The Large Hadron Collider ended its 2010 physics runs on 6th December last week and has now shut down until February. For the rest of the week the LHC beam teams met up in Evian to open discussions on how to run the LHC throughout 2011 with the goal of collecting at least 1/fb of integrated luminosity. The slides from the full set of talks can be found on the CERN indico server.
Towards the end of the proton runs this year a peak luminosity of 0.2/nb/s was reached. If they could run very efficiently during the year this would be enough to collect 1/fb, but that is too optimistic. To be confident of reaching the target they will aim for a peak luminosity of 1.0/nb/s. There will also be a small increase of the collision energy from 7 TeV this year to 8 TeV in 2011.
Reaching a luminosity of 1.0/nb/s will take some time and hard work. The increased performance will come from several factors, but the main ones will be an increase of bunch numbers up to 930 per beam and a tighter beam squeeze down to beta* of 2.0m or less. Last year they reached 368 bunches using a bunch spacing of 150ns. To get more bunches in they must decrease the bunch spacing and at the end of last year they already tried 75ns and 50ns spacing. This was not without problems, but they think they can manage 75ns runs for next year to fit in the 930 bunches.
It is less clear what value of beta* they will aim for. The luminosity is inversely proportional to the beta* value and last year they squeezed to 3.5m. If they can get this down to 1.75m it will give a factor of 2 in luminosity. This plot from a talk by R. Bruce gives some idea of what is possible. Notice that the higher energy helps to get to lower beta
To implement these new parameters there will have to be a completely new beam and collimator setup. Last year it took about three weeks in August to perform the commissioning for the parameters used in the final runs and they expect a similar time next year. That will be followed by a gradual build up of luminosity by increasing the number of bunches. According to a scenario suggested by Mike Lamont, first they will go up in steps of 50 bunches until they reach 300, then in steps of 100 until 900 bunches. This would take another six weeks to complete (This obviously not intended to be a precise plan since they normally go up in multiples of 24 bunches.) This means that full operating conditions will not be reached until May. After that the amount of data they can collect depends on how efficiently they can run the collider for the remaining 6 months of proton runs. A draft schedule can be found in the talk by J. Wenninger . Unfortunately there are no slides for the closing words from Mike Lamont and Steve Myers.
There remain a number of issues that will make the run for 2011 challenging. The famous “Hump” is still causing some problems and its cause is unknown. The UPS (Uninteruptable Power Supply?) is a suspect but the unusual pattern of the interference is hard to explain. They are working to pin down the true frequency of the signal by oberving its effects and are using magenetic probes around the ring to try and pick it up directly. The beam damping systems have largely removed the problem of the hump, but it still has some potential to destabalise the bunches and reduce luminosity. Hump spectators can read the slides of G. Arduini for full details.
Other problems considered include the UFOs (Unidentified Falling Objects) which may be cleaned out by running some high intensity beams for a while. The e-cloud is another big issue that was found at the end of the runs this year. It is a build up of electrons that leads to unwanted background collisions near the detectors. No doubt some work will take place during the shutdown to help address these and other problems for next year. Final decisions on run parameters and schedule will be taken at the Chamonix meeting in January.
Meanwhile, a conference on LHC First Data is scheduled for next week!