Hey,

when computing the flow over an aerofoil at alpha = 0.5° with the "laminar" option switched on in Fluent, the streamline plots render a number of consecutive separation bubbles on both the upper and lower surfaces.

Initially, I thought that the first separation bubble may shed vortices, which would then look like more bubbles. However, when forcing transition at a certain point downstream of the first few bubbles, the "remaining" bubble structures disappear. Hence, I concluded that the large number of bubble structures actually are bubbles rather than shed vortices convecting downstream...

Has anyone else seen this phenomemon? Do you know how Fluent computes "fully laminar" simulations - what does it simulate instead of undergoing transition (as in reality)? Why is this large number of bubbles formed?

Many thanks for any input and best regards,

CC

Hi CC,

I've tried fully laminar simulations. But the solution didn't converge and my results also oscillating. I also noticed what you have experienced. Fluent user guide says that when the 'laminar' option switched on, it will disable the turbulence production. So it is not possible for the separation to occur during full laminar simulation(it is my assumption). When I plot the k, can see some kind of spike in the bubble area, other than that, the value of k almost remains zero. So I always use the full turbulence simulation and give some value in the source terms, which was able to predict the transition. In this case I can see the k value remains zero until the separation occurs and then k starts to increase.

Can you give some further details about how you forced the transition. I'm intrested in that part.

Cheers, Alagesan.

Laminar to turbulent transition bubbles occurs cause the adverse pressure gradient taking place on upper surface. Standard FLUENT versions are only able to hand fully laminar of fully turbulent calculations. Two years ago I was working on this subject, I found that the v2f turbulence model (which can be purchase and introduced into FLUENT) could hand the transition process (a turbulence sensitive factor is the friction coefficient, so is a good idea to see this factor in order to get the transition point, also is a good idea compare the transition point with Xfoil, it predicts the transition point quite well).

In my research also found a turbulence model called turbulece potential model, which model only the divergence of the reynolds stress tensor which is actually the term need by the momentum equation. This turbulence model can also hand the laminar to turbulent transition. I added this model to OpenFOAM so you could add this model into FLUENT trough UDFs.

regards.