The overall part is basically made up of four distinct components:
- The front and back halves of the "shoulder dome", created by layering Aves over styrene cross-section guides filled in with clay
- The "Spike Fringe" parts - the three are duplicates created from a single original (sculpted in Bondo) by resin casting. They are counter-sunk into the shoulder domes - one in front, one in back, one on the center plane.
- The spikes: these, too, are resin-cast copies of a Bondo original. Each is mounted to the "spike fringe" part via an allegedly 2mm styrene rod.
- The center plane - after the two shoulder dome parts were created separately, each was attached to a 1mm styrene plate in the shape of the shoulder dome's front-view cross-section for further refinement - and then the two plates were joined together and, most recently, largely cut away.
One of the basic problems here is establishing a frame of reference: this would have been a lot easier if I'd done it back when the two halves were separate, and each was attached to a whole center-plane plate. As it stands, I'm not sure I can reliably establish a consistent frame of reference.
What I'm trying currently is this: I rigged up a thick styrene plate with a tube on one of its edges such that I could put a long 2mm styrene rod through it - then I cut it out such that I could do this around the shoulder part - running the rod through the tube, through the front spike fringe, and then into the tube again on the other side. Then, with the shoulder part in place in this rig, I added a "floor" to the rig, aligned to the bottom surface of the shoulder part.
This is enough to get the elevation for the spike fringe, as well as its vertical tilt - but where I'm having a lot of trouble is getting the other two axes of its position, and the horizontal tilt (yaw) to also match. Being able to reliably locate the part's center plane along the "floor" would be a great help, of course - but I've already cut most of the center plane away. The front and back halves don't quite match, either, which makes the whole process tougher. Ultimately if I want this kind of precision I'll have to improve the back half to match the front...
Anyway, the whole process is rather frustrating. I've taken some photos of where I'm at with the whole thing - when I have a bit more time I'll post 'em and hopefully I'll have made some more headway on this problem, as well.