This pre-release adds CST-Kulfan-based airfoil support and integrates NeuralFoil as a fast alternative to XFOIL-based polar generation.
The AirfoilEditor is a fast airfoil viewer, analyzer, and advanced geometry editor with integrated Xoptfoil2-based optimization. The app provides three operating modes:
- Browse and view airfoils in subdirectories
- Analyze curvature of airfoil surface
- Show polars generated using XFOIL or NeuralFoil
- Export airfoil to DXF format
- Repanel and normalize airfoils
- Adjust thickness, camber, high points, and trailing edge gap
- Blend two airfoils
- Set flap
- Generate airfoil replicas using Bezier, B-Spline or CST-Kulfan curves
- User Interface of Xoptfoil2
- Graphical definition of polar based objectives
- View results while optimizing
The app was initially developed to address artifacts found in other tools when using XFOIL geometry routines. The aim has been an intuitive, user-friendly experience that encourages exploration.
The app, developed in Python with the Qt UI framework, runs on Windows, Linux, and MacOS. On Linux and MacOS, View and Modify mode work directly after installation, while XFOIL-based polar generation and optimization require compiled external binaries when those features are used - see 'Installation' for details.
- Windows (recommended): download and run the latest installer from the GitHub releases page.
- Python package:
pip3 install airfoileditor, then runairfoileditor. - Linux/macOS note: View and Modify mode work directly after installation; XFOIL Polars and Optimize mode require compiled
workerandxoptfoil2binaries.
The AirfoilEditor utilizes various strategies to represent the geometry of an airfoil.
-
'Linear interpolation' – Using the point coordinates from the airfoils '.dat' file, intermediate points are calculated through linear interpolation. Used for quick previews and simple tasks.
-
'Cubic Spline' – A cubic spline is created from the airfoil's point coordinates, enabling precise interpolation of intermediate points.
-
'Bezier curve' – An airfoil is modeled using two Bezier curves, one for the upper surface and one for the lower surface. Nelder-Mead optimization is used to fit these Bezier curves to an existing airfoil profile.
-
'B-Spline curve' – Quite similar to Bezier using two (segmented) B-Splines of degree 4 to represent an airfoil (experimental).
-
'CST-Kulfan' – A compact airfoil-shape formulation that combines class/shape transformation with Kulfan design variables for smooth, controllable geometry.
The 'Cubic Spline' is applied for coordinate based airfoils to determine the exact position of the actual leading edge, which can vary from the coordinate-based leading edge defined as the point with the smallest x-value. Airfoil normalization rotates, stretches, and shifts the airfoil so its leading edge based on the spline is at (0,0) and trailing edge at (1,0).
For thickness and camber geometry operations, the airfoil is divided into two separate splines representing the thickness and camber distributions. A mapping spline—similar to that in XFOIL—is applied to shift the high point of thickness or camber. The airfoil is then reconstructed from the adjusted thickness and camber splines. This method also enables separate adjustment of the upper and lower surfaces' high points.
One of the major views in the AirfoilEditor is the airfoil's curvature. It provides quick assessment of surface quality and helps detect undesirable artifacts like a 'spoiler' at the trailing edge.
Tip
Have a look at the Xoptfoil2 geometry documentation for more information about an airfoil's geometry.
A 'curvature comb' can also be displayed around the airfoil as an alternative way to visualize the surface curvature.
Besides .dat files, the AirfoilEditor seamlessly handles .bez files defining Bezier-based airfoils, .bsp files for B-Spline-based airfoils, and .cst files for CST-Kulfan-based airfoils.
While traditional airfoils are defined by coordinate points, curve-based airfoils are defined by control points or compact shape coefficients. This approach provides inherently smooth curvature along the airfoil surface and makes the geometry easier to edit parametrically.
