The polarimeter is designed to be a 
polarimeter
for Hall C, operating at energies between 1 and 6 GeV, for a large range of beam
currents ( 
, higher beam currents can be envisaged at a later time).
To achieve this, the beam will be rastered over 
1mm in horizontal and vertical
direction, and the target will be continuously moved. At the same time, the layout
(see figure 9 is planned to have a 
geometry independent
on energy.
These boundary conditions, together with a maximal length of 12 m requires a setup
with two quadrupoles. The lower energy Møller electrons (large scattering angle)
get focussed by the first quad (which needs only little bending power) in order to
pass through the second one, where they are bent away from the beam. The higher
energy Møller electrons (at small scattering angles) are only affected by the
second quad and make use of its maximal bending power. With the two-quad setup a
separation of 50 cm between the beam and the 90 
CM Møller electrons is
achievable for all incident beam energies.
Figure 9: Setup with two quadrupoles which produces a separation of 50 cm between the
beam and the detectors with a total length of 12 m.
Two total absorption lead glass shower counters are used as detectors. Energy discrimination and left-right coincidence will ensure identification of a real Møller event. To check the optics and the settings of the quadrupoles, a hodoscope in front of the shower counters gives the exact position information of the Møller pairs. The hodoscope is available during all data taking runs and allows an online check for the quadrupole settings. The setup for the electronics of the two shower counters and the 16 channel hodoscope is shown in fig. 10.
Figure 10: Setup of the electronics for shower counters and hodoscopes, insert:
Hodoscope spectrum
The count rate used in the determination of the polarization is simply given by the coincidence rate between the shower counters, which is counted in a fast scaler. The hodoscope is read out only if a valid shower counter coincidence was detected and then stacked in two memories which stack 16K events before read out. The hodoscope positions of the Møller pairs produce, if the setup is correctly tuned, the event distribution shown in fig.10 (insert).
With the layout planned (Fe foil thickness 20 
, beam current 100nA) a
determination of the beam polarization to an accuracy of 
takes 12 minutes.
The setup to measure the Møller target polarization has been developed and is
running in Basel; studies to improve stability are still underway. The split
superconductive solenoid has been procured from Oxford instruments and is available.
The large quadrupole has been refurbished at CEBAF and is ready for installation by
late 93. The optics of the 2-quad system have been studied, and the electronics for
the readout of the detectors has been purchased. The mechanism to move the target
and additional rastering of the beam by 1 mm (for spreading the heat),
collimators, vacuum enclosures and the hodoscope detectors are being worked on.