LHC dipole magnets in the tunnel
A chain of LHC dipole magnets inside the tunnel at point 1 (ATLAS) towards the end of Long Shutdown 2 (LS2).When the Large Hadron Collider (LHC) begins Run 3 next year, operators aim to increase the energy of the proton beams to an unprecedented 6.8 TeV. (Image: CERN)

It was at 9.30 a.m. on 10 September 2008 that the LHC’s first beam was injected, in the full glare of the global media spotlight. Just under one hour later, a beam had been successfully steered all the way around the ring, to scenes of great emotion at the Laboratory. A long wait was over, LHC page 1 became the focus of everyone’s attention around the Lab, and a new era of research seemed about to get under way, but the sense of euphoria was to be short-lived.

In the days that followed, things went well, but then disaster struck: during a ramp to full energy, one of the 10 000 superconducting joints between the magnets failed, causing extensive damage that took more than a year to recover from.

It was unheard of to start a machine like the LHC in the public eye, but I’m assured we had little choice. In the months and weeks before the start-up, particle physics had never seen so much media attention. A small number of individuals on social media had managed to stir up the myth that the LHC would create a world-eating black hole, and the newspapers were full of it. They were going to come to CERN whether we asked them or not, so we invited them in on the basis that it would be better to have them inside the Lab than outside, telling the world that CERN was starting up the “black hole machine” behind locked doors. Over 300 media outlets came, BBC Radio 4 did an unprecedented full day of outside broadcast from CERN, and an estimated billion people watched as I gave the countdown to that first beam. I thought I was just talking to physicists in the main auditorium!

home.cern,Life at CERN
The scene in the CERN Control Centre on 10 September 2008, when beams went round the LHC for the first time. (Image: CERN)

Those joyful events of 10 September firmly established CERN’s place in the public eye, while the failure of a magnet interconnection just over a week later ensured the Laboratory would stay there. There was, and there remains, fascination with the human endeavour that particle physics represents, and the media were kind to us on the whole. But for me, the most important part of the story was somewhat lost.

The LHC is unique. Like any energy-frontier accelerator, it is its own prototype, and building it was a learning experience from the start. Despite the serious nature of the setback in September 2008, it was really just another step, albeit a big one, on a long learning curve. As with previous setbacks, the LHC team was hard at work the next day to ensure that we could recover as fast as possible. We soon understood the problem, and we had all the spares we needed. It took a year to put right, but we knew straight away what we had to do.

Image of a LHC beam screen recorded on 10 September 2008, showing two spots corresponding to the successful circulation of protons once around the machine. (Image: CERN)

It’s a great tribute to the global particle physics community that setbacks are confronted with a confident, positive approach. In 2004, after we’d installed a full sector of the cryogenic distribution line (QRL), it failed and had to be removed from the tunnel. To me, this was a much bigger issue than the 2008 event, since it required the whole LHC installation schedule to be rearranged while the contractor made good the problem with considerable help from CERN. Our Director-General at the time, Robert Aymar, was an engineer, and he understood the magnitude of the problem perfectly. He was the unsung hero in liberating the resources needed to get it fixed. It’s also thanks to him that we have Linac4, a key part of the HL-LHC project, whose construction began during his mandate. Later, in 2007, one of the so-called inner triplets, which perform the final focus of the beams, failed a high-pressure test in the LHC tunnel. It was remarkable how quickly CERN staff came up with an innovative and elegant solution, and implemented it with the help of colleagues from Fermilab, KEK and the Lawrence Berkeley National Laboratory.

Following repairs and consolidation, on 29 November 2009 there were beams circulating again in the LHC, and full commissioning could get under way. The experiments had had an extra year to prepare, and although I’m sure they’d have preferred beam in 2008, they were in perfect shape to start data taking. Every cloud has a silver lining. This time, start-up went very quickly. The injector chain worked beautifully, as always, with even higher performance than we’d anticipated: a great tribute to our predecessors who built those machines from the 1950s onwards. We’d also learned a lot from LEP, and instrumentation was very much improved. The LHC physics programme, at an initial energy of 3.5 TeV per beam, began in earnest in March 2010.

I’m an accelerator physicist, but I want to finish by talking about the experiments. It’s not only the LHC that took technology way beyond anything that had ever been done before. Like the accelerator team, the experimental collaborations had also learned much from their predecessors. The previous generation of hadron collider experiments had luminosities two orders of magnitude lower to deal with, they had around a million readout channels compared with the LHC experiments’ up to 100 million, and their data rates and volumes were also much smaller. It’s thanks to the efforts of a global, multidisciplinary collaboration that the LHC project delivered so well on its promise right from the moment data taking began, re-measuring everything we’d learned before about the Standard Model of particle physics in the first few months of operation, and then going on to new discoveries. But that’s a story for another day.