Let me clarify. I sorted out my problem with the tapered spokes. Yay me. There never was a problem creating a single solid using booleans.

See screenshot. The red arrow points to an area where the fillets need to be very heavy *between* the spokes but not as heavy at the top and bottom of the spoke. I'm sure it's a blend or network or something else I also don't know how to use.

3dm file lives here for your viewing pleasure: https://www.dropbox.com/s/kkldadi27wf7y0t/NYC%20999%20tender%20wheel%20v2a.3dm?dl=0

Thanks

Chris

Hi Chris,
Very nice wheel.
May I ask where you found the "blueprints"?
Signed by Buchanan, they appear to be originals circa 1893, with some notes from 1919 and 1933.

Presumably when the 86" drive wheels we replaced with 70" wheels, the smaller wheels were changed also.
Spoke count must have changed also?

- Brian

I looked at some photos on the Internet and there is definitely a web structure joining the spokes on each end. I think the way it all blends together will make it difficult for NURBs modeling - perhaps Sub D is the way to go on this one.

I attempted to make an approximation by creating filler pieces to go between the spokes. Maybe with some additional effort the filler pieces can be sized correctly and made such that they blend into the spokes more gracefully.

Ed Ferguson

Nice poster of the 999. (Original size wheels).

https://www.ebay.com/i/353260486381?chn=

Also found a beautiful scale model, but it only had 7 spokes.

- Brian

Hi Chris,
perhaps this method may help.
After boolean union spokes to hub select the edge and join.
Add circles sized to the radii you need and sweep joined edge curve.
Trim Hub and blend.
Note when I produced the hub I rebuilt the curved section
so the surface was complete (no seam lines) this helps the blend operation.
Cheers
Barry

Barry, you are WAY better at Moi than I am. LOL

Is there any way you could show me the process step by step? This is EXACTLY what I need.

Thanks

Chris

Hi Chris,
you can just use the fillet tool as shown in the photo.
I did the collar trim to give a variable fillet as I was unable to get it with the fillet tool.
I'm basically trimming the solid with a collar produced by sweeping desired fillets.
The sweep path is the edge where spoke joins the hub.
If your still unsure post me your curves and I will mark up what's needed.
Cheers
Barry

Edit:
It's possible to do it with the fillet tool with variable fillet my problem was filleting across the seam of the hub.

Hi
have you try ONSHAPE to do the bevel ?

@ Barry-H, how did you get the sweeped circles to meet the spokes and hub at a tangent with your first technique? Or are they just roughly positioned close enough to tangential to suffice?

Hi,
the circles are snapped to the edge to be filleted.
I used 2 small circles front and back & 2 larger circles on the sides.
Then sweep the edge curve to produce the trim collar.
The blend operation produces a tangent fillet.
Hope this helps.
Barry

Ah, the sweeped circles are just to trim the part, then a blend is actually used for the fillet, ok I get it now. Thanks. At first I thought you somehow managed to sweep between the 2 different sized circles in a way that the resulting torus-like form stayed perfectly tangential to the spoke and hub all the way around the sweep...

DanC I'm glad you get it. I'm still confuzzled. LOL