Bezier and B-Spline airfoils are especially useful for interactive geometry editing because the control points can be moved directly in the diagram while preserving a smooth shape. B-Spline-based airfoils were introduced in version 4.3 of the AirfoilEditor to gain more flexibility in airfoil optimization due to the local character of B-Spline control points compared to Bezier control points, which act globally by definition. However, achieving acceptable C3/C4 continuity between B-Spline segments is challenging and essential for generating smooth XFOIL polars. Therefore, B-Spline airfoils are currently in an experimental state.
CST-Kulfan airfoils use a compact class/shape transformation together with Kulfan design variables to represent the upper and lower surfaces with a small set of parameters. This formulation is well suited for smooth, stable geometry generation and is particularly convenient for rapid design exploration and for NeuralFoil-based evaluation. For CST-based airfoils, there is no dedicated match optimization in the sense of a Nelder-Mead curve fitting routine. Instead, a simple least-squares fit is used to recover the CST parameters from an existing airfoil. For the intended use in the AirfoilEditor, this is completely sufficient and keeps the workflow fast and robust.
In 'Modify Mode', control points can be moved directly in the diagram with the mouse to adjust the airfoil geometry. Each modification creates a new 'Design' with newly generated polars.
The match function (experimental for B-Spline) fits the Bezier- or B-Spline-based curve to an existing airfoil as accurately as possible using simplex optimization (Nelder-Mead) to:
- Minimize the root mean square (rms) deviation between the curve and target airfoil
- Align the curvature at leading and trailing edges to the target
- Avoid bumps of the curvature
- Ensure equal curvature at the leading edge on both surfaces
For CST-based airfoils, the equivalent step is a lightweight least-squares fit rather than a full match optimization.
Curve-based airfoils are especially useful for downstream work in 3D CAD, for example when building a wing from multiple airfoil sections.
The Export to DXF feature converts all supported airfoil representations to spline-based geometry suitable for CAD workflows. Bezier- and B-Spline-based airfoils are exported as uniform B-splines without loss of geometric precision. Since CAD systems also store curves internally as uniform B-splines, this format is well suited for lofted 3D bodies.
CST-Kulfan-based airfoils are exported in the same way as classic .dat profiles: they are converted to a spline representation for CAD import and downstream geometry processing. This keeps the exported geometry consistent across coordinate-based and parameterized airfoils.
Classic .dat airfoils are exported as cubic splines, which are widely supported and interpreted correctly by CAD software.
During export, the airfoil can be scaled to the chord length of a wing section and assigned a specific trailing-edge thickness.
The AirfoilEditor can generate polars using two methods: XFOIL-based polars through the Worker tool from the Xoptfoil2 project, and NeuralFoil-based polars for fast interactive evaluation.
The Worker-based workflow uses XFOIL to compute the polar data and is the reference method for higher-fidelity results. NeuralFoil is a learned aerodynamic model that predicts airfoil polars from the geometry and operating conditions. It is designed for rapid evaluation, which makes it well suited for interactive workflows where a polar should update immediately after geometry edits. For a more detailed description of the underlying concept, please refer to the NeuralFoil project page.
Compared with XFOIL-based generation via Worker, NeuralFoil allows near-real-time polar updates and is therefore particularly useful when modifying an airfoil and immediately checking the corresponding polar response. In general, the agreement with XFOIL is very good, especially when larger models such as xlarge are used. However, NeuralFoil polars can become slightly optimistic near cl_max, which is related to the CST-based geometric transformation used for the airfoil, where the nose radius can occasionally be represented too large.
For polar generation, the Worker's auto_range feature optimizes the alpha range to show the complete T1 polar from cl_min to cl_max. For T2 polars (constant lift), the range starts just above cl=0.0 up to cl_max.
In both workflows, a polar is generated automatically when it is needed for display. For XFOIL-based polars, the computation runs asynchronously in the background and typically takes about 1-5 seconds. For NeuralFoil-based polars, the calculation is performed synchronously and usually takes only around 10 milliseconds.
Each generated XFOIL polar is stored in an individual file using XFOIL's format for fast reuse. This method enables the sequential review of airfoils or airfoil designs, displaying the polars without requiring additional user input.
