(In collaboration with W. Flegel PPE/EC)
The central ALEPH magnet
is complemented on both sides by a dipole magnet with a horizontal field of B =
0.7 T orthogonal to the beam. We propose to re-use
the flux return yoke of the UA1 magnet for these two magnets,
see Fig. 3.2. The magnets will
be arranged as the existing UA1
magnet, but each with half of the available number of yokes. To adapt to the
required length of the new magnet, two short flux return yokes have to be
added. This requires completely new windings which, in addition, have to
provide a large opening of 1.8 m diameter around the beam axis at the end
closest to the
interaction point (see Fig. 3.3). At the opposite end, the beampipe
is of 
20 cm diameter, it fits between the two halves (see
Fig. 3.3) and
needs no special shaping of the winding. A support and moving system
similar to the one for the UA1 magnet has to be constructed for each magnet.
Figure 3.3: A view of one-half of the UA1 magnet, which will be located
next to the central ALEPH magnet.
The new magnets will consist of two halves which are mirror symmetric about a vertical plane containing the beam axis, as is the case for the UA1 magnet. Each half consists of 4 rectangular, C-shaped flux return yokes of UA1 plus an additional one to adapt to the length of the new magnets. The opening of the C-shaped yoke is facing the beam. Each yoke is constructed of 16 low-carbon steel plates of 50 mm thickness and separated by 17 mm gaps. For UA1 they were instrumented with scintillators for hadron calorimetry, but they will not be so equipped for FELIX. Each magnet will be placed onto a support system which permits one to horizontally separate the two magnet halves orthogonal to the beam, in order to gain access to the magnet volume for detector installation without dismantling the beam pipe. In addition, in its open position it has to allow for the axial displacement of the ALEPH end caps.
The new coils were calculated and designed to obtain the required acceptance
angle and the necessary bending field. In order to avoid a strong saturation
of the iron in the return yoke, a field value of B = 0.7 T is envisaged. For
reasons of cost, aluminium is
proposed as the conductor material with an acceptable
current density of s = 3.2 A/ 
.
Finally, the pressure drop of the cooling water in
the underground hall of ALEPH has to be kept well below 10 bar.
These considerations then lead to a similar conductor cross-section as in
UA1. Each magnet half will contain two coils with the same number of turns.
The coils near
the beam axis on the side facing the interaction point provide a large
opening of 1.8 m diameter for particle acceptance.
Fig. 3.2 shows in a perspective view the C-shaped return yokes of
one-half of a magnet half
and its pair of coils in its position next to the ALEPH magnet.
Each coil is built of 7 pancakes, each with two layers of 5 turns.
A list of the parameters of the proposed magnet layout is given in Table 3.1.
Table 3.1: Parameters of the proposed UA1 magnet.
One way to reduce the cost of the magnet is to re-use the existing material and equipment from the UA1 magnet. Additional iron is, however, needed to adapt to the required length of the new magnets. Because of their complex shape, the coils will be a major cost item. A first optimization of the costs for coil manufacture versus operation cost has been performed. The resulting current density determined the choice of the conductor cross section. The following costs were assumed for this calculation:
The combined field of the ALEPH solenoid and the UA1 dipole magnet at nominal current has been calculated with the TOSCA [2] programme. For these calculations, the main field component of the dipole magnet is in the x-direction, that is in the horizontal direction orthogonal to the beam axis. The main field component of the ALEPH solenoid is in the z-direction, that is the beam axis. Fig. 3.4 shows the horizontal component of the field. In the dipole region, the asymmetry in the field is due to the coil asymmetry between upstream and downstream currents. The interference between the ALEPH solenoid and the UA1 dipole has also been checked. Figs. 3.4 and 3.5 show the x and z field components, respectively, resulting from both magnets. In the region of ``ALEPH cap & UA1 coils'' the main fringe field effect in the z-component is from the dipole, while in the horizontal direction, the biggest effect is due to the radial component of the field in the solenoid. In this simulation, the shims and grooves, currently made in the ALEPH end-cap in order to reduce the field inhomogeneity, have not been taken into account [1].
Figure 3.4: 
component for the total field: the dotted curves correspond
to different emittance angles from the interaction point in the horizontal plane; the
solid curve is the field along the z-axis.
Figure 3.5: 
component for the total field: the dotted curves correspond
to different emittance angles from the interaction point in the horizontal plane ; the
solid curve is the field along the z-axis.
