LHC Update May

The Large Hadron Collider crawled out of a scheduled technical stop two weeks ago and passed through a rocky patch. There was a series of cryogenic failures that slowed the build up back to normal luminosity. They are currently running with the worst hit sectors around point 8 at a temperature of 2.0 Kelvin rather than the normal 1.9 K. This appears to have fixed the problem but as an uninformed outsider I can’t help wondering what extra risks this entails. Another issue was emittance blow-up from the SPS that was limiting peak luminosity to around 4.3/nb/s. This was fixed in the last couple of days and now luminosities have returned to the record levels set before the technical stop of around 5.7/nb/s with bunch intensities up to 138 billion protons per bunch. previous discrepancies between luminosity recorded by CMS and ATLAS have been resolved by data from the Van de Meer scans run just before the technical stop. The two experiments are now in perfect agreement and previous record numbers from CMS have been rescaled downwards. The present luminosity should be close to the maximum they can achieve this year unless they have kept back some tricks for later.

On the plus side, minimum turnaround times are well under two hours which is about half last years waiting time. Recovery from loss of cryogenics also looks much faster than before. This means that if they can avoid problems with cryogenics and RF they should be able to accumulate data at a high rate. As I write they are passing the 2/fb mark for this year’s total with a little under 5 weeks before the next technical stop. It should be a breeze to reach the stated 5/fb target in time for the summer conferences.

There are a few conferences coming up over the next three weeks that could be opportunities for the experiments to present some early results using the 2012 data at 8 TeV. In particular Recontres de Blois opens on 27th May and Physics at LHC begins in Vancouver on the 4th June. However we may need to wait for the big ICHEP conference in Melbourne where they should be able to add about 5/fb from this years data at 8 TeV to last years similar total at 7 TeV. This looks likely to be a watershed moment for the Higgs search with a likelihood of at least an unofficial discovery moment if the combined significance exceeds 5-sigma (it is currently around 4.2 sigma) There is even the possibility that one of the two experiments could pass the discovery threshold with the diphoton decay mode. It depends on how lucky they are with the stats. There is also the possibility that this years data will tell a different story from last year and we will be left waiting for the full year’s dataset to complete the story. Whichever way it goes the ICHEP conference is billed as a historic moment for the Higgs boson, and it is just seven weeks away.

19 Responses to LHC Update May

  1. JollyJoker says:

    Two weeks to get everything analyzed for ICHEP seems like a tall order. Obviously they’ll time it to get some amount of data for each channel analyzed by then, and “some data” is still far more than 3/fb.

    It’s more than likely we’ll have a month where everyone knows the Higgs has been found but there’s no official combination that shows that and no official announcement from Cern. Will they try to avoid that situation somehow?

  2. wl59 says:

    So we can hope to enter in the next year being rather sure that Higgs don’t exist 🙂

  3. Murod says:

    4.2 sigma? Where this number comes from?

    • Philip Gibbs says:

      Unofficial combinations

      • anna v says:

        It is fun to look and see whether unexpected from the SM excesses combine to give a significant difference from its predictions.

        I remember though how surprised we were when the Psi was discovered at how narrow its width was. Our prejudices at the time expected it to be wider.

        I am trying to say that the signal being Higgs presupposes that this excess of events comes from one single ” resonance” with a large width. Until we see in individual channels the shape of the beast I think it is too soon to declare the Higgs found, even if the excess becomes 6 sigma. It will tell us that something significantly different from the SM is happening, and not more, until the decay channels are classified and are consistent with a Higgs decay, imo of course.

        Not that I will not be happy with a 6 sigma and some channel showing a resonance :).

  4. Tony Smith says:

    Has the LHC released a combination analysis for the diphoton channel for the 2011 data from ATLAS and CMS ?

    If so, how did they deal with what Matt Strassler described in his blog on 9 Feb 2012:
    “… ATLAS’s largest excess (126-127) is at a point where CMS has a deficit,
    and … ATLAS has a deficit across some of CMS’s excess …” ?


    PS – Andrew Elwell’s comments seem to indicate problems getting worse after peak luminosity/day went up to 5.7/nb/s.
    Would stability be improved by staying below 5/nb/s ?

    • Philip Gibbs says:

      They have not published that combination, perhaps for those reasons.

      I don’t know if a luminosity limit would help but they seem to hate such artificial limits. 🙂

  5. Chris Austin says:

    “that was limiting peak luminosity to around 4300/nb/s”

    I think that should be 4.3/nb/s. The LHC design luminosity is 10^{34} cm^{−2} s^{−1}, i.e. 10/nb/s.

  6. carla says:

    If things go reasonably well from now, they can deliver 3-3.5/fb by the first week in June in time for ICHEP, 4.5-5/fb for the other summer conferences. You’re being too optimistic as usual, Phil 😉

    • Philip Gibbs says:

      Maybe, depends how late they leave the cutoff

      • anonymous says:

        Phil—several experimenters told me a couple of days ago that the cutoff will be June 10th, roughly 3.5-4/fb delivered. However, they are unlikely to combine the 7 GeV run and the 8 GeV run—instead they’ll just keep going. That means a five sigma announcement will not happen before September.

      • Philip Gibbs says:

        10th June is three weeks away and they have just provided a record fill that delivered 170/pb in 15 hours. I make it 875/pb in the last week and IF they repeat that three times they will reach 5/fb (delivered) by the 10th. The recorded figure is about 10% behind and can be made up with a couple of good fills. We will see.

        It was said after Moriond that they would not initially combine the 7 TeV and 8 TeV. It would avoid the situation that one of the experiments gets the discovery first. However I will be doing the unofficial combinations so we will know where they stand

      • carla says:

        @Phil it’s surprising that the decrease in beam life time with increasing luminosity doesn’t have quite the reduction in integrated luminosity as I’d expect:


        Fill duration pk L Int L L half life
        2644 14:45 5640 170798.8 9 hours
        1836 14:55 1154 46578.6 18 hours?

        So a x4.9 increase in luminosity from L = 1.154 nb/s gives a x3.9 increase in L over 15 hours – which looks good!

      • JollyJoker says:

        3/fb in three weeks means around 12h / day in stable beams. 2/fb would be only 8h/day and 4/fb 16h/day. Given those numbers 4.5 – 5.5 / fb seems most likely for June 10th.

      • Philip Gibbs says:

        They just gave to keep those pesky gremlins away.

  7. kevin says:

    Yes, and they still have some margin, they can raise the bunch intensity to 1.6, what would give a 30% increase in luminosity. They can also improve the squeeze after the second or third TS. I think the figure will be 1.5 fb-1 per week at the end of the run. For a total of 20-25 fb-1.

%d bloggers like this: