[Mikevb]

I recently trialled FloEFD for use in modelling modifications to aircraft but struck a fundamental problem right from the first benchmark test. I have used naca windtunnel reports in the past to validate cfd modelling approaches/tools before committing to a project or product.

The first test for FloEFD was a NACA23012 2D model and despite fine tuning the mesh size and accuracy I couldn't get even close to the real world data - Clmax being well short of the test data and the Cl vs. alpha data also nowhere near reality. The moment coefficient data was very close in all cases tried for what it was worth though.

The feedback from mentor graphics was a little contradictory in that with sufficient mesh density the results should match reality but also that the boundary layer modelling wasn't ideal for external flows.

When a reference 3d wing was tried we achieved similar results.

A 2d supersonic inlet did work well (actually our next project) and I do love the tight integration with proe.

But if I have no confidence in the results...

Does anyone use this product for external flows and have validated their modelling with windtunnel or flight test data?

[Boris_M]

Can you write me a private message with whom you spoke from Mentor Graphics?
I work for MG and can check internally on your Problem and find out more.
Was it really someone from Mentor Graphics or from our reseller?

Regards,
Boris

[Chris Kriel]

Hi Mike,

I have seen some validation results of FloEFD on a NACA profile, with very good correlation of lift coefficients to experimental data. It is however important to specify an appropriate turbulence intensity.

Rgds,
Chris

[Chris_321]

Naca Profiles will work fine but if you want to look for more complex geometry it will fail because of the turbulence modelling.

[Boris_M]

Hi Chris_321,

this is not true, we have customers using FloEFD for civil airplanes, rockes & missiles and helicopters in external aerodynamics applications. Surely there are limitations to the turbulence model but that is not linked to the complexity of the geometry but the flow type. As for every code the main task in aerodynamics calculations is the stall region prediction.

Boris

[Chris_321]

In my company, we use flow efd for cooling systems and easy geometry, we did some analyses of flow efd for complex external simulation. After some simulations with complex regid bodies, Ahmed and naca profiles and comparing it to our windtunnel data, we came to the result that it is not good for ca, cl or detachment predicitions. My personal opinion is that flow efd have these troubles because it is just possible to make simple meshs. For rough calculation or just easy geometry it can be a good choice though.

[Boris_M]

It will certainly not be a case of the mesh type.
Certainly there is a difference in the applications of cooling systems and external aerodynamics, simply different requirements on the simulation and flow types but this has nothing to do with geometry complexity or mesh type.

Also external aerodynamics is not external aerodynamics. Considering race boats, cars, race cars, civil airplanes, helicoperts, rockets and reentry vehicles. All of them have different needs on simulations by boundary conditions, flow regimes and flow types (strong separations or very laminar etc.).

The key is to know which code to use in what application and at what stage of the design process.
You would wonder what our customers could tell you about the software for various applications and geometries copared to other codes.
I cannot tell what your external aerodynamics application was and who did the simulations and if you were guided by our experts. It is crutial to get guided in the beginning as I have seen often how people that used other traditional tools and meshing methods created way to big meshes where a fraction would give the same results in much faster time etc.

Regards,
Boris

[Chris_321]

Hi boris,
I can just tell from my experience with flowefd and compare it to ansys,ccm+ and powerflow. We did this analysis of using flow efd for external aerodymamic because we had good results from flow efd in pressure drop and flow distribution in piping systems and wanted to know if its possible to do external aero with efd. But we compared the results to our native cfd software and windtunnel data and came to the result that efd is a nice and fast software but not for external aero. Im interessted in some reports of projects where efd were used as external aerodynamic simtool. We use cfd for racecar engineering.
Regards

[Boris_M]

Hi Chris,
I understand completely which is why I wanted to emphasise that it is completely application and design phase dependent on if a tool is suited for it or not.
One of our customers "Voxdale" uses our tool for some time now and are very successful with it for external aerodynamics. They designed a speed boat with FloEFD that won the speed record of its class and also used it in race car aerodynamics. Others used it in re-entry vehicles from space and rockets where our results exceed the accuracy of the other tools and are used by experts that correlate it to measurements but then these are other flow regimes as you are working in.

What type of racecar are you designing, racecar as in NASCAR, Indycar or F1 or the street cars such as Ferrari etc.?

Of course in some simulations the task is to consider underhood flow and all the small details on the underbody, others simply take the simple external shape without the detail of underhood and underbody but still complex free form surface designs or a car.

Which product do you use exactly, FloEFD or Flowsimulation from SW?

I would like to exchange with you in more detail if you like and you can contact me in a private message so I can share some information with you on reports I have.

Regards,
Boris