The LHC has to operate at temperatures near absolute zero, otherwise the superconducting magnets that hold the proton beams in place won’t work properly. On Friday, around 100 of these magnets suffered a failure which caused a rapid rise in temperature, known as a quench. The magnets went up almost 100 °C, leaving them around a still comparatively chilly -170 °C, but much too hot to retain their superconducting properties.
The tunnel sector where the failure occurred will now have to be warmed up even further to allow engineers to perform repairs, and then cooled back down to its operating temperature of 1.9K. The process will take two month, says CERN spokesman James Gillies:
“A full investigation is still under way but the most likely cause seems to be a faulty electrical connection between two of the magnets which probably melted, leading to a mechanical failure. We’re investigating and we can’t really say more than that now. But we do know that we will have to warm the machine up, make the repair, cool it down, and that’s what brings you to two months of downtime for the LHC.”
As I recall, the original plan was to shut down the LHC over the Christmas period, presumably as researchers would be taking a break for some festivities. If they aren’t up and running until two months from now that will be approaching December, so I wonder if the planned shut down will still go ahead.
You can actually follow the temperature status of the LHC online. In the bottom left-hand corner, you can see that sector 3-4 has experience a spike in temperature up to nearly 100K. This page gives more detail, showing the spike occurring between lunchtime yesterday and midnight. I think it’s pretty cool that all of this information is accessible to anyone online.
As an aside, the Telegraph have a story about the difficulty in communicating the energy used by the LHC. Apparently, with a beam energy of 10 trillion watts, the LHC could defrost a pizza in just 30 nanoseconds, according to J R Minkel of Scientific American. Dr David Sankey of the Rutherford Appleton Laboratory disputes his figures, saying that because the protons are arranged in bunches it would actually be almost 1000 times quicker, at 250 picoseconds. I’ll keep you up to date with an further LHC pizza news, as it happens.