CERN Accelerating science

LHC TDI - injection absorbers

Concerning the control in general, and the alignment method in particular, for the TDI, TCLI, TCDQ and TCSG (IR6) collimators, it should be clearly born in mind that these are part of the machine protection system and not part of the collimation system. These collimators will need to be set up independantly to positions which are determined by the protection requirements, not by any beam cleaning issues. It will not be possible to have these collimators in a control system which will adjust their positions for beam cleaning reasons, however cconvenient this might seem. As I see it, these devices are set up according to the protection requirements, and the collimation system proper must then be set up in order that the beam load on these protection elements (like for the rest of the machine) is acceptable. There is no question that these protection devices will be included in some kind of general collimator adjustment application, with feedback from BLMs etc. We will set them up together - the injection protection elements TDI and TCLI together, and the dump protection elements TCDQ and TCSG together. Our baseline for the setting up process for all these elements assumes that we will do this EITHER by a transmission/remote loss measurement for a single injected bunch, OR by scraping through a circulating beam and deriving the beam size and centre by the resulting intensity vs position curve, OR finally by loss levels recorded on local BLMs. Hence we certainly do not want that the TCLI and TCSG can only be set up using this local BLM loss feedback that is (I think) intended for the main collimator system.

SSM (safe settings management) has no role in the 'control' per se of the protection elements; we have to have some reference values resident in the front-ends, in order to be able to generate interlocks if the jaw positions are wrong - SSM is the high level application which is being discussed to manage these references (for many other systems, not just protection devices). We already discussed this concept concerning the settings for the extraction/transfer/injection elements, and Mike will make a proposal about how this is to be done. However (contrary to what is stated in the minutes) I don't think it's the Cocost meeting where this should be developed/followed up, since it is a general issue for all machine-protection related equipment.

c/o Brennan

Information sources

Objectives

Explore use of LHC TDI during operational cycle in an attempt to identify control requirements.

Background

The TDI absorber - 2* 4.2 m jaws each of which is adjusted by two independent step motors.

The TDI must be accurately aligned to the vertical crossing angle orbit (IR2), separation bump (IR8). For the current crossing angle separation scheme (Version 6 of the LHC lattice) the vertical orbit changes by 0.2mm over the length of the TDI which is of the same magnitude as the required alignment accuracy.

Quench limit : 38 mJ cm3   Damage limit: 87 Jcm 3
Operating conditions
"Nominal bunch : 1:05 10¹¹ protons, every 16.8 at  32 mm from edge.
Full 2:55 1013 protons or ultimate batch 4:13 10¹³ p on various spots of TDI.

• CASE 1: at some 32 mm from edge "on purpose" or full MKI failure (e.g. no trigger). Estimated
  rate: several/yr.
• 75% kick strength - one module missing - 23 mm impact parameter
• CASE 3: Grazing impact, swept in case of wrong MKI timing, i.e. around 10 bunches bunches down the pipe, some on the edge, many at 32 mm). Estimated rate: several/yr. OR more or less grazing impact of bulk of bunches (with initial sweep) in case of internal fault of 1 module (flash over) . Estimated rate: practically excluded, 1/10 yrs (?).
• 25% kick strength - beam at some 4 mm from edge in case pre-fire of 1 out of 4 modules. 1/yr (?).

The injection protection scheme presented by Volker Mertens requires an orbit control within 0.5 sigma of the beam size at the TDI (-> 0.2mm). This orbit accuracy must be provided during the whole injection process and requires either a good knowledge of the effect of the persistent current decay on the orbit or requires an active orbit feedback system during the injection process. The TDI absorber and the collimators must be accurately positioned without having a circulating beam as a reference for the absorber position.

The TDI absorber - 2 jaws independently controlled each is adjusted by two independent step motors. The TDI must be accurately aligned to the vertical crossing angle orbit (IR2) and vertical separation bump (IR8).

Operational Use Case (first pass)

Pilot

• Initially the TDI will be retracted, not necessarily fully out but far enough to avoid affecting injected beam. This "parking" position will be at least 10 s or around a 5 mm retraction.
• Per ring: the closed orbit of the pilot will be measured, corrected and re-measured. The TDI position and angle will then be adjusted with respect to the closed orbit. They will be positioned at something like 8.5 sigma with a accuracy of 0.58 sigma.
• Given nominal emittance and anticipated b value at TDI this translates to a required resolution of 0.1 mm. The ability to set to 0.1 mm was thought to be possible but not incrementally because of the large inertia of the system. To go to a new position accurately the TDI would need to be retracted somewhat and then set.

TDI position: 5.1 +/- 0.3mm (10 +/-0.5 s)

Dense

• Before high intensity beam is injected the TDI must be in position. An interlock must be thrown if not.

After injection before ramp

• The TDI must be withdrawn to its parking position. If left at injection setting, however, it shouldn't be an problem.