Helium Leak Forces LHC Shutdown for at Least Two Months

It’s this sort of news I really did not want to wake up to. At 0927 GMT Friday morning, a fault known as a “quench” resulted in the leakage of a tonne of helium coolant causing 100 of the LHC superconducting magnets to heat up 100°C. The fire services had to be called and it was some time before engineers could access the tunnels to assess the damage. It was worse than they were expecting. Although no one was hurt and there was no danger to the public, the once-supercooled magnets were one hundred times warmer than they should be and optimal vacuum conditions had been lost. To perform repairs, the rest of the damaged sector will need to be warmed up and then slowly cooled down again, resulting in a shutdown of LHC operations for at least two months

The leak occurred between the Alice and CMS detectors (sectors 3-4) after repairs to the faulty 30-tonne transformer were being finalized and the systems were being powered up to begin a new series of commissioning tests. According to the LHC logbooks, temperatures rose by 100°C and the vacuum required within the equipment for particle circulation to be possible was lost. Engineers had to wait for oxygen levels to return to normal within the tunnels before they could investigate the “meltdown.”

Although last week’s fault with the transformer caused frustration, setting LHC experiments back by a few days, scientists were optimistic the incident would have minimal effect on the first scheduled particle collisions in October. Friday’s quench, however, is a serious incident, knocking the largest experiment mankind has ever attempted offline for at least two months. Although this is sad news, many scientists are keeping a positive attitude:

This kind of incident was always a possibility with such a unique and demanding project, that’s why we were so tense on the 10th [of September]. Having seen those tantalising first signs of beam in our detectors, everyone is raring to go. So it’s really disappointing, and hard for us to keep in perspective right now. But a delay like this in a 20-year project isn’t an utter disaster and I’m sure the team at Cern will fix it, and make it more robust as they go.” – Prof Jonathan Butterworth of University College London, the UK head of the Atlas detector.

So what happened? The basic operating conditions for the LHC depend on very low temperatures and a very high vacuum state. It would appear both key conditions were lost as engineers tested the electrics of the LHC in the run-up to full commissioning. There was a faulty connection between two of the superconducting magnets, so when the system was switched on, the high current melted the connection, causing the helium leak. The loss of supercooled helium caused a rapid release of stored energy (an event known as a quench), heating the magnets and destabilizing the vacuum conditions.

After such a smooth start to the first proton circulation on September 10th, these setbacks may come as a surprise. However, probing the frontier of physics rarely happens without a few hiccups along the way, so let’s hope this incident will be the last and we can once again look forward to the first particle collisions toward the end of the year…

Sources: BBC, Telegraph

31 Replies to “Helium Leak Forces LHC Shutdown for at Least Two Months”

  1. Black holes, the 2012 doomsday prediction, the annual euchre tournament at the Cydonian complex on Mars…..

    Where will it all end?

  2. @ Tammy: LOL! I should have shown that to all of the “black hole” phobics before news of the shutdown occurred.

    @ Ian: I’m guessing they will encounter another hiccup, and we will see this thing become operational in January, right when the next President is elected.

    And then right after that, a black hole will form. 😉

  3. A quench occurs when a portion of the superconducting wire goes normal. That normal section of wire suddenly goes from no resistance to nano-ohms of resistance. It doesn’t sound like much but when there are hundreds of amps running through the wire and the temperature is single digit Kelvin, it doesn’t take many watts to cause the surrounding wire to go normal. A chain reaction loops completely around the magnet till the entire coil is normal. The remaining energy boils off the remaining liquid helium. Once the helium is gone, the rest of the energy raises the temperature of the magnet.

    I worked on a superconducting medical cyclotron in the ’90s. The cryostat held 80l of liquid helium. We ran the coil at 200 amps during normal operations. Occasionally, the coil would quench on us. Usually, it was a problem with the cooling in the electrical leads into the coil. The entire coil would go normal and boil off all of the helium in about a half second. There was enough current in the coil after that to raise the temperature to 70 Kelvin (or degrees Kelvin if you want to watch a physicist’s blood pressure rise 🙂 ). Each liter of liquid expands to more than 800l of gas (I think, that was a long time ago) so we had to vent the coil with a burst plate to prevent something else acting as a burst plate. Quenches were pretty dramatic and expensive.

    What ever the root cause of the quench at LHC was, the fact that the coils jumped to over 375 Kelvin tells me they have a very small cryostat in those magnets. I am sure the engineers purposely quenched a magnet during the initial production stages to verify the engineering and to ensure the magnet can survive the event. When ever a new magnet was built at the National Superconducting Cyclotron Lab at Michigan State (where I also worked during the ’90s), we would quench it at least once just to make sure everything worked like we wanted it to.

  4. NO! NO! NO! NO! NO! NO! NO! NO! NO! NO! NO!

    I am in denial… 🙁

    We wait years and years, and then they FINALLY have it up and starting to run, and it breaks…

    Well, now we can create some anti-conspiracy theorist theories, as in:

    The I-fear-the-black-hole-will-eat-me-ers lainched a terrorist attack on the LHC!!!

    @mcenhillk: Great post, very informative!

    I do have a question/observation. If you have 80l of LHe, each l gives you 800l of gas, that’s 64,000l of gas or 64 cubic meters or a cube of 4 meters side length.

    I half a second?? Sounds pretty much like a small explosion!! Not to mention the fact that the oxygen would be gone…

  5. question:
    I have a more practical one. If the LHC does indeed produce a mini black hole, could we then prove that Hawking radiation exists? If so, what would be the consequences?

    It is possible. The eventual end of every black hole could well be the same evaporation that will be seen with the quantum effects that should be the result of these mini-black-holes as they disappear. The possible problems in detecting them, however, are enormous. Not only is the “black hole” only supposed to last for about 0,00000000000000000000000001 seconds, the measurement process is still in the theoretical stage. There’s also the possibility that the Hadron collision might not produce a black hole. At this point that is still only a theoretical possibility.

