Mobilizing for the LHC

27 January 2009

Investigations following the incident in Sector 3-4 of the LHC on 19 September have confirmed that the cause was a faulty electrical connection between two magnets. This resulted in mechanical damage and the release of helium from the magnet cold masses. CERN has published two reports on the incident and confirmed that the accelerator will be restarted in summer this year.

An interim report issued on 15 October gave the result of preliminary investigations. A more detailed report followed on 5 December, confirming that a small resistive zone developing in a bus connection in the circuit that conducts current between magnets probably caused the incident. Arising during the ramping-up of current in the main dipole circuit at the nominal rate of 10 A/s, in less than a second the zone led to a resistive voltage of 1 V at 9 kA. The resistance was small – 200 nΩ – dissipating of the order of 10 W at high current intensity. The power supply, unable to maintain the current ramp, tripped off and the energy-discharge switch opened, inserting dump resistors into the circuit to produce a fast decrease in current. In this sequence of events, the quench-detection, power converter and energy-discharge systems behaved as expected. Within a second, an electrical arc developed, puncturing the helium enclosure and leading to a release of helium into the insulation vacuum of the cryostat. After three and four seconds, the beam vacuum also degraded in beam pipes 2 and 1, respectively.

The insulation vacuum then started to degrade in the two neighbouring subsectors. (A vacuum subsector consists of two lattice cells, each with six dipoles and two quadrupoles, with “vacuum barriers” at both ends.) The spring-loaded relief discs on the vacuum enclosure opened when the pressure exceeded atmospheric, thus releasing helium into the tunnel, but they were unable to contain the pressure rise below the nominal 0.15 MPa in the vacuum enclosure of the central subsector. This resulted in large pressure forces acting on the vacuum barriers separating the damaged subsector from its neighbours.

Investigation teams confirmed the location of the electrical arc and, while they found no electrical or mechanical damage in neighbouring interconnections, they discovered contamination by soot-like dust, which propagated over some distance in the beam pipes. They also found damage to the multilayer insulation blankets of the cryostats. In addition, the forces on the vacuum barriers attached to the quadrupoles at the subsector ends were such that the cryostats housing these quadrupoles broke their anchors in the concrete floor of the tunnel and moved, with the electrical and fluid connections pulling the dipole cold-masses in the subsector from cold supports inside their undisplaced cryostats. The displacement of the quadrupole cryostats – short straight sections (SSS) – also damaged jumper connections to the cryogenic-distribution line.

As soon as the gravity of the incident was clear, a campaign for cryostating and testing of spare cold masses – both dipoles and quadrupoles – was immediately launched. After the sector had been warmed up, by the end of October, the repair programme began in earnest with the inspection of all the affected magnets – first underground and then at the surface. The programme also includes the inspection of the beam pipes and screens for contamination by soot-like metal dust and debris from the damaged insulation blankets. All the affected sections will be cleaned.

Virtually all the cold masses of magnets in the affected zone seem to be intact, with the possible exception of the bus bars in the end zone. The damage has been mainly to components located between the cryostat and the cold masses, as a result of the displacement that occurred. In all, from a total of 57 magnets (42 dipoles and 15 SSS) in the affected zone, 53 magnets (39 dipoles and 14 SSS) have been removed from the tunnel for inspection and/or cleaning or repair. Of the magnets to be re-installed, 39 (30 dipoles and 9 SSS) will have new cold masses, almost depleting CERN’s stock of spares. The decision was taken to reuse spare cold masses as much as possible to enhance operational safety. Nine of the dipoles removed are believed to be undamaged and will simply be inspected and retested. Five SSS will be reused after reconditioning of the cryostat (i.e. change of multilayer insulation blankets and the cold supports).

This work is being carried out in building SMI2, where the cryostat facility is based, and also in B904 at the Prévessin site. Meanwhile a temporary line for decryostating dipoles has been installed in B180 (West Hall) to recover quickly cryostat components that will be used for new cryodipoles based on new cold masses.

All magnets will undergo complete warm and cold testing in building SM18, where they were tested before original installation. They are being tested up to 12 850 A, which corresponds to a field of 9 T, compared with the 8.3 T for nominal LHC operation at 7 TeV. It will be possible to test up to five magnets a week, once more cryogenic capacity has been brought on line in February. In addition, the circuits of the main magnets are undergoing power tests to detect any abnormal resistances. As a result, a magnet in Sector 1-2 will also be replaced.

As of mid-January, all 53 magnets had been brought to the surface and the first eight replacement units had been installed in the tunnel. The goal is to have all the magnets in place in the tunnel by the beginning of April. Making the interconnections will start at the beginning of February, with enhanced quality control.

New electronic boards will protect the magnets by constantly measuring the resistance of the busbars and the interconnections. These additional electronics will also measure other parameters. Installation will start at the beginning of April. In addition, a better use of the present quench-protection scheme will help to single out possible bad connections inside cold masses already installed in the tunnel. The final stage will be the testing of the entire sector in June and July.

• For the two reports, see

bright-rec iop pub iop-science physcis connect