Using the Standard Linear Ramps of the CMS Superconducting Magnet for Measuring the Magnetic Flux Density in the Steel Flux Return Yoke

The principal difficulty in large magnetic systems having an extensive flux return yoke is to characterize the magnetic flux distribution in the yoke steel blocks. Continuous measurements of the magnetic flux density in the return yoke are not possible and the usual practice uses software modelling of the magnetic system with special three-dimensional (3-D) computer programs. The 10,000-tonne flux return yoke of the Compact Muon Solenoid (CMS) magnet encloses a 3.8 T superconducting solenoid with a 6-m-diameter by 12.5-m-length free bore and consists of five dodecagonal three-layered barrel wheels around the coil and four endcap disks at each end. The yoke steel blocks, up to 620 mm thick, serve as the absorber plates of the muon detection system. A TOSCA 3-D model of the CMS magnet has been developed to describe the magnetic field outside of the solenoid volume, which was measured with a field-mapping machine. To verify the magnetic flux distribution calculated in the yoke steel blocks, direct measurements of the magnetic flux density with 22 flux loops installed in selected regions of the yoke were performed during the CMS magnet test in 2006 when four ``fast'' discharges of the CMS coil (190 s time-constant) were triggered manually to test the magnet protection system. No fast discharge of the CMS magnet from its operational current of 18.2~kA, which corresponds to a central magnetic flux density of 3.8 T, has been performed that time. For the first time, in this paper we present measurements of the magnetic flux density in the steel blocks of the return yoke based on the several standard linear discharges of the CMS magnet from the operational magnet current of 18.2~kA. To provide these measurements, the voltages induced in the flux loops (with amplitudes of 20-250~mV) have been measured with six 16-bit DAQ modules and integrated offline over time. The results of the measurements during magnet linear ramps performed with a current rate as low as 1-1.5~A/s are presented and discussed.