LHC Incident

So it's been ages since I've written about work. I suppose it kind of looks like I'm on holiday from my recent entries. But I have been working fairly hard, maybe not as hard as other people around CERN, but definitely harder than I was at home earlier this year. And this is despite the LHC accident on the 19th of September. A lot of people have asked me whether this delay in the machine schedule will affect my PhD. The answer is unfortunately yes; because while I don't need real data to write a thesis, it would be preferable. But there is still heaps to be done to prepare for data. And things which can be done with the limited first beam data which was collected and the extensive amounts of recorded cosmic ray data.

This week at CERN is an ATLAS Week, and as is customary during these collaboration weeks, there is always a talk by someone from the LHC project to tell us about the machine status and plans. It has been strange for the past few months, not knowing what is going on, and hearing about things from outside media sources, rather than internal emails or notices. I thought the speaker did a really good job of providing us answers with lots of details, which will hopefully stop all the speculation of what happened and what will happen from here. The talk really addressed all the questions which have come up in all the lunchtime and coffee conversations. If anybody is interested, the slides of the talk can be found here.

More interesting than this talk, however, are these photos, from the talk presented last week by the CERN Director General, of which the slides can't be accessed anymore, but the permissions were changed too late, and people had already taken photos from the talk and posted them online. So I figured it would be okay for me as well, as seeing the damage is very different from just reading about it.

I must make a disclaimer here and say that I am no expert on the incident, or even accelerators, and what I'm going to say here is my paraphrasing of various talks and reports.

On September 19th, seven out of eight sectors of the magnets on the ring were fully commissioned with for 5TeV operation, while one has been commissioned up to 4 TeV. The timing here part of the speculation, as while a transformer was being replaced on the other side of the ring, the commissioning of this final sector resumed.

While the current was being ramped up in the magnet circuit, a resistive zone developed in the electrical bus region between a dipole and a quadrupole, an area which is not monitored by the magnet quench detecting system. Then the resistive voltage grew to 1V and the power converter, unable to maintain the current ramp, tripped and current started to decrease in the circuit and an energy discharge switch opened, inserting dump resistors in the circuit to produce a fast power abort, which is what is supposed to happen in an event such as this.

An electrical arc then developed and punctured the helium enclosure, leading to release of helium into the insulation vacuum of the cryostat. An electric arc is an electrical breakdown of a gas which produces an ongoing plasma discharge, resulting from a current flowing through normally nonconductive media such as air. This activated the pressure relief system on the vacuum enclosure, however the system wasn't able to handle such a high pressure, which resulted in large pressure forces acting on the vacuum barriers separating neighboring subsectors. These forces displaced dipoles in the subsectors affected from their cold internal supports, and knocked the cryostats housing the quadrupoles and vacuum barriers from their external support jacks at positions, in some locations breaking their anchors in the concrete floor of the tunnel.

This damage is what you can see in the photos below, this first photo shows what happened to the region between two magnets when they moved.


and this photo shows the the movement of a magnet off its floor support.


It's kind of amazing to see the damage, and looking at the photos it is easy to despair, particularly knowing at there were 39 dipoles and 14 quadrupoles in the incident area. But repair is well underway and all of the magnets in the area and the surrounded segments will be removed and new ones will be installed, after been cold tested. A lot of effort has also been directed to being able identify resistive splices at low current, to avoid another such incident. Many additions have also been planned, the quench detection system will be improved so that the bus regions will be monitored, the ground support of the quadrupoles will be improved, and more devices will be added to aide in alleviating high pressures in the vacuum chamber, should they occur.

The plan is have the machine cold by the end of July next year. Good luck to all the physicists and engineers working on it, they definitely deserve acknowledgement and credit. It's easy to forget and to treat the machine and the detector as a black box which will provide numbers to analyse, but lots of people have worked very hard on it. And looking at what they have to fix, you can't help but respect them and not be impatient for data, because it is a very complicated system.