It would be much easier for this station to be redone from scratch.
The kinds of mesh corruptions done by boolean union functions are too much for a human being to clean up.
Why would anybody do this?
Smells of Wings to me...
And it doesn't make sense that someone would ruin a mesh by pressing a single button
and then I should have to spend days cleaning up the disaster.
Anyways, I added a new, clean torus and I did some cleaning up of the other stuff; but there's days of clean-up
work still to do.
There's like ultra-thin triangles hanging from other geometry from a single vertex, in many places (left),
pairs of co-planar, overlapping triangles (right)...

There's ultra-thin triangles along edges, with normals facing in random directions, edges meeting at a T (or H

Too much.
I uploaded the new blend file with the new torus. The old torus is still there, moved it to layer 11 for those wanting to be
horrified by looking at it.
http://wcjunction.com/temp_images/rlaan ... y_v2.blend
IMO, it would take a lot less time to model this again from scratch.
Note: I often recommend welding features to the main geometry, but as I said many times,
1) It's only justified when the features cover significant areas. Spikes and tiny things do NOT cover significant areas; therefore they
should NOT be welded.
2) When welding is done, it should be done manually and cleanly; NOT using boolean ops.
In fact, ideally, meshes should be made entirely of quads; without a single triangle. Boolean operations produce bazillions of triangles, --in
addition to corruptions--, and so they should NEVER be used.


