Sometimes I’m amazed at what I can do with Rhino. Other times I’m ashamed that even after around ten years of using it, I still can’t do everything I want with it. While I have drawn aircraft propellers in the past, this one (see attached photo) is a head scratcher, since the blades morph into, and are integral, with the hub. After a few days of studying it, I still haven’t a clue how to begin laying out the curves. If any Rhino ninjas out there have a suggestion or two, I would be most grateful.
1 Like
SubD 
1 Like
any further info for the profiles ? sections ?
some open cad / stl data ?
I found this (linked image to woodenpropeller.com. - i hope this is within the copyright.
(not exactly the same model …)
I agree with @martinsiegrist that it s tempting to use subD. but there might be a nice Nurbs approach as well - what s the final data used for ?
1 Like
Final data will be STL output, then to my resin 3D printer. Thanks for the drawing, that helps me visualize things a bit better. I was hoping to avoid SubD, since that’s another whole science to me, at the moment.
2 Likes
Dunno… might be worth a look…
-Pascal
2 Likes
Hi HAL9000,
i have experimented a bit with a sub-d approach. Maybe this could help as a first step to create the first volume, from which you can extract isocurves.
sub-d propeller.3dm (4.5 MB)
I have also attached the file if you want to try for yourself.
Kind regards
Sergej
Well I watched the video, but it was well beyond my level of understanding, as I have no idea what Grasshopper is, or how to use it. 
This looks very promising. Looks very close to what I need to accomplish. Thank you. I might now have to throw myself into learning SubD. Hopefully it’s not beyond my grasp!
You could use _Radiate instead of copy + rotate and stitching both parts…
1 Like
You’re welcome. It was a nice challenge.
Here’s one detailing a practical NURBS approach to connecting the blade and hub regions -
1 Like
Wow, that looks gorgeous. You should consider making this into a tutorial video for those of us who need to learn SubD (wink-wink!). Thank you so very much!
1 Like
Great video! Thanks so much for posting this!
Thank you for your kind words, but I am still learning too.
I think the available tutorials cover everything up about sub-d modeling. If you make your resin 3d print, please post some photos of the result, as I would like to see it.
I did not like the result, which sub-d → nurbs gave me:
I made some mesh refinements in another software, called GOM Inspect.
Now it is called ZEISS Inspect, as the company GOM (Braunschweig - Germany) was bought by Zeiss some years ago.
I still use GOM Inspect 2018, which has all the killer algorithms for mesh manipulation, like keep a sharp edge in your model while reducing the overall amounts of polygons. You can even remodel missing areas to some degree.
Companies like Steinbichler were angry about the fact that GOM gave their software away for FREE ! They did it to attract customers.
If anyone needs the installer without the hassle of registration at ZEISS, I can share the installer via wetransfer ( 600 MB).
1 Like
First, note that the hub is barrel-shaped. So model it as an oblate spheroid (American football) having its long axis parallel to the axis of rotation and extending 3 or 4 (or more) times the thickness of the propeller. Model it so that its “middle” part is a close approximation to your finished hub as shown in your illustration. Do all your modeling at full scale—you can shrink it later for 3D printing or whatever.
Model the blades at four to six stations from hub to tip using a simple and likely airfoil section for the era and nationality, such as a Gottingen 387(?) or a Clark Y or RAF 6. Calculate the blade section angles (“Beta angles”) by the known RPM and airspeed of the airplane. If unknown, guess something like 1800 RPM and 100 mph for WWI vintage (this isn’t intended to be accurate or flight worthy, eh?). Don’t bother with fancy or super-accurate calculations, just fix the angles to the nearest tenth of a degree. Draw lines inclined at these “pitch” angles at their respective blade stations. These lines are your “Beta-angle reference lines” or just “Beta lines.” Now layout your airfoil sections so that their flat bottoms are aligned on these Beta lines.
This particular propeller looks to have been designed using a flat, straight trailing edge line as the reference. So, use the (front of propeller) view to place these airfoil sections’ trailing edges along the trailing edge line. Use the (side of propeller) view to position these fore/aft if the trailing edge is not a “flat” line (parallel to the plane of rotation). Then, using the front view and perspective views, scale these airfoil sections along their Beta lines so that their projections in the front view appear to just “touch” the leading edge curve. This sets the chord length of each blade station.
If you did your airfoil sections correctly, they should be lightweight curves having relatively few control points, perhaps a maximum of about 16 for top and another 16 for bottom. You do not want them over-specified. Half that number would be better.
Okay, now create a leading edge guide curve and one or two trailing edge guide curves depending on whether the trailing edge is indented to be sharp or truncated. Sweep your airfoil section along the leading and trailing edge guide curves. You should obtain separate top and bottom surfaces. Sweep a flat closeout for the blunt trailing edge variant. At this point you only need one blade.
Old propellers like this had their hubs hand-blended using large-diameter drum sanding machines. Watch the Culver Propellers videos on YouTube to see how Alaina masterfully carves and sands these beautiful surfaces. Her drum sander must be at least four inch diameter (two inch radius), and nothing she makes has a surface radius even close to that small. So get out your fillet tool and set the radius to three inches or more and fillet your top blade surface to your football. Use a variable fillet if you wish. You’re going to fiddle with this a TON until it looks right so have at it for a while. Every prop is different. Then fillet the bottom surface similarly.
Sometimes the fillets just will not work. I’ve found that you can sometimes use the Pipe command (or other commands) to cut the blades and hub to where you think a blend surface might work. BlendSrf can work, but I’ve also manually built degree five or even degree seven input curves for a sweep using the surface edges resulting from the Pipe command. Do what it takes to make smooth transition surfaces from the blades to the football. Extend guide curves fore and aft as needed to make any sweeps long enough for later trimming. Trailing edges can be finicky, too. Do not be afraid to temporarily extend the blade surfaces aft so as to get the fillet or sweep to be long enough for later trimming. I’ve even extended leading edges to close a persistent gap in the fillets.
Your next task is to cut off the parts of this mess that protrude ahead of and behind the hub faces. Establish cutting planes parallel to the plane of rotation and use them to trim the front and back parts of the football and fillets that extend past them. Then use the cutting planes you made to close the fore and aft faces of the football and the inboard parts of the blades and fillets. At this point you could put a simple rounded tip on the blade so you don’t forget.
So far, you’ve got one blade. Use your Rhino skills to cut your hub in such a way that you have half a hub. I say this awkwardly because there’s a chance that one of your fillets extends around the hub a little past the halfway point, making a simple planar cut infeasible. You might just cut the football without cutting the fillets. Whatever, you now need to Array your one blade and fillets and half-hub to make two blades. Cut away the redundant inner football surfaces beneath the fillets and likewise for the blades. Fillet all “sharp” edges with a 1/4” or larger radius and you should be able to make a solid out of this thing by now. “Drill” and fillet your center bore and your bolt holes (mind the orientation or “clocking” of that bolt pattern!) and counter-bore the bolt holes for the drive lugs if the original was so equipped.
Alright, I hope this is helpful. Your question anticipates a video series I’m working on that isn’t ready yet, but I hope that these words frantically typed on my phone at bed-time will get you a useable product. I’m sure I left out some details….
One last thought: don’t be too concerned with surface continuity. The standard of manufacture for these old propellers was akin to tangent continuity, maybe not even that good. It was definitely not a visual criterion like Class-A surfaces or curvature continuity. Tangent continuity is just fine.
1 Like
Hi Sergej
Could I be one to receive the tool with killer algorithms?
Always in need😁
Thank you
1 Like
3 Likes
Thank you