A polar can be 'flapped', meaning the airfoil has temporary flaps set before XFOIL or NeuralFoil computes the polar. This is convenient for comparing different airfoils at a specific flap angle without manually adjusting and recalculating.
Alternatively, a flap can be set in 'Modify Mode' for an individual airfoil and saved as a separate airfoil. This is useful when the flapped airfoil is needed for further processing.
When designing airfoils as wing sections along the wing span, they must be compared at different Reynolds numbers corresponding to each section's chord length. This is supported by defining a 'scale' value (in percent) for the reference airfoils of the main airfoil.
This allows comparison of wing airfoils using a single polar definition for the main airfoil.
Upon launch, AirfoilEditor opens in 'View Mode', which serves as the app’s default mode.
The 'View Mode' provides an overview of the geometric properties and polars of an airfoil. Since all airfoil parameters are read-only, there is no risk of making unintended changes to the airfoil file.
Using the mouse wheel on the airfoil selection combo box allows for efficient browsing of airfoils within a subdirectory.
Reference airfoils can be added to compare the current airfoil against other airfoils. This enables side-by-side analysis of geometric properties and polar characteristics.
The current view settings can be saved for an airfoil. When opening the airfoil again, these settings will be applied.
Modify Mode enables you to change the geometry of an airfoil.
In Modify Mode, many airfoil parameters can be changed by entering new values in data fields or moving helper points in the diagram.
A key feature is that every modification creates a new 'Design' version of the airfoil, saved in a subdirectory related to the original airfoil. This allows you to leave 'Modify Mode', re-enter later, and find all your Designs from the last session. At any time, you can step through the created Designs and compare the changes and their effects on the polar.
As the polar(s) of each Design is created automatically, it becomes easy to see how airfoil modifications relate to polar changes.
Tip
Adjust the camber highpoint position and observe its impact on polars at different Reynolds numbers. This approach helps you understand airfoil geometry interactively…
One of the possible modifications is to set a trailing edge flap: either permanently or to assess its influence on the airfoil's polar.
Note: A flap cannot be set on an already 'flapped' airfoil. The app remembers the initial unflapped design airfoil, enabling multiple sequential flap settings to be applied during a design session.
Bezier-based and B-Spline-based airfoils can also be adjusted in 'Modify Mode'. Since their geometry is defined by curves rather than coordinate points, traditional parameters like thickness cannot be changed directly.
For trailing edge gap adjustments on curve-based airfoils, new control point coordinates are calculated using a smooth interpolation function to preserve curve quality.
Instead, control points of the curves can be moved directly in the diagram with the mouse. Each modification results in a new 'Design' with newly generated polars, making it easy to observe how adjustments impact aerodynamic performance.
CST-Kulfan airfoils are also based on polynomial shape functions, but unlike Bezier curves they are defined on fixed x-stations. The profile shape is controlled primarily by the CST weights, which govern the vertical distribution of the curve along the chord. In the app, these weights can be edited interactively in a similar way to Bezier control points, but the adjustment is effectively vertical and acts at fixed x-positions. This results in a compact parameterization that still allows smooth and intuitive airfoil shaping.
In 'Optimization Mode', the AirfoilEditor serves as a wrapper for Xoptfoil2.
Xoptfoil2 is a particle-swarm-based airfoil optimizer that supports different 'shaping methods' to modify the airfoil during optimization:
- Bezier curves defining the shape
- Hicks-Henne shape functions
- B-Spline shape functions (experimental)
- Geometry constraints for defining allowed variations
The AirfoilEditor covers all steps needed for airfoil optimization with Xoptfoil2:
- Define an optimization case with the objectives and boundary conditions
- Run, control and watch an optimization
- Analyze the results
- Improve the specifications and re-run
Compared to manual editing of Xoptfoil2's input file, the user interface greatly streamlines defining operating points and objectives.
Multiple versions of an optimization case can be created, making it easier to select the best version at the end of your optimization sessions.
