The Large Hadron Collider has just finished its best run yet with 19 hours of stable beams providing an integrated luminosity of about 75 inverse nanobarns. This takes the total luminosity accumulated to about 350 inverse nanobarns. The run with 13 on 13 bunches and 8 collisions per turn began with a new record peak luminosity of 1.6 MHz/barn for ATLAS and CMS.
The LHC has now entered a planned technical stop that will last two or three days before restarting for the last push to increase luminosity before a period of steady physics runs during August. According to the medium term plan set a few weeks back they should be aiming to double the number of bunches up to 24 over the next two weeks. However the luminosities have been a bit lower than expected because they have settled for bunch intensities below nominal in order to improve stability. A Physics Programme Meeting this afternoon will tells us what they plan to do next.
As we’ve said before, optimising the filling schemes for the LHC is difficult because the collision point of the LHCb experiment is 11 meters away from its ideal position exactly one eighth of the way round the collider ring. This is because the experiment has an asymmetric design and had to be placed in an old cavern used by a more symmetric experiment for LEP, the previous collider in the tunnel. If it were not for this they could place the 4N proton bunches equally spaced around ring and get 4N collisions per turn in all the experiments. Because of the asymmetry this is not possible. Up to now they have used mostly a filling pattern that provides 2 collisions per turn for every three bunches in each beam (see my previous post for a video of how that works). That is just 67% efficient compared to the optimum.
That was a democratic solution because it gave the same potential luminosity in each experiment which was great for the calibration stage. Now as they move into a more serious phase where potential new physics becomes possible it is more important to give the maximum luminosity to CMS and ATLAS. ALICE can take a back seat because its real role is for heavy ion collisions and that will not be tried until much later. LHCb can also do with less but not too much less.
Another filling scheme with offsets the positions in such a way that ATLAS and CMS get the optimum number of N collisions per turn with N bunches, while LHCb gets just half the number and ALICE gets none. This is described as scheme 1 in the presentation of this afternoon. Then there is also scheme 2 that gives two collisions per turn for every three in LHCb. By alternating between these two schemes and possibly the current one, they can get a good mix of luminosities.
Update: The technical stop is due to end on Thursday. The plan now is to attempt ramping the energy up with 24 bunches on Friday.