Stuck on a workflow to design a robotic fish tail

[nils_tack]

I am currently designing a fish robotic analog as part of my research. The project was going well until I needed to design the tail. The tail is a separate element that will later be bolted to the caudal peduncle (posterior section of the body). The horizontal cross-section of the tail lobes was chosen as a standard NACA0012 hydrofoil. The thickness of the chord is wider near the peduncle and gets thinner closer to the tip of the lobe (see 'Additional images' for reference). If only designing a hydrofoil, a simple guided loft would probably suffice. However, the flat, vertical cross-section of the peduncle must have a precise shape/dimensions consisting of a 3mm exterior offset from a 37x12mm ellipse. Ideally, I need the (symmetrical) superior and inferior tail lobes with a NACA0012 chord profile to seamlessly transition into the thicker caudal peduncle while ensuring the caudal peduncle maintains the desired shape and dimensions.
This is where all my trouble started. I have tried many combinations involving solid lofts and surface lofts with interior and/or exterior guides. I have tried to model an entire lobe at a time, only one side of the lobe followed by a mirror and even serial lofts, but none of the solutions produced acceptable results. I have come close in some cases, but the curvature map and section analyses revealed slight imperfections (i.e., concave leading edge). Additionally, I couldn't figure out how to appropriately include the NACA0012 hydrofoil cross-section. I essentially left the loft features' figure out' an approximation. But this is not what I want.
I have been stuck on this for a couple of weeks now, so I decided to seek help/feedback from the Fusion 360 community. I generally design simpler 3D objects, so I may be unaware of a specific workflow that would enable designing this fish tail. I am attaching a file for reference, including the main sketches that define the overall shape and dimensions of the tail (superior tail lobe only). I'm also including screenshots of some common issues I found when running curvature and cross-section analyses.

Thank you all in advance for your help, suggestions, and assistance with this project!

NBT

A visual reference of how the superior lobe of the fish tail is constructed. The CAD does not include any of the (many) attempts I've made to generate a solid/surface.

[davebYYPCU]

I don’t have access to your file, atm.

second pic with tangency at the panel lines would be ok?

when selecting body edges, Tangency condition is available.

So 3 lofts, top fin, and the other two from front and back transition piece.

Might help….

[nils_tack]

Hi davebYYPCU,

Thank you very much for your prompt reply. To answer your question, ideally, I do not want that tendency line seen in pic 2 as it breaks the curvature (though, not very dramatically). This picture was a composite of two lofts using this line as a guide. First, I lofted the vertical front part (peduncle) to about 1/3 of the top horizontal profile and then performed a second loft using the posterior sections of the horizontal bottom and top profiles using the edges of the previous loft as a guide. While this works and allows me to loft all the way up to the tip of the lobe, I still end up with slight surface artifacts. However, it is the closest I've gone to achieving the results I wanted, and your reply suggests it might be a good compromise to get where I want. I will explore your solution further and see how it improves things!

P.S. Is the CAD file not accessible/downloadable, or are you having issues opening the file?

[TrippyLighting]

You'll get much better and faster responses if you embed images directly into a post. 

On a Windows computer, you can paste it directly from the clipboard or drag it from a folder into a post.

Drag-&-drop also works on a mac.

[davebYYPCU · Accepted solution]

My access problem is due to not being near a desktop whilst travelling.

Maybe you didn’t understand my suggestion, 

your 3 loft process is spot on, the use of body edges with tangency, and not sketch curves being selected is the way to go, your continuity creases will be eliminated with appropriate constructions.

When selecting a body edge, the option - tangency is in a drop down where is says Connected.

Search the forum for dummy body for mirror.

Might help…..

[nils_tack]

Thank you for your clarification. I did not realize that selecting body edges vs. sketch curves could make a difference. I suppose I was a bit naive for thinking they'd act the same. I'll go through the ref you suggested and give it a spin.

Thanks again for taking the time to help with this.

[nils_tack]

Thanks for the suggestion, will do!

[nils_tack]

Your suggestions solved my problems. The tangency option got rid of the continuity crease. The issues I had with the surface analysis showing erroneous curvature were due to the inconsistent thickness of the cross-sectional profiles of the upper lobe.
Thanks for the help!