Supersonic_Zero - lift folder missing

[Sebastian Leblic]

Hi all,

I have recently been trying to employ the "Supersonic_OpenVSP_Wave_Drag" function and have run into a few issues, namely that the "vortex_lift.py" is missing; the entire "Lift" folder is missing from "Supersonic_Zero" module. I have found it on an older version of SUAVE and have placed it in my "Supersonic_Zero" folder to fix the issue.

I'm just wondering if that "Lift" folder was removed for a specific reason (although I could not find the "vortex_lift.py" file anywhere else) and re-adding that file is not the present solution. If not, I would just like some confirmation that I should be adding this back into my library to be able to use the "Supersonic_OpenVSP_Wave_Drag" function.

Thanks,

Sebastian

[ebot...@gmail.com]

Hi,

vortex_lift should have been removed from that analysis. We upgraded the VLM with SUAVE 2.4, which now has a built in corrections for leading edge suction that arise from vortex lift. We found the new in VLM correction to be much more accurate.

For now, when you setup your mission analyses, set compute.lift.vortex to SUAVE.Methods.skip

-Emilio

[ebot...@gmail.com]

One other thing. If you do expect to get vortex lift from a wing (highly swept delta wings), you must set: 

wing.vortex_lift = True

[Sebastian Leblic]

Hi Emilio,

Thanks for getting back to me. Those settings have been updated now. Just saw the concorde tutorial suggesting this - wish I'd gone to look at that beforehand...

Anyways, a quick side problem I've been trying to fix is this:

Not sure if I'm supposed to go ahead and define those settings as well, or if something else needs to be setup to avoid this issue. I have replicated what is setup in the Supersonic_Zero file and then run into some further issues with a " 'NoneType' object is not callable " error.

Any advice here? At this point, I'm wondering if I'm going about my setup all wrong as every time I fix one issue, I run into another one (all seem to be minor, but tedious nonetheless). If I just need to keep powering through, then I will - I just don't want to be going down an unnecessary rabbit hole.

Thanks again,

Sebastian 

[ebot...@gmail.com]

Yeah, that looks like that was also not updated. This function should look just like this.

I'll put a fix in for it now.

[Sebastian Leblic]

Okay sounds good.

I have another error that I can't seem to fix. This one I am not seeing how to approach it best:

For a refresh on context: I am trying to use the OpenVSP_Supersonic_Wave_Drag analysis for the concorde vehicle; the file coming from the regressions folder. I've printed the wave drag coefficient and see that it is empty, printing a set of empty brackets. This is the code surrounding the issue:

I've commented out a few lines to focus on the issue itself. Any further help would be much appreciated.

Thanks,

Sebastian

[ebot...@gmail.com]

Hmm. I started taking a look, these scripts are way out of date. It looks to me that the vehicle file was never written. Maybe VSP has nothing to compute? It doesn't seem like the WaveDrag analysis is run, hence no results (the error you're getting).

I'll see if someone who actually works with supersonics can have a look.

[Sebastian Leblic]

Yes that makes sense - it's definitely not creating a vehicle file. Thanks for looking into this more!

[Sebastian Leblic]

Hi Emilio,

Just wondering if anyone was able to look into the problem I'm having with running higher-fidelity analysis (higher than fidelity-zero).

Let me know if more details or a refresh of the problem is needed.

Thanks,

Sebastian

[Tim MacDonald]

Hi Sebastian,

Looking at your results here I think fixing the missing vehicle write step would probably solve the issue. You can do this by calling the vsp write function when you set up your vehicle. It's not included in this script directly because writing the vehicle to OpenVSP for every condition would be unnecessary and very time consuming. I haven't personally used the script in quite a while, but let me know how it goes.

As an FYI, there have been some changes to the wave drag output format since the supersonic scripts were written. This shouldn't matter for API calls, but if you find scripts that directly extract from text files, you may see related errors.

Tim

[Lukas Programme]

Hello Tim,

I also tried to integrate the function Supersonic_OpenVSP_Wave_Drag() into the concorde tutorial to later use it for my own airplane. As I first got errors like "AttributeError: 'Data' object has no attribute 'use_surrogate' " I defined the attributes after calling the function. I also call the vsp write function at the end of the vehicle setup.

    aerodynamics = SUAVE.Analyses.Aerodynamics.Supersonic_OpenVSP_Wave_Drag()
    aerodynamics.geometry = vehicle    
    aerodynamics.settings.use_surrogate = True
    aerodynamics.settings.propeller_wake_model = False
    aerodynamics.settings.use_bemt_wake_model = False
    aerodynamics.settings.initial_timestep_offset = 0.
    aerodynamics.settings.wake_development_time = 0.05
    aerodynamics.settings.number_of_wake_timesteps = 30
    aerodynamics.settings.model_fuselage = False
    aerodynamics.settings.model_nacelle = False

However I get this error message (see screenshot). The function drag_div defined in Methods.Aerodynamics.Supersonic_Zero.Drag.compressibility_drag_total indeed just takes 4 input parameters but in Methods.Aerodynamics.OpenVSP_Wave_Drag.compressibility_drag_total 5 input parameters are given.

Do you have an idea how to fix this?

Thanks,

Lukas

[Emilio]

Hi Lukas,

Looks like you found a bug. This file has diverged a little from the original. I think all you need to do is delete k as an input here. k is no longer used.

I'll make an issue on github.

-Emilio

[Lukas Programme]

Hi Emilio,

still about the integration of the function Supersonic_OpenVSP_Wave_Drag - I work on the develop branch.

I deleted the k however I still get the following error message:

CL = conditions.aerodynamics.lift_breakdown.inviscid_wings_lift[wing.tag] doesn't work anymore in the wave_drag_lift function from the folder Methods-Aerodynamics-Common-Fidelity_Zero-Helper_Function. Do I need to define inviscid_wings_lift somewhere?

Lukas

[Emilio]

Hi,

I've made some fixes in this branch here. Let me know how it goes.

-Emilio