LHC switch-on: a preview

The LHC control room (heavy lever obscured from view).

By Jon Cartwright

Until a few months back I had an excited vision of the moment the great LHC “switch on” would take place. Here’s how it goes: The control room, normally frantic with the workings of scientists, falls under tense silence as a lone technician grips a heavy lever. Just as the quiet becomes unbearable, the director general mutters: “OK, let’s go.” Beads of sweat trickling down his temples, the technician heaves back the lever while averting nervously to a dial that has coloured bars going from green to yellow to red (450 GeV…5 TeV…7 TeV…DANGER) . “Faster!” cries the director general, his eyes glowing with a sort of manic intensity, “Faster!” Then the control room begins to shake and the scientists dive under their workstations to avoid the plaster and tiles falling from the ceiling.

Needless to say, the real event tomorrow will not satisfy onlookers with any cinematic clichés (and nor will the beams break any speed records — they will be strictly cruising at their injection energy of 450 GeV). But that’s not to say the event will be without drama, as I found out today when I went to CERN’s Meyrin site for for lunch with Paul Collier, head of the accelerator operations team.

“It’s not like blasting off from Cape Canaveral,” he said, referring to the fact that there is no definite countdown for performing certain tasks. Rather, the operations team will be learning as it goes, and we will get to watch — milestones, mistakes and all. The current plan is to inject the first beam into the ring at around 9:30 am, but it could happen anytime between 9 and 10 am (keep an eye out on this blog for the decisive moment). From then on, the team will take the beam round the LHC’s 27 km-long ring in a dozen or so sections, each initially fenced-off by a physical barrier.

At each section the team will monitor the beam’s x–y displacement and spread before letting it into the next. If the beam looks like it is veering too far off course (between 12 and 15 mm), one of the technicians will apply a correction algorithm, which will tell exactly how to adjust the magnets. Correction completed, the team will remove the barrier to the next sector and let the beam flow in.

Part of the trick to this trail-and-error process is to have humans, rather than, computers, decide when a correction is necessary. “There’s certain things that a brain inside a machine can’t do, that a human can,” said Collier. The trouble with computers is that they see all the noise, whereas humans are more adept at seeing past certain errors and getting an overview of the beam’s progress.

So how long will it take to get a proton beam all the way round? If the 1989 switch-on of CERN’s previous cutting-edge accelerator, the Large Electron–Positron Collider, is anything to go by, it could take 12 hours. But, as Collier points out, the LHC is far more advanced both in controls and instrumentation. “We could get a beam round in a few hours if all goes well,” he says — although that’s without any “showstoppers”.

Indeed, if the team does get a beam all the way round tomorrow it should be able to try to do the same for a beam going in the other direction (they haven’t yet decided which to try first: clockwise or anticlockwise). “The principal goal for Wednesday is to get a beam all the way round the machine,” notes Collier. “But our personal goal is to get beams going both ways.” Over the coming days, the four main experiments on the LHC ring could even see their detectors light up with some low-energy beam collisions.

I noticed when I was speaking to Collier that he didn’t seem in the least phased by his big media day; he gazed directly ahead and spoke about tomorrow as though is was just another date in his diary. So at the end of our lunch I asked why he thought there was so much interest in the event. “The LHC is,” he began, and then paused to search for the right word. “The LHC is fundamentally special.”