Those responsible for the LHC machine are having their yearly meeting this week in Chamonix to discuss the state of the project and plans for the future. Last week a subgroup met to discuss plans for beam commissioning to 3.5 TeV/beam, starting next month. The current schedule envisages beam commissioning to restart around February 19, and best estimate is that it will take about a month to establish safe, stable 3.5 TeV beams and begin extended runs for physics purposes. There’s a plan for a big media event when first collisions are achieved at 3.5 TeV/beam, something that may require discouraging experiments from announcing observation of high-energy collisions that happen before the planned moment (evidently this is what occurred last year, when Atlas saw 1.18 TeV/beam collisions before they were supposed to…).
This year’s schedule includes a possible one-month stop mid-year to increase the beam energy from 3.5 to 5 TeV, but based on the discussions at Chamonix, this looks very unlikely. The most serious problem with the LHC remains the bad splices which are known to exist in the machine, as well as sectors where definitive measurements of all the splices have not been possible (they would require warming up the sector, causing delays of months). The current knowledge of the splices leaves no room for error, even at 3.5 TeV, and going to 5 TeV would require warming up parts of the machine, something which cannot be done during a 1-month stop.
Discussions are beginning about how long a stop for repairs should be planned for after this year’s run ends in November. To be able to run at 5 TeV/beam will probably require keeping the machine off until May 2011 to fix splices. Going to the design energy of 7 TeV may require even more extensive work on the splices, work that could keep the machine off for all of 2011, with startup again in 2012. To get above 5 TeV, work also needs to be done on retraining the magnets through repeated quenches. Not much of this would be needed to get to 6.5 TeV/beam, but to go all the way to 7 TeV, problems that are still not understood with magnets from one manufacturer will have to be addressed.
Update: From the Chamonix summary talk, there are two main scenarios now being considered. In the first, the energy of the machine would stay at 3.5 TeV/beam this year and next, with .1-.5 fb-1 integrated luminosity in 2010, 1 fb-1 in 2011, then a year-long shutdown in 2012 to fix all splices before moving to 6.5-7 TeV/beam. In the second, splices would be fixed in stages, running for only 5 months in 2011, at 5 TeV/beam, 1 fb-1 integrated luminosity.
There will be a summary session at CERN next Friday.
Peter, thank you, this blog really did at some point become the best source of LHC updates I’ve found anywhere.
One question I’ve had about all of this since Resonaances first broke the “one manufacturer’s splices” part of the story — these things under warranty?
Amos,
I’m pretty sure that these things come with no warranty. They’re extensively tested before being put in the tunnel, but I’d guess there’s no warranty after that.
6 to 12 months to repair splices sounds pretty pessimistic – warm-up about 1 month, cool-down about 2 months, so that’s about 3 to 9 months of actual work?
If it really is that major, then putting the work off until other planned work (e.g., the 2014ish upgrade of the inner triplets?) is carried out, would sound attractive. There’s plenty of interesting stuff that can be done at 3.5TeV/beam at the mean time (especially if working at a lower energy allows being more aggressive with the beam current).
Let me try to balance the somewhat gloomy cast of Peter’s post by pointing out Ralph Heuer’s (Director General of CERN) new set of goals for the 2010-11 run: there should be 500 pb-1 delivered to each experiment at high energy (which is expected to be close to 10 TeV, ie, maybe 9 TeV) and one considers even 1 fb-1. With such a data set, the LHC experiments could surpass the achievements of the Tevatron experiments, both in terms of standard model measurements and searches for new physics.
Michael,
Thanks for pointing this out. The splice problem directly affects the possibility of increasing beam energy, but not the prospects for increasing luminosity. Beam commissioning went very well last year, and the same may be true this year.
For the most relevant Chamonix presentation about the effects of the splice problem on prospects for 5 TeV/beam, see
http://indico.cern.ch/getFile.py/access?contribId=2&sessionId=0&resId=1&materialId=slides&confId=67839
My reading of it is that going to 5 TeV/beam will require warming up sectors and is not going to happen in 2010. There are ideas discussed that could be implemented with a 1 month stop, but it looks like they would only allow an increase to 3.8-4 TeV/beam.
Ralph,
I think the fundamental problem is that there are 24000 splices in the LHC main circuits, and if something has to be done to a significant fraction of them, that’s a time-consuming job.
It will be very hard to hide the fact that they have collided at 3.5 TeV unless they cut public access to most of the status displays and logs.
You must have a magic wand for reading technical slides..I couldn’t find those 2 run plans in the Chamonix summaries at all…inside info?? 😉
Look at Mike Lamont’s talk on integrated luminosities, and the summary Steve Myers talk.
Pingback: LHC Update, More « Not Even Wrong
On splice fixing, I just noticed the following from http://indico.cern.ch/getFile.py/access?contribId=8&sessionId=2&resId=1&materialId=slides&confId=67839 :
Conclusion: For safe running around 7 TeV, a shunt has to be added on all 13 kA joints
In other words, it looks like its not just a matter of fixing a limited number bad splices, they’re going to put an additional copper piece around every single high current splice between magnets.
(Thankfully they don’t seem to be talking about the splices inside the magnets also – presumably the protection diodes give extra margin there).
i have one question why they have down till february ? what is that for
robert,
Last year the new quench protection system was not yet fully operational, but they decided to go ahead and start beam commissioning at lower energy (1.18 TeV). This month they’re finishing commissioning the quench protection system, and getting things ready to operate at 3.5 TeV. They should be ready to start beam commissioning again sometime in February.