Converts APC propeller's PE0 files into BEM files

[Haoliang Yu]

Hello!
I discovered this program on GitHub, based on the Julia programming language, which converts APC propeller's PE0 files into BEM files. This allows importing them into OpenVSP for aerodynamic analysis. I want to use this method to calculate the performance of APC electric propellers and compare them with UIUC wind tunnel experimental data to validate the accuracy of VSPAERO in propeller calculations.
Link: https://github.com/jamesjhu/APC2BEM

I have started using it, and I find this tool to be quite effective. However, I feel that the generated shape looks a bit strange when compared to the actual images of the propeller. So, I'm sharing it with the group, hoping that everyone can give it a try in their spare time and see if it is indeed useful.
Additionally, I plan to further analyze the generated propellers using VSPAERO and compare the results with UIUC wind tunnel experimental data.

Many thanks

Knight

[Rob McDonald]

The instructions on the GitHub page say:

After importing the BEM file into OpenVSP, set the following Propeller ⎡Design⎤ properties:

• Construction X/C 0.000
• Feather Axis 0.125

Did you try that?

Rob

[Haoliang Yu]

Hello, Rob!
It's evident that you're right! I was too focused on modifying the airfoil later on and completely forgot about setting these two important parameters. You've given me a great reminder!
After setting them, the shape aligns properly, and I'm very grateful for your guidance. I will continue with the validation analysis.

Knight

[Joep de Boer]

I'm a bit confused as It seems that POE files contains rake in inches and this is then parsed to rake = 0  by the APC2BEM file why is rake omitted?

line 56 : rake_R = zeros(num_sections)

Thanks for creating the parser by the way 

Op vrijdag 19 januari 2024 om 07:49:18 UTC+1 schreef Haoliang Yu:

[Rob McDonald]

I suspect that James Hu just didn't need it and so did not put it in.

That said, it is quite possible that APC uses a different convention for defining rake than does OpenVSP.  If they differ, you can likely figure out a way to convert one definition to the other.

I suggest you fork APC2BEM and try to make the changes -- see what happens.

Rob

[Joep de Boer]

I played around with APC2BEM like you set and compared it to a optical 3D scan I found somewhere (I forgot but a quick search will get you there) of the APC 8X8E.

I found the best match to be when I did the following: 
    

    rake = apc[:,7]             # RAKE(Z) (IN)
    axial = -rake/R

    sweep = apc[:,6]            # SWEEP(Y) (IN)
    # NOTE: SWEEP IS DEFINED WITH (MOLD) LE PARTING LINE.
    tangential = -sweep/R

I also had some problems with the script i think maybe due to a change in structure of the table and white lines  in the APC files. I forked the repo here: https://github.com/JoepdeBoer/APC2BEM.git and made a pull request to the original. 

Results keep in mind the scan was roughly transformed as datum and axis did not line up , additionally I want everything in meters not inches and just used the scaling a few times on the stl to make it the right size and  lastly I did not put the right sections in. 

Really cool that APC provides this data and together with windtunnel data of UIUC :https://m-selig.ae.illinois.edu/props/propDB.html can do quick validation studies. I will only have time to some validation but maybe after october when my thesis is done will run them all. If anyone is intrested.

 

Op zondag 26 april 2026 om 20:25:58 UTC+2 schreef Rob McDonald: