!

[Far]

Originally Posted by eishinsnsayshin

Hey I really appreciate the reply! Thanks!

We would be interested in your your research and results, as my knowledge is upto 2011 and may be you can bring new ideas and new concepts being discussed in 2012. So please do share the latest developments.