Voir en


Accelerator Report: Leak repaired, cooling in progress

Despite challenging working conditions, the vacuum leak caused by the quench that occurred on 17 July was repaired in less than 10 days


The interconnection between the Q1 (left) and Q2 (right) magnets when closed. (Image: CERN)
The same interconnection when opened, revealing the various lines for cables, helium and beam vacuum. The arrow indicates the location of the leak. (Image: CERN)

In our last Accelerator Report, we referred to a quench of an inner triplet magnet located to the left of Point 8 (LHCb) that had caused a small leak in the insulation vacuum of the inner triplet assembly. This vacuum barrier is crucial for preventing heat transfer from the surrounding LHC tunnel to the interior of the cryostat. We now know that the quench was triggered by the quench protection system (QPS) following an electrical disturbance on the general electricity grid.

The insulation vacuum reached atmospheric pressure on the morning of Monday, 17 July, but it took another week to bring the magnets to room temperature, ready for a possible intervention. During that week (week 29), the cryogenic and vacuum teams identified the source of the leak, located between the magnet cold mass and the insulation vacuum. The size of the leak was estimated at approximately 1 mm2, sufficiently large to “hear” the sound of the gas leaking out.

Now, before I go any further, let me remind you what an inner triplet is. Before they enter an experiment detector, particles must be squeezed closer together in order to increase the collision rate – this is the job of the inner triplets. Three quadrupoles are used to create an inner triplet. There are eight inner triplets in the LHC, two on either side of the four large LHC detectors: ALICE, ATLAS, CMS and LHCb.

Equipment to measure sensitive vibrations was installed in the inner triplet in question, in the interconnections between the quadrupoles, which indicated that the probable location of the leak was at the interconnection between the Q1 (the closest quadrupole to the LHCb interaction point) and the Q2 quadrupoles.

In parallel, the cryogenic team drew up various possible recovery scenarios. The standard procedure would have implied the full warm-up to room temperature of the entire sector, in which case more than three months would have been required to bring the sector back to beam conditions. So an alternative, less restrictive, scenario was developed: the sector would be left to drift up slowly in temperature with all liquid helium removed from the magnets and all cryogenic lines depressurised for a limited duration intervention – estimated at 10 days maximum.

The new bellow is in place. (Image: CERN)

Just one week after the incident, the magnet and vacuum teams opened the large bellows around the interconnection between Q1 and Q2. The exact location of the leak was identified that same day: it was located on a flexible bellow installed on one of the lines between the two magnets. The decision was taken to perform an in-situ intervention to replace the faulty bellow with a spare one.

Easier said than done… as this type of bellow is delivered as an integral part of the triplet magnets. A completely new in-situ welding strategy had to be developed as the work progressed. Despite challenging working conditions for the welders, the new bellow was in place and the absence of a leak confirmed by the end of the week. On the evening of Friday, 28 July, the interconnection was closed again.

Over the weekend, the vacuum team successfully pumped down the insulation vacuum. After a final high pressure and electrical integrity test, cool-down started on Tuesday, 1 August, just in time to avoid a complete warm-up.

As I write this article, cool-down is in progress. Despite the challenges and the unprecedented nature of the incident, we succeeded in limiting its impact on the run: LHC beam operation is expected to resume in the first half of September, in time for the 2023 LHC ion run. Once again, this was made possible by the hard work and collaborative spirit of all the teams involved.


Take a look at the various stages of this incredible operation in pictures, and for full technical details, watch this video interview with Paul Cruikshank, from the TE department.