What Happened with Spectrometer Magnet 2B

Abstract

The spectrometer solenoid is supposed to be the first magnets installed in MICE [1],-[4]. This report described what happened during the test of the MICE spectrometer solenoid 2B. First, the report describes the temperatures in the magnet, the cooler top plate and the shield during the run where the magnet quenched at 258 A. During this quench, a lead between the bottom of the HTS leads and the diode bank burned out causing the magnet to quench. Second, three methods for measuring the net heat flow into the cold mass are described. Third, there is a discussion of possible resistive heating in the HTS leads between liquid helium temperature and the copper plate, which is at about 50 K. Fourth, there is a discussion of the measured first stage heat loads in the magnet, when there is no current in the magnet. The first stage heat load calculations are based on knowing the first stage temperatures of the three two-stage pulse tube coolers and the single stage GM cooler. Fifth, the estimated heat load to the first stage when the magnet has current in it is discussed. Sixth, there is a comparison of the stage 1 heat loads in magnet 1A [5], magnet 2A [6], and magnet 2B [7]. Finally there is a discussion of recommended changes for improving the spectrometer solenoids so that the coolers can keep them cold. IntroductionThe second version of Magnet 2 (called magnet 2B in this report) was tested between 15 March and 31 March 2010. The cool-down of the magnet was successful. The magnet had a number of training quenches starting at 169 A. The first low current quench was believed to be due to power supply issues. On 19 March 2010, the five coils in series were trained to a current of 258 A (94% of the magnet design current). During this quench, one of the low temperature superconducting leads for coil M2 burned out or was disconnected. The break in the lead occurred between the voltage tap inside of the cold mass (near the diodes) and the bottom of the HTS lead. It is not clear whether the quench was caused by the lead break or whether the quench caused the lead break. The liquid level in the magnet cryostat during the quench test was low (less than zero on the upper level gauge). The magnet cryostat was about half full. Depending on the location of the lead break, the fact that the liquid level was low may have been a contributory factor in the break of the M2 lead.The three-coil set (E1, center, and E2) was subsequently quenched at a current of 270 A. (This is ~99% of the design current for the three coil set.) After the three coil set was quenched at 270 A, the magnet cryostat was completely filled up. The magnet cryostat was kept full during the period of time that the people from LBNL were at the MICE collaboration meeting in Riverside. From 20:25 on 22 March until 8:50 on 29 March about 250 liters of helium were boiled away, at an average cryostat pressure of 1.17 bar. The average boil-off rate during the period of 156.4 hours was between 1.58 and 1.68 liters per hour. During this period, the tank was always k...

Using a single-stage GM cooler to augment the cooling of the shields and leads of a magnet cooled with two-stage coolers

“…The Cu plate temperature near the coolers, the estimated ΔT between the Cu plate and the cooler first stage, the first stage temperature and the heat flow into the cooler first stages during tests 2A and 2B. The first stage heat load plus heat that boils the LN 2 in the tank equals the total shield heat load [18]. The total heat load for test 2B is larger than for test 2A at zero current.…”

“…TABLE 1 also presents the temperature at three locations on the 6061-Al shield around the cold mass. TABLE 2 shows the effect of current in the five magnet coils on the heat removed by the four cooler first-stages [18]. TABLE 1.…”

“…TABLE 2. The Cu plate temperature near the coolers, the estimated ΔT between the Cu plate and the cooler first stage, the estimated first stage temperature and the heat flow into the cooler first stages as a function of the current in the five coils during the magnet test 2B [18]. There is no LN 2 The heat that goes into the cooler cold heads increases as the magnet is powered.…”

Results from the Cooler and Lead Tests

“…The cooler lead test did a great deal to increase our understanding of what happened with the spectrometer solenoids during the tests of the last two years [34], [35]. From the results of the latest spectrometer solenoid test [35] and the cooler-lead tests, we can improve the operating performance of the coupling magnet [36].…”

“…From the results of the latest spectrometer solenoid test [35] and the cooler-lead tests, we can improve the operating performance of the coupling magnet [36]. Knowing how the copper and HTS leads behave while the system was connected to a PT415 cooler greatly enhances our understanding of how the coolers and the magnets behave as a system [18], [21], [35].…”