Help with mismatch in CeRAS aircraft model.

[Calioth]

Hi everyone. I am new to OPENVSP and VSPAERO, and I am currently working on a project that utilises the CeRAS CSR-01 model (https://ceras.ilr.rwth-aachen.de/tiki/tiki-index.php?page=CSR-01&structure=CeRAS). However, after creating the model and used the VSPAERO analysis, there seems to be a mismatch with the lift distribution, drag polar graph as well as the L/D polar graph.

 

One of my guesses right now for the mismatch is because my aircraft model is not trimmed so, that might impact the L/D graph, but I could not explain the reason why there is a mismatch on the other graphs. I have attached my .vsp3 in here. Any help/hints/suggestions would be greatly appreciated.

Thanks in advance!!!
Calix

[Brandon Litherland]

With the information that you've provided, it's difficult to really understand what you did to set up your analysis to produce your results.  When I ran your model as-is, I got some very odd dCP on the horizontal tail that caused a very poor solution.  

For starters:

* Did you include the fuselage and engines in this analysis or just the lifting surfaces?  It's almost always best to start with a single wing and find some reasonable results before moving on to a full aircraft. You're using Sets, which is good.  Start simple and build up from there.

* You're running fairly large alpha sweeps at transonic conditions.  VSPAERO is essentially a potential flow solver with compressibility corrections that can solve for supersonic flow in VLM (Thin) mode but will not give very reasonable results with Thick components at supersonic speeds (for now). Transonic conditions are problematic for many solvers and VSPAERO is no exception.  Try to recreate the results at M 0.5 and see what you get.

* I suggest increasing your wing resolution (Num U, Num W, clustering) to try and capture more of the pressure changes along the wing.  I suspect that your wing is a bit too coarse to get a good load distribution.

* VSPAERO has some parasitic drag (CD0) included in the results but this is essentially a corrected flat-plate model based on the NACA 0012 for wings and a regular flat plate for bodies.  You may have better luck adding in some drag from the parasite drag tool.

* VSPAERO does not model separation but does include some experimental stall modeling.  You should find that the stall effects will significantly alter your results at higher alpha.

Work with these suggestions and then report back with your results and some more detail about how you set up your analysis. Then we can help you more.

Good luck!

[Calioth]

Thank you so much. Let me try your suggestions and see what the results are.

Calix

[Calioth]

One quick question before I send out the results after changing the model.

After I get the parasite drag from the parasite drag analysis section, could I just add that into the CDtot in VSPAERO analysis? Also, what is the difference between the CD0 in VSPAERO analysis and the parasite drag in parasite drag analysis?

Thanks in advance!

[Brandon Litherland]

CD0 in VSPAERO is a "simple" skin friction drag method for either bodies or wings.  The parasite drag tool is MUCH more extensive in the ability to work different drag formulations, atmospheres, grouping, etc. See the Ground School videos on the topic to learn more (https://www.nasa.gov/reference/openvsp-parasite-drag-tool/).  VSPAERO's CD0 term is lift-dependent, you will find.  This is because there is a correction to the parasitic drag component that also accounts for viscous behavior of a NACA 0012.  My recommendation is often to perform an alpha sweep to locate the minimum CD0 from VSPAERO, subtract that minimum value, then add the result from the parasite drag tool.  I've posted in the group about this several times, I think, so you should be able to find my other responses on this topic.

[Calioth]

Thanks very much!