Important
Before you start your own airfoil optimizations with the AirfoilEditor, you should fully understand the key concepts of Xoptfoil2 and the special terms like 'seed airfoil' or 'operating point'. Please read carefully the chapters Getting Started and Airfoil Optimization of the Xoptfoil2 documentation. You will find the example of 'Getting Started' is ready to go in the AirfoilEditor making it easy to watch and modify your first optimization.
The main task when setting up a new optimization case is to define the 'operating points' on a (virtual) polar and choose the objective type for each of these operating points.
Within the polar diagram of the AirfoilEditor operating points can be added, deleted or moved with the mouse. A little dialog allows to enter additional specifications for the selected operating point.
If a different polar (e.g. Reynolds Number) is defined for an operating point, this polar will be automatically added to the list of polars and displayed in the diagram.
An individual weighting is visualized by the size of the symbol in the diagram.
In the lower data panel of the AirfoilEditor nearly all of the numerous options of Xoptfoil2 can be modified according to the needs of the optimization.
The button 'Input File' opens a text editor showing the current Xoptfoil2 input file that would be used for the optimization. You can tweak it with this editor (or an external editor) to cover special situations.
Once the definition of the optimization case is finished, the optimization is ready to go.
When an optimization starts, the diagram area is automatically maximized for full visibility of what's happening. Since Xoptfoil2 optimization runs as a background task, you can change view settings, pan, and zoom the diagram as needed during optimization.
After the optimization finishes, a new final airfoil is created. You can review the numerous Designs from the optimization process, analyze the final airfoil's properties, and if necessary, modify objectives and re-run. Creating a new version when changing parameters allows you to roll back to a better version if needed.
A ready-built Windows app including Worker for polar generation and Xoptfoil2 for airfoil optimization is available in the releases section on GitHub.
Download the Windows installer and run it to install the app.
If you already have Python version >=3.12 installed, it's advantageous to install AirfoilEditor as a package. It will start faster than the standalone .exe file and already includes Worker and Xoptfoil2.
Install the app:
pip3 install airfoileditor
To upgrade to the latest version, use pip3 install airfoileditor -U.
Run the app by typing airfoileditor on the command line.
If you want to try the app and ensure the installation doesn't affect other packages, you may prefer installing in a virtual environment. For daily use, a normal installation is more convenient.
The app is installed as a Python 'package'. Please ensure to have a Python version >=3.12.
For View and Modify mode, no additional compilation is required. NeuralFoil is included as part of the package and can be used for fast polar evaluation without any extra setup.
XFOIL-based polars and the optimization workflow require compiled worker and xoptfoil2 binaries. If these features are not needed, the package can be used without building those external programs.
pip3 install airfoileditor
To upgrade to the latest version, use pip3 install airfoileditor -U.
Run the app by typing airfoileditor on the command line.
To use XFOIL-based polar generation and airfoil optimization, the two programs worker and xoptfoil2 have to be compiled and made available for the AirfoilEditor by copying them into /usr/local/bin.
Please have a look into Xoptfoil2 README Installation for further information.
As a bonus for the extra setup effort, polar generation and airfoil optimization typically run 2-3 times faster on Linux than on Windows.
If there is warning message like "Failed to create wl_display" when starting the app, set a QT environment variable with export QT_QPA_PLATFORM=xcb.
If you want to clone the AirfoilEditor repository from GitHub for local development, install the following packages in your Python environment:
pip install "numpy~=2.2.0"
pip install "packaging>=24.0"
pip install "requests"
pip install "pyqt6>=6.11.0"
pip install "pyqtgraph>=0.14.0"
pip install "f90nml>=1.4.4"
pip install "termcolor>=2.3.0"
pip install "platformdirs>=4.3.0"
pip install "ezdxf>=1.4.0"
See CHANGELOG.md for history of changes.
I hope you enjoy working with the AirfoilEditor.
Tip
For Windows: Use the "Open with ..." Explorer command to associate AirfoilEditor.exe with the .dat extension. Then double-clicking a .dat airfoil file will open AirfoilEditor and allow browsing other files in the same directory. If you use the Python package version, create a small batch script to open .dat files.