Well, I got two quotes on the custom chamber last week. One was about
where I projected the cost to be, the other was about half again as much as the
first quote. Obviously, there’s quite a range in the market place.
Got another 3 bars milled out. Much faster this time, but I really need
to cut smaller lengths. Trying to mill more than I can clamp at one time
is for the birds. As the piece gets milled, the clamp holding the milled
end torques a little so the milling angle is off a little. Nothing
hideous, but I spend too much time trying to compensate for this during the
clamping that I’m just going to cut the bars and take the waste under the
clamps. No biggy.
out the cost of having a custom chamber built got me thinking seriously about
the geometry. Up until now I’ve been just trying to duplicate Bussard’s
original cuboctahedron. But I want to investigate a magnetic cage geometry
with 5 fold symmetry. If there’s anything you learn from polyhedral
geometry is that 5 fold symmetry doesn’t mesh with 3 or 4 fold symmetry.
Thus there is no common chamber configuration for the radially directed ion and
electron guns – they are either/or. So I can’t really do both.
So I found an Archimedean solid, the
shown on the picture to the right. This polyhedron has the correct
property in that each vertex is surrounded by 4 faces. Strangely enough
(well, not really that strange) the icosidodecahedron is one of two convex
quasiregular polyhedra. The other is the
Thus I’m changing plans. The new CAD repository can be found
chamber, shown on the right, also now uses bar magnets rather than the ring
magnets. As you can see, the cross sectional area required by the magnets
is quite a bit smaller than with the rings. Most of the magnet is
protected by the steel cage, although a small amount of each magnet is exposed
to the resulting fusion products (i.e. hot and angry ions).
I’ve colored the poles of the magnets blue and red to make it clear how the
magnetic fields are oriented. I think it looks pretty cool and will work
out better than the cuboctahedron.
I had to redesign the vacuum chamber. I’m getting much better at using
Inventor, so things went much quicker. Now if I can just get it to stop
crashing on me… I have figured out the tell tale signs of a crash – at
least for now – and I can close and just restart when things are going south…
The nice thing about the chamber is that the edge lengths are now 3", rather
than 4.5". Naturally, I’ll need more of them. But I can recycle all
the pieces I have, plus I can recycle all the pieces that didn’t make the cut –
well, all except the ones I bent. All told I need 120 edges. More
milling. ‘sokay as the custom chamber has a lead time of anywhere from
four to eight weeks. I can spend that milling.
So I hope you weren’t expecting anything dramatic real soon If I had
5 grad students, scheduling would be a bit easier to contract. <heh>
Now I know why mad scientists always have an Igor or three helping them out with