Defining ‘trailing_flap_deflection’, and ‘trailing_flap_fraction’ for "linear" airfoils

[Rahmathulla P]

Hi Cory,

I used 'database' airfoils from AirfoilDatabase previously and it works fine with MachupX but was taking time. I'd like to try 'linear' airfoils but I've some questions related to degrees of freedom.

In AirfoilDatabse, to get the 'database' airfoil properties (to be used in MachupX), I defined the airfoil (4 digit NACA) as 'databse' and then used 'generate_database' and 'export_database' methods to output as a text file. In 'generate_database', I can define the ‘trailing_flap_deflection’, and ‘trailing_flap_fraction’ in the parameter, 'degrees_of_freedom'. In MachupX, when I set control deflections, it can extract the correct aero coefficients for the appropriate flap deflection degrees of freedom, in my understanding.

But, to get the 'linear' airfoil properties in AirfoilDatabase, I defined the airfoil as 'databse' and then used 'generate_linear_model' and 'export_linear_model' methods to output it as a json file. But in 'generate_linear_model', there is no option to define the ‘trailing_flap_deflection’ and ‘trailing_flap_fraction’ as degrees of freedoms. The output json file has only one set of values (perhaps for flap deflection of 0). In MachupX, when I set control or flap deflections, how does it get the coefficients for the deflected control surface?

Also, while generating the linear model using 'generate_linear_model', if I chose narrow limit for the linear portion of the curve (to accurately predict the slope), I get the CLmax low. 

I have attached airfoilDatabase input file for your reference. You can see dof is a parameter to 'generate_database' method but no dof parameter for 'generate_linear_model'. Could you please clarify this?

Thanks,

Rahmat

[Cory Goates]

Hi Rahmathulla

Excellent question. For linear airfoils, the flap deflection model is a semi-analytic (based on thin-airfoil theory), semi-empirical (based on some CFD data) model that accounts for flap deflection and size on top of whatever aerodynamic coefficients you supplied. So if you're using a "linear" airfoil model, you don't need to tell it anything about how it will behave with a flap. MachUpX will take care of that under the hood. I know that makes things a little limited, but that's the current state of things.

The low CL_max issue makes sense with how I've implemented things. I may look into changing that. For now, you can just manually update this CL_max value.

I hope that helps.

Thanks

Cory

[Rahmathulla P]

Hi Cory,

Thanks for the clarification. It really helps to understand what is going on inside. That means, I only need to specify alpha as the only degrees of freedom, right?

I have a related question regarding the linear coefficients AirfoilDatabase created for NACA 4412. I'm getting "aL0" as a positive value which I think should be negative. See the graph,

The attached is the python script (naca4412_linear_generator.py) that I used to generate the linear coefficients and I get the following as the coefficients.

{
    "aL0": 0.48004459770114954,
    "CLa": 6.2916121053165925,
    "CmL0": -0.08842209131888379,
    "Cma": 0.02938323870428498,
    "CD0": 0.008230774398761336,
    "CD1": -0.006715987975004134,
    "CD2": 0.006159507877676799,
    "CL_max": 1.0207874999999997
}

I'll update the CL_max, as you suggested. But I'm wondering what is happening with aL0. I'm creating the airplane json file using the linear airfoils created using the python script above (after necessary modification like CL_max and adding other parameters like geometry). See an airplane json file (platform0_airplane_test.json) attached.

Could you please have a look at the sign of aL0 and the process?

Regards,

Rahmat

[Cory Goates]

Hi Rahmathulla

Yep you should only need to define angle of attack when you're creating a database.

Good catch! I fixed the zero-lift angle of attack calculation. The updated code is on Github. Let me know if you have issues with it.

Thanks

Cory