[CellZone]

Hello,

this question is more related towards panel code: so when can I use panel code over CFD simulation? So CFD captures all effects of viscid and compressible effect with flow separation. So it is valid over a broad band of velocities.

I am wondering why the panel methods are used for analyzing airfoils and aircrafts? So normally they fly at speeds much faster than Mach=0.3 , so compressibility plays an important role. So despite, it is used for a lot of codes like http://www.3dpanelmethod.com/home.html .

Is it just that for subsonic flight the calculation time has a much bigger effect than the errors one would make?

Thanks for your feedback
Cellzone

[LuckyTran]

It's just simpler and way faster for the task that it needs to do. From start to finish. You can write a panel code in a day.

It's also very easy to train one person to use a panel code. For CFD, first you gotta CAD your plane, which is its own training course. Then you gotta mesh it, which is its own training course. Then you finally get this thing loaded into your solver and it diverges, which requires its own training course. And then you come to cfd-online and spam the forum with questions about what under-relaxation factors you should use, what solvers you should use, etc. Finally you get results, now you gotta figure out how to automate the whole pipeline to even begin optimization.


I can also do CFD to figure out the pressure drop in a pipe. Or I can just look up a friction factor correlation and a calculator and get it in a few seconds.

[CellZone]

Okay thanks.

And what about the issues about incompressible flow, no friction ? So actually panel methods are not useable for aircraft problems (Mach > 0,3 )? Are they just used for academic research or for quick design parameter studies where exact values are less important?

Thank you

[FMDenaro]

Originally Posted by CellZone

Okay thanks.

And what about the issues about incompressible flow, no friction ? So actually panel methods are not useable for aircraft problems (Mach > 0,3 )? Are they just used for academic research or for quick design parameter studies where exact values are less important?

Thank you

There are some theories extending the results of incompressible flow to compressible one.

[sbaffini]

Panel methods are CFD, just another way to solve the flow problem when the assumptions and the resulting simplifications lead to a specific form of equations, amenable to be solved with the boundary element method. The exact same form of equations could be solved in a "traditional" matter, leading to the same solution (within numerical approximation). Indeed, Fluent does it to initialize the flow field in certain cases.

So, it's not the method but the assumptions: incompressible, inviscid, irrotational. Turns out that, for external aerodynamics, where you want to disturb the flow as less as possible, most of the flow field is indeed as such.

In particular, most such flows are attached, with viscous effects limited to the attached, and thin (because high Re), boundary layer. So, it turns out that most pressure forces can be reliably predicted as long as the hypotheses hold. And viscous effects can be added later (or iteratively) trough boundary layer methods.

As mentioned, compressibility effects, as long as the flow stays subsonic, can be somehow taken into account (e.g., trough the Prandtl-Glauert transformation).

When applicable the method is, indeed, very practical and effective.

[FMDenaro]

And, of course, panel method exists also for supersonic flows...