Spalart-Allmaras and transition
 

Hi!

I am trying to compute a transonic airfoil testcase (RAE 2822 Case 9) with Spalart-Allmaras and transition set at 3% chord (by defining a laminar zone) and get an error message of the type

Error: > (greater-than): invalid argument [2]: wrong type [not a number]

Error Object: nan

immediately, no matter whether what options I use. On the other hand, computations with k-e work perfectly fine and yield good results. As far as I understand the manual, laminar zones are handled similarly for the eddy-viscosity models implemented.

I use version 5.5 on SuSE Linux 6.4 (Kernel 2.2.14-SMP). I applied the conditions and options according to the validation case that comes with the release; however, I tried both explicit and implicit solvers, but to no avail. The grid employed has about 20000 points and a low-Re wall resolution (Y+ approx. 1).

According to my expericence with other solvers, this case is not very critical, so that I am somewhat running out of ideas.

Any help is appreciated, thank you very much,

Martin.

Re: Spalart-Allmaras and transition
 

The k-e model does not model transition very well and o must have a transition set by setting a laminar zone.

the spalart-allmaras model on the other hand models transition well, and so often does not need a transion point specified.

i'm not sure if the Spalart-Allmaras model works with a forced transition within Fluent, and I can tell you that better results will be obtained with the spalart-allmaras model without forced transition than the k-e model with forced transition.

Re: Spalart-Allmaras and transition
 

What do you mean by forced transition ?

Re: Spalart-Allmaras and transition
 

Why K-e model dose not work well with transition it should work well even with laminar how about a case where you have turbulant in the inlet and then the flow slow dowan to transtion or even laminar. i benchmarked a flow in a pipe where the flow was in transition region there is only a problem how to defined your inlet intesity and hydrolic diameter if it is an internal flow and you must set up your mesh in a way that you can use two layer zonal model in the near wall treatment.

good luck