The BES III detector consists of a helium-based, small-celled drift chamber, time-of-flight (TOF) counters for particle identification, a calorimeter of thallium-doped caesium iodide CsI(Tl) crystals, a super-conducting solenoidal magnet with a field of 1 T, and a muon identifier that uses the magnet yoke interleaved with resistive plate counters (RPCs). Figure 3 shows a perspective view of the detector.
The wiring of the drift chamber has been completed, and the assembly of the chamber has started. Beam tests of prototypes have been performed at KEK and IHEP with electronics prototypes, and both tests show that all design specifications have been satisfied and that the single wire resolution is 110 μm. CsI(Tl) crystals are being produced by Saint-Gobain Crystals, by the Shanghai Institute of Ceramics, and by Hamamatsu (Beijing). More than two-thirds of the crystals have been delivered, with satisfactory light yield, uniformity and radiation hardness. A beam test shows that the electronics noise from the preamplifier, main amplifier, charge digitizer and 18 m of cable was less than 1000 electrons equivalent per crystal, corresponding to about 220 keV of energy.
The scintillator and phototubes for the TOF system will be delivered before summer. All the RPCs for the muon identifier, made of bakelite but without the linseed oil surface treatment, have been manufactured, tested, and installed (figure 4). The average dark current and noise level for all chambers installed after one week's training is 1.6 μA/m2 and 0.095 Hz/cm2, respectively, for a high voltage corresponding to an average efficiency of 95%.
The BES III superconducting magnet is the first of its kind built in China. The vacuum cylinder and the supporting cylinder are made in China, in collaboration with the Wang NMR company of California. The wiring of the superconducting cable and the epoxy curing, the assembly and testing were all done at IHEP, with advice from experts all over the world. The superconducting magnet coil has now been successfully installed into the detector (figure 5) and will be cooled soon.
