Both of the coil designs are two layer serpentine designs using “six
around one” cable. The skew dipole is the first coil to be wound as it
is the simplest from a coding and e/m analysis point of view.
Figure 1
Figure 1 shows the first layer of the skew dipole nearing completion.
Of note is the lack of harmonic correction spacers within the body of
the coil. This is made possible by the the two step wiring process,
which allows the two layers of the coil to be entirely independent of
each other. Previously, the nested wire type of design locked the second
layer of the coil into the same pattern already established on the
first. By isolating each, it is now possible to eliminate the harmonic
error correction function from the first layer, and incorporate it
entirely within the second, thereby speeding up the assembly process by
the elimination of additional g-10 piece parts.
Figure 2
The second layer of the skew dipole, figure 2, shows a far more
complex pattern. It is here that the harmonics for the 2-d cross
section, as well as integral field, are reduced to zero. Note that the
first layer start lead comes out from under the second layer, and has
been placed directly alongside the exit lead of the second layer. This
detail is required, as the contribution of the leads to the field cannot
be ignored for magnets of this short length and large radius. Both coils
are 600 mm in length, and 81 mm in radius. This coil will remain as
seen, until the completion of the combined function coil, at which point
all the second layer g-10 pieces will be applied and the entire package
over-wrapped with s-glass and cured.
The initial design of the combined function corrector required
careful attention to detail. Since it is a combined function corrector,
to be used in a combined function main magnet, the first issue to be
addressed is the coupling of the corrector to the main field. Quenches
occurring within the main cable magnet can dump significant energy into
the corrector, destroying it. To prevent that, the corrector must be
orthogonal to the field of the main magnet at the radius of the
corrector coils. Initial coding efforts starting with a quadrupole
serpentine are easily capable of adding low levels of dipole field.
However, the levels of quad and dipole for this magnet require dipole
fields close to 50 percent of the quad level. This called for a drastic
solution, creating the combined function magnet as a quad first layer,
with a dipole second layer.
Figure 3
Figure 3 shows the first, quadrupole layer, of the combined function
magnet after winding. Since the quad layer harmonics cannot be addressed
easily within the dipole layer, it has been designed for zero harmonics
on its’ own. This is evident by the introduction of several pattern
spacers. The quad layer start lead is taped within the pole, and will be
re-routed prior to second layer winding. Note the completed skew dipole
coil on the right of the figure.
Figure 4
Figure 4 shows the first layer of the combined function coil after
spacers and matched expansion epoxy has been added, but before the
second, dipole layer.The two thicker wire leads are the stabilized skew
dipole leads. They have been routed through the open section of a quad
pole, along with the start lead for the combined function coil. This
assembly and wire routing sequence allow both coils to be made and
connected without using any more radial space than that required for the
two layer coils.
Figure 5
Figure 6
Figure 5 and 6 show the completed second layer of the combined
function coil. To cancel the solenoid component of the first layer quad
serpentine pattern, the second layer dipole coil requires the same
number of turns. To maintain the correct quad to dipole ratio, the
dipole coil length has been adjusted, and is shorter than the underlying
quad. Of note is the large number of pattern spacers on the dipole
portion. This is required to provide the correct two dimensional cross
section required for low harmonics in the dipole layer.
Figure 7
Figure 7 shows the final magnet, with all coils completed. On the far
right side, the exit leads for both coils are visible; the larger pair
is the stabilized skew dipole, and the two smaller are the newly
completed combined function leads.
This support tube is now complete with respect to computer winding,
and will now be removed from the machine for fitting of the g-10 piece
parts on both coils, lead stabilization, final overwrap and cure, and
then warm measurements.
For more information contact
John Escallier
