(LEMIC)

1. MINUTES OF THE LAST MEETING

The minutes were accepted without modification.

2. BEAM CROSSING ANGLES AND MAGNET COMPENSATION IN V.6.1 (W. Herr - see transparencies in Annex 1)

Counter rotating beams must be kept separated during the injection and energy ramp-up phases. A beam crossing angle is required in collision mode, to minimise the beam-beam effect due to parasitic crossings. Werner Herr recalled the inconveniences of the two separation/crossing schemes that were studied in the past:

The solution retained for Version 6 of the LHC lattice is a combination of the two schemes that provides all the required flexibility within the performance of the available bumpers. There is a possibility to increase the crossing angle beyond ± 160 m rad, which is the revised nominal value in collision mode.

The cases of insertions 2 and 8 (ALICE and LHCb) call for special attention because of the experimental spectrometer magnets. There was a provision for two correctors in Version 5: this corresponded to a three magnet bump scheme that required a very high integrated field to put the beams back into collision. In Version 6, the experimental dipoles are compensated with a four magnet bump scheme that ensures that the beams stay in collision, independently of the tuning of b * at the IP. This correction scheme also increases the beam crossing angle and furthermore reduces the required bumper strengths in these even insertions. The location of the extra correctors has been discussed with the experiments concerned and solutions have been found in both cases.

A small tilt angle between the plane of the LHC rings and the direction of the field of the spectrometer magnet is acceptable, provided that the correctors stay aligned with the field of the experimental dipole. This would for instance allow the alignment of the spectrometer magnets along the local vertical axis.

A question was raised about the crossing planes of ATLAS and CMS: these need to be perpendicular to compensate for long range beam-beam effects. It is presently assumed that beams cross in a vertical plane at point 1 and in a horizontal plane at point 5, but the crossing planes could be exchanged if necessary.

The compensation of the LHCb magnet requires a ~1 Tm dipole between the experiment and the low-b triplet. Standard correctors are stronger and Pierre Lefèvre confirmed that the space optimisation demanded by LHCb is possible, but that a special magnet (and spare) has to be constructed. Werner Herr was asked if the crossing angle between the two beams need to be inversed when one flips the polarity of the experimental magnet. He thinks that there is enough strength in the crossing bumpers to over-compensate the spectrometer magnet. However, the aperture left for the beams would be marginal, and the same crossing angle could not be maintained in this case.

Lars Leistam presented a layout of the ALICE area with the supplementary corrector (Annex 2). He confirmed that the compensation scheme is accepted by the Collaboration but he underlined that :

• Some access and installation into the volume of the L3 magnet will require the removal of that extra corrector.
• This solution compromises the space available for future upgrades of the experiment.
• There is a forward detector (CASTOR) on this side of the collision point and there will be a demand for a larger aperture corrector.

(P. Lefèvre – see transparencies in Annex 3)

There are 3564 bunch positions in the LHC, separated in time by 24.95ns. Collisions at point 1,2,5 and 8 require global symmetry of 4 for the bunch trains. However, this symmetry is broken by:

1. the necessity to provide a large gap corresponding to the rise time of the beam abort kicker.

2. the shift of the collision point in LHCb and we have certain bunches that do not collide in IP2 and/or in IP8.

The LHC filling scheme presented in the Yellow Book includes, for each ring, 12 SPS pulses, each of 3 PS pulses (2 for the last batch) with 81 (or 80) bunches: the structure "333 333 333 332" thus contains 2835 (or 2800) bunches. Out of these, 183 bunches do not cross in point 8. The number of stored bunches drives the LHC luminosity and the number of special non crossing bunches increase the overall tune spread: both numbers are furthermore important parameters for the final performances of the machine.

The 24.95ns bunch spacing is obtained in the PS after a debunching/rebunching process that leads to an increase of the longitudinal emittance. This process occurs at 26 GeV/c, on harmonic 84, and 3 bunches (most probably 4 bunches) need to be kicked out of the PS before the transfer of the remaining 81 (or 80) bunches to the SPS.

An alternative is presently under study, where the PS is filled with 6 bunches on harmonic 7, thus leaving an empty location. The technique to split the bunches into 3 has been developed at the PS. Two consecutive "traditional" splittings then allow to structure the PS with 6 ´ 3 ´ 2 ´ 2 = 72 bunches and a gap of 12 positions on harmonic 84, with the same 24.95ns bunch spacing as in the previous scheme. This is obtained with no bunch lost and there is no increase of the emittance due to the debunching/rebunching process.

The next steps are to fill the SPS with 3 or 4 PS batches, and to reproduce the operation 12 times to fill one LHC ring. The final beam structure could be "334 334 334 333", leading to 39 ´ 72 = 2808 bunches in each LHC ring, and 186 non colliding bunches at point 8. These numbers are very close to those corresponding to the previous scheme.

This new filling scheme is very much preferred because it is "cleaner", but the RF gymnastics in the PS need to be studied and the resulting performance confirmed. The lengthening of the LHC injection kicker flat top, to accommodate 4 PS pulses in some SPS batches, is already foreseen. The intensity in the PS Booster rings has to be increased by 15% in order to keep the same bunch intensity (10¹¹ppp) in the LHC. Concerning TOTEM special runs, the new filling scheme could provide 39 bunches separated by 2 m s.

In conclusion, Pierre Lefèvre confirmed that the bunch spacing is and will remain 24.95ns, but he warned the users that the bunch train structure in the LHC can still evolve.

(E. Cennini – see transparencies in Annex 4)

The access control in the LHC buildings and underground areas results from legal prescriptions, from TIS recommendations and from the policy defined in the INB (Installation Nucléaire de Base) document. The basic concept is a hierarchy of access rights:

0. to the site and surface buildings,
1. to underground areas without interlock of the machine,
2. to underground areas with interlock of the machine, but no radio-activation,
3. to underground areas with interlock and with possible radio-activation,
4. to tunnel and experimental areas dedicated to the passage of cables and pipes.

The corresponding access control is dependent on an ID card, on the possession of a film badge, on an interlock token delivered at the access gate of the corresponding zone and on the possession of an operational dosimeter. Enrico Cennini reviewed the experimental sites and the classification of the underground areas according to these access control rules. In case of emergency access to the UX areas (AC3 level º operational dosimeter) can be forced to provide safety exits for the USA caverns.

The access to the LEP caverns will be modified (October 2000) for the dismantling of the experiments and during the LHC civil engineering period: the low turnstiles will be removed and the doors opened with a push button, until new card readers (Y2K compliant) are installed.

Ongoing studies concern the hardware of the reader devices and of the access gates:

• All LHC areas are RP controlled areas: the check of the film badges must be done at the surface (AC1 level) and the INB authorities may also request full height turnstiles even for these access points.
• Tokens for the interlocked accesses (AC2/AC3 level): it would be preferable to install a system that allows entry and exit through different gates.
• Operational dosimeter (AC3 level): automatic reading systems are available, but one needs to check if these solutions are cost effective.
• The definition of test areas (e.g. RF cavity, machine cryogenics, etc…) still has to be done, as well as the machine equipment to be interlocked with the access system.

The next LEMIC meeting is scheduled for the 23^rd of November.

S. Weisz