  6. @Don Alexander: I think the cryostat had a total internal volume of 150l. It had something to do with the fact that we flipped the cyclotron 360 degrees around the patient. With that volume and 64,000l of gas, things get “sporty”. The nice thing about helium is it’s compressibility. Instead of a detonation and it’s associated shock wave, we got a quick pressurization. The burst plate popped at 5psi so it blew very quickly. This opened up a 2″ vent to the outside of the building. The only thing that could hurt those of us working on system is holding on to that vent during a quench. Nothing give you frost bite faster than holding on to a copper pipe that is venting 20K helium. 🙂

  7. I’m guessing your all saying all of these things with high squeaky voices !

    .. that is helium right ?

    /ignorance ..

  8. “Saved the drama for your mamma” This is a new machine – never has their been any cyrogenic device on this scale. Stuff happens. Stuff will happens again. I’m disappointed but these thing teach and I am sure CERN will overcome this setbeck. Research the early days of space flight and the ole Vanguard project! Or the many attempts to land a craft on the Moon (the Ranger program). This is just was complex…if not more.

  9. Kind of reminds me of the early days of Hubble! Once the glitches are ironed out, I’m sure the LHC will deliver up a Higgs boson with as much excitement as (or more than) the ‘resolved’ Hubble problems have provided for observational astronomers. Hang in there you particle physicists. Your time will come!

  10. I would have been amazed if something like this didn’t happen. The complexity of the apparatus in incredible. I just hope that the quality control has been top notch enough that this is an isolated incident. A failure so early on would tend to suggest otherwise though, and if so there may be many more such events…

    Ho hum – I was thoroughly looking forward to the first collisions…

  11. Check out the novel Einstein’s Bridge, about a parallel universe where the Superconducting Super Collider (SSC) was built and operated. The SSC somehow made beings in another universe aware of us, and they aren’t too happy with the idea of others existing even in another universe.

    So the scientists have to find a way to make the SSC never happen so we don’t get wiped out.

    Hmmmm – of course, the SSC was much bigger than the LHC, so maybe we are still safe.

  12. Astrophysical implications of hypothetical stable TeV-scale black holes

    Authors: Steven B. Giddings, Michelangelo M. Mangano

    (Submitted on 20 Jun 2008)

    Abstract: We analyze macroscopic effects of TeV-scale black holes, such as could possibly be produced at the LHC, in what is regarded as an extremely hypothetical scenario in which they are stable and, if trapped inside Earth, begin to accrete matter.

    We examine a wide variety of TeV-scale gravity scenarios, basing the resulting accretion models on first-principles, basic, and well-tested physical laws. These scenarios fall into two classes, depending on whether accretion could have any macroscopic effect on the Earth at times shorter than the Sun’s natural lifetime.

    We argue that cases with such effect at shorter times than the solar lifetime are ruled out, since in these scenarios black holes produced by cosmic rays impinging on much denser white dwarfs and neutron stars would then catalyze their decay on timescales incompatible with their known lifetimes. We also comment on relevant lifetimes for astronomical objects that capture primordial black holes.

    In short, this study finds no basis for concerns that TeV-scale black holes from the LHC could pose a risk to Earth on time scales shorter than the Earth’s natural lifetime. Indeed, conservative arguments based on detailed calculations and the best-available scientific knowledge, including solid astronomical data, conclude, from multiple perspectives, that there is no risk of any significance whatsoever from such black holes.

    Subjects: High Energy Physics – Phenomenology (hep-ph); Astrophysics (astro-ph); General Relativity and Quantum Cosmology (gr-qc); High Energy Physics – Experiment (hep-ex); High Energy Physics – Theory (hep-th)

    Report number: CERN-PH-TH/2008-025

    Cite as: arXiv:0806.3381v1 [hep-ph]

    Submission history

    From: Michelangelo Mangano [view email]

    [v1] Fri, 20 Jun 2008 12:57:24 GMT (222kb,D)


  13. man they had put the black hole fears to rest, finally got rossler and wagner to STFU, and were getting ready to do some actual science. and now this happens! oh well i supose a device this size there are a million thing that could break

  14. @mcenhillk: Thanks for the answer!!

    And I should say, venting ultracold helium gas will additionally cause, due to the highly non-adiabatic expansion of the gas, the vent to cool down vent faster.

    Reminds me of the greatest non-physicist I ever met. 😀 Guy at a retirment home was teaching us safety precautions for, I think, oxygen tanks? Anyway, he has this pressurized bottle with gas in it, opens the valve, whoooosh, and the valve suddenly gets frost on it.

    His explanation: “Because the gas in the pressure tank is at ABSOLUTE ZERO, IF NOT EVEN BELOW THAT.”

    I actually did not really understand why the valve cooled down so rapidly back then. But heck, yeah, I was sure that a liquid gas in a metal tank is going to be at room temperature…

  15. Oh thank god…. or what ever most physicists believe in now days. Here I was just waiting, any day now, for an event horizon to show up in my neighborhood to suck my house into a point of infinite density formally known as the LHC. Now though, I’ve got a two, (more likely eight), month reprieve thanks to obvious sabotage by a concerned scientist. I can finish reading my new book, and enjoy the fall foliage here in Kansas, which really just means watch the grass turn brown. So live it up people, we’re on borrowed time! If the black hole doesn’t get you, then 2012 certainly will. (Don’t forget, 2012 is another election year, end times for sure)

  16. It looks like you have changed the temperature increase to 100C, but you still state that the magnets were hot enough to boil water. Can you change it to whatever boils at -173C? :o)

Comments are closed.