Not able to stitch surfaces together

[Quefelsees]

Hello all,

at the moment I'm trying to learn about form and surface modeling. So I tried to model a small UAV. The task at hand seems simple enough: create a fuselage (form), create a wing (surface) and stitch them together. I attached the project file.

However, the latter is not possible - even with a ridiculous tolerance setting (see image below). I can see some potential problem areas in the model (e.g. where the trailing edge of the wing meets the sharp edge of the fuselage) but I am not sure whether they cause the issue.

The end goal of course would be to have a shelled fuselage where all components like battery, etc. will be placed and a wing that's solid. The wing structure would be added later. The nose and aft of the fuselage can just be a plain patch for now. So is it just simpler to:

1. patch the fuselage and the wing to create a solid body individually and
2. to then combine the solid bodies,
3. splitting them again at roughly the wing root,
4. shelling the fuselage (which will probably fail because of the sharp edge of the fuselage) and
5. to then create fillets where they intersect (after recombining them)?

Or is an approach with a manually created fillet between wing and fuselage necessary?

As always, I greatly appreciate your time and help!

[dsouzasujay]

Hi @Quefelsees ,

There may be a better approach that more experienced users or Expert Elites can recommend.

In the meantime, here’s how I stitched the surface, as shown in the video below.

[TrippyLighting]

Here is a video explaining some problems already in the model. Those will interfere with continued modeling and need to be addressed now, before the timeline gets too long.

[johnsonshiue · Accepted solution]

Hi! As Sujay already pointed out, the issue here is due to non-manifold edges (an edge is shared by more than two faces). This geometric condition can lead to inoperable body. There is another way to resolve the condition. You may use Trim command to remove the unwanted portion. Select the wing surface as the tool and select the two faces on the body to remove. Then Stitch will work.

Many thanks!

[Quefelsees]

So first of all: thank you very much @dsouzasujay and @johnsonshiue for pointing out ways to stitch these surfaces together! I am going to mark the reply by @johnsonshiue as the "solution" since I find it a tiny bit more elegant.

--------------------------------------------------------------------------------------------------------------

And huge thanks to @TrippyLighting for the in-depth analysis of my model(s)! I greatly appreciate it! It's very valuable information to learn more about this topic.

I had to chuckle a couple of times because you pointed out some model quirks that you absolutely correctly pointed out as unintentional 😄 The crease that just vanished for example would have totally gone unnoticed to me. I would have probably written another forum post after hours of troubleshooting!

Some lessons learned

• Don't import airfoil data! Instead, import and calibrate a canvas of the airfoil and recreate it with the least amount of splines possible.
  • This was actually quite the surprise for me since I thought using plugins for importing airfoils was the right thing to do. I assumed that manually recreating them would be the kind of "unprofessional way".
• Finish forms before adding features to the timeline. Changes to the form after multiple features in the timeline after the form has been created have been added may result in huge problems.

Follow-up questions @TrippyLighting 

1. Box view mode:
   1. I kind of struggle to understand its utilization. When I have the model displayed in box view I am severely less able to model after my template. Maybe this gets better with experience but as of now it feels like a hurdle rather than a tool. I am probably lacking background in this field. I would appreciate if you could summarize what the benefits of modeling in box view are and why one should use it. The areas you pointed out in box view are indeed areas that would have gone unnoticed in smooth display. But to me box view seems like a tool to fix up the model at the end instead of using it from the beginning.
2. "Good" Curvature
   1. At the beginning of the video and towards the part where you talk about the airfoil you mention that the model has mostly "good" curvature but there are some areas that could be better. You also mention that the curvature of the imported airfoil splines is very bad. Could you please elaborate on that? What would this exact airfoil made from splines with "good" curvature look like?
   2. The reason I'm asking is that to me it seems like the purple areas in the curvature analysis represent curvature inflections. Is that correct? Because if it is, what makes it bad? To me there seems to be models that there would be no way around it, right? Take a rocket nozzle for example in its most basic form:
      
3. Symmetrical design
   1. You correctly pointed out that it makes little sense to model a form and then cut it in half only to shell it later on for which you need the second half again. However, for me there are some problems that I am not sure how to tackle correctly:
      1. In the front view (looking directly onto the nose of the fuselage), when I would only model one half, how would I make sure that the upper and lower point, where both halves would touch after mirroring, would not have a seem? In other words: how would I make sure that the curvature of the T-splines at the top and bottom would be continuous in the front view?
      2. How would I test for "shelling" this way? (In order to prevent discovering after hours of modeling that the shell fails and have to start from scratch)

Viele Grüße aus Braunschweig!