OpenVSP Git Model Repo

[deenri...@gmail.com]

In an effort to allow users to continue to share model files while the Hangar is down, I created a Git repo where people can contribute their vsp3 models. It's not the same, but its something. 

I threw in whatever models I had, but I welcome any pull requests. I also welcome any feedback to make this repo maximally beneficial to others. 

Git Repo

- Daniel

[Brandon Litherland]

This is a fairly open and honest community so I don't think you'll have problems with providing the library.  However, you'll run into the same issues as the Hangar that Rob has already mentioned many times.  Often it takes less time and effort to simply create your own model from scratch than to hunt down another user's mistakes in an existing model.  

There has been interest in providing a library of Validation models and data but these efforts are still very much in the ideation stage.  If people are going to add to your repo, I strongly recommend that you require documentation and some preliminary validation to go with it.

[Alan Teeder]

On 01/02/2024 16:02, Brandon Litherland wrote:
> This is a fairly open and honest community so I don't think you'll
> have problems with providing the library.  However, you'll run into
> the same issues as the Hangar that Rob has already mentioned many
> times.  Often it takes less time and effort to simply create your own
> model from scratch than to hunt down another user's mistakes in an
> existing model.

I fully agree with Brandon. The old hangar was full of unmaintained and
often buggy models. Many of them were pretty 3-D models, but were not
suitable for analysis with OpenVSP.

There are currently 3 examples in the distribution (directory
OpenVSP\vspaero_ex). Perhaps this should be extended, starting with
a basic simple wing-body-tail aircraft configuration to get folk
started. Examples answering frequently asked problems  ( panel method,
controls, propellers, structures FEM etc) could be added.

If there is a new hangar or git repo, any user added examples should be
reviewed before being included.

These are just my observations. I appreciate these ideas would add yet
more workload for the OpenVSP team, and finding resources may not be
possible.

While I am on the subject of the hangar, I found the NASA X57-Maxwell
example interesting, but could not get it run. What are the memory and
processor requirements for this model?

Regards

Alan

[Rob McDonald]

I'm happy to see the community hash out what they'd like to see in a Hangar replacement.

I might suggest that each model should be put in a separate directory with a unique file name.  This will allow multiple files associated with a single model to be kept together.

GitHub will display STL files in 3D.  While it won't maintain the color and transparency information that the Hangar could display via X3D files, it will give a view of a model that can be rotated and zoomed.

A README.md file in each directory can contain all kinds of information.  It should even be possible to point at the included *.stl file to display it 'inside' the README when you visit the directory.

You need to decide if contributors can choose a separate license per-file, or if all files in the repository must use the same license.  If the former, then the license can go in each directory.  If the latter, then you need to make sure the requirements for contributing a model are clear.

If all the files in a directory were named identically, then there could be a generic README.md file that would contain appropriate links.  Maybe something clever or more sophisticated could be done...

model_repo/

     f-16/

          README.md

          model.stl

          model.vsp3

     B747/

          README.md

          model.stl

          model.vsp3

etc.

Maybe that is a terrible idea because all the files will have the same name if you download many of them...

Maybe there is some magic way to render a website out of it using GitHub pages -- a simple script could create the required website that might tie it all together.  Someone well versed in the ways of GitHub pages could probably come up with a great solution very quickly.

I suggest coming up with a clever name for the site -- we've had the Hangar and the Ground School.  Perhaps this should be the Museum or the Flight Line or the Airshow....

I fully support the community coming together to figure out what you want to do for a generic model repository.  I am not going to voice an opinion -- and I am certainly not going to maintain it.

I am willing to 'nest' this repository under the OpenVSP GitHub username.  I will also give the maintainer(s) membership to the OpenVSP group as they need in order to maintain this site.

That said, I do have a project on the back burner that I need to work on over the next six months.  This will be a GitHub repository with example analysis cases (that include models).  These will serve both as Verification and Validation test cases (that will get re-run automatically using scripts) but also as documentation of how-to and best practices for various analyses.  I am calling it the Wind Tunnel (see what I did there?).  A lot of the cases will be VSPAERO centric, but the goal is to include Example / V&V cases for all OpenVSP analyses -- mass properties, wave drag, parasite drag, etc.

I am open to suggestions and help on the OpenVSP Wind Tunnel, but it is something that I will be heavily involved in and that I will commit to maintaining over the long term.

If you think you can trick me into making the Wind Tunnel into a Hangar replacement, I am wise to your ways and I will not be fooled.

Rob

--
You received this message because you are subscribed to the Google Groups "OpenVSP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openvsp+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openvsp/8fb81c76-76dd-4d0f-b313-781b9bbf66d2%40gmail.com.

[Mark Evan Salutin]

Good Day to All,

I agree with Brandon about the documentation for the VSP3 models since most of my data are taken directly from NACA/NASA reports. 

best regards,

Mark

--

You received this message because you are subscribed to the Google Groups "OpenVSP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openvsp+u...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/openvsp/9569efcd-d3e1-442f-85ca-4b412f03877an%40googlegroups.com.

[Tim Swait]

I like this idea, and have a hang glider model I can add, which is based on laser scan data and gives performance figures that are at least within the right ball park. I agree with the suggestion of having a folder for each model though, including a readme (including attributions and contact details for the author) and stl file as per Rob's suggestion. I'd vote for Airshow for the name (as in "show everyone what you've got").

[Haoliang Yu]

Hi,

Regarding the establishment of the hangar, I prefer the wind tunnel testing part mentioned by Rob because I have been interested in and continuously using OpenVSP to calculate things and do V&V like:

- APC electric propeller compared to UIUC Exp Data
 

- Simple wing:  https://ntrs.nasa.gov/citations/19980228368
- Propeller Blown Wings , such as Delft's PROWIM model

- Ducted fans placed above the wings

- And general aviation aircraft and other low-speed innovative aircraft.

- However, I'm not very familiar with mechanisms like uploading and merging branches on GitHub because I'm not good at coding.

In the future, I will try to organize my comparative results of calculations and experiments into a document and upload it. I hope it can be helpful to the community.

Best regards,
Knight

[deenri...@gmail.com]

Wow, did not expect this much attention on this. I will work on refining the repo given the suggestions by everyone.

I'll post back once I've made some updates.  

[Brandon Litherland]

Before going too deep on structuring your repo, I would take Rob's offer to make the repo hosted by the OpenVSP username.  For several reasons.

License and attribution absolutely comes into play here.  Anyone looking to leverage these results or models for professional work must be able to show proper use and permission.

Repo pull authority also matters.  From a NASA/Govt. perspective, we won't be able to upload files or publish to a personal account under any circumstances.  However, I've been able to publish to OpenVSP resources having a history with us.