Minutes of the 2nd LHC Insertions Upgrade Working Group held on 27th September 2007

Present: V. Baglin, F. Bordry, O. Brüning, S. Fartoukh, M. Giovannozzi, J.-P. Koutchouk, K.-H. Meß, L. Tavian, E. Todesco, J. Kerby, F. Zimmermann

Excused: R. Ostojic

    1. News and approval of the minutes of the last meeting

Due to the absence of Ranko, SF chaired the meeting.  The creation of the LIUWG web site was announced to the members in a separated e-mail, with no specific comments received during the meeting. The minutes of the last meeting were also approved without any comment.

    2. Vacuum stability issues in the wide-aperture inner triplets (V. Baglin, ppt file)

Vincent started his presentation by recalling the three main physics phenomena which have to be taken into account in order to assess the need of installing beam screens in the new triplets foreseen for the LHC upgrade phase I, namely: the photon, electron and ion stimulated gas desorption, induced respectively by the synchrotron radiation of the off-axis beam inside the triplet, the so-called electron multipacting effect (with a secondary electron emission yield larger than unity) and the interaction of the beam with the residual gas.  In the absence of beam screen, at cryogenic temperature, the gas produced condensates onto the cold bore, forming one or several atomic mono-layers. As a result, the effective desorption yield of the cold bore surface increases considerably, and quickly reaches a threshold where the vacuum stability can no longer be granted. Considering only the photon stimulated desorption, and taking rather optimistic assumptions for the new triplet (a cold bore of 150 mm diameter, and a beam excursion of 5 mm in the inner triplet), Vincent demonstrated that this instability threshold could be reached in a few days of operation, with a gas density exceeding after less than one day the average density target  of 10¹³ m^-3 in the LSS (in the case of hydrogen), and then becoming 10 times higher in only two days.

As in the case of the nominal LHC, the installation of a perforated and actively cooled beam screen is therefore highly recommended in the new triplets. In a first step, Vincent proposes a simple scaling with respect to the present design with a pumping capacity (given by the fraction of  holes per unit of surface) still to be determined depending on the expected gas load for Phase I.

The presentation of Vincent triggered several comments and questions.

• SF questioned the simple scaling proposed by Vincent to define the geometry of the new beam screen, possibly too conservative concerning the difference between the gap height and the diameter of the beam screen which is essentially given by the minimum possible size of the capillaries, the latter varying very slowly with the required helium flow. LT agreed but already anticipated that the thickness of the new beam screens and/or the number of the capillaries might have to be doubled which, in the case of a square beam screen to accommodate four capillaries instead of two, would be detrimental for the mechanical acceptance of the inner triplet. It was generally agreed that Vincent should make as soon as possible a proposal of the beam screen geometry (shape, thickness and clearance with respect to the cold bore).

=> ACTION: Vincent.

• JPK reminded that heavy fluxes of pions are expected on the beam screen surface and that, in general, the debris coming from the IP might become a non negligible source of gas load already for Phase I. Vincent agreed to have a look.

=> ACTION: Vincent.

• Pending a detailed estimate of the gas load expected in the interaction regions for Phase I, which will determine the needed pumping capacity of the new beam screens, the question of the maximum possible fraction of holes per unit of surface was also raised (~ 4% in the present design), possibly impacting on the machine impedance.

=> ACTION: ABP (Elias Metral)

• Frank suggested to ask the LHC experiments for a statement on the acceptable vacuum pressure inside the triplets, i.e. the average density target  of 10¹³ molecules per cubic meter in the LSS as quoted by Vincent. In view of the results obtained by Vincent (see in particular the 9th slide of Vincent's presentation), Stephane however replied that a gas density threshold  even relaxed by a factor of 10, 100 or 1000 would be reached after 2 or 20 and 100 days, respectively. Therefore this would not change substantially the conclusions obtained by Vincent concerning the need of implementing  beam-screens in the new inner triplet quadrupole magnets. Then Frank asked whether the formalism predicting the threshold of the vacuum instability, which had been developed at the time of the ISR, has ever been, or could be, checked at any other machine, e.g. at modern high-intensity proton rings. Following this meeting, Vincent calculated that according to this same formalism one should be able to observe the predicted vacuum instability when switching off vacuum pumps at the SNS. A dedicated MD at the SNS could therefore be envisaged.

=> ACTION: potentially Frank & Vincent.

• Finally, OB mentioned the possible need for absorbers in Q1/Q2/Q3 which, already for Phase I, might have to be  installed  in between the cold bore and beam screen, contrary to the present design where only one absorber has been accommodated in the 2K helium bath of Q1. Therefore, Oliver asks the possible impact on the pumping capacity of the new beam screens. Vincent replied that depending on the temperature (e.g. if actively cooled down) and  the geometry of this absorber, the impact could be more or less dramatic. After the meeting, SF and VB agreed to start investigating the alternative of a very thick beam screen with therefore an increased density of holes, and possibly multi-layers to play also the role of absorber.

    3. IR aperture requirements versus beta* (S. Fartoukh, ppt file)

SF presented the main aperture requirements of LSS1 and LSS5 with the target of n1~9 for the normalized aperture of the magnets and a b^* of ~25 cm in IP1 and IP5. Due to strong aperture limitations in the LSS magnet, it turns out that this analysis allows as well to obtain an upper and lower bound for the aperture of  the inner triplet. The main results obtained are summarized below:

• Q6-Q13: from Q6 onwards, no specific changes (magnets, beam screens, power supplies, layout,...) are a priori required.

• Q5: the beam screen orientation  needs to be modified in both apertures of all Q5 magnets (otherwise their aperture will be in between n1=6 and n1=7 for the squeezed optics). As a result, the mechanical acceptance of Q5 will degrade at injection. The remedy could be to change b^*  at injection to 8-9 m (compared to 11 m for the LHC injection optics V6.500) which is fully compatible with the increased aperture of the new inner triplet quadrupoles (see below).

• Q4-D2: the  beam screen orientation  needs to be modified in one of the two apertures of each D2 and Q4 magnet, in which case the aperture will be optimum both for injection and collision optics. Contrary to Q5, this modification could even be performed before commissioning of the nominal LHC.

• D1: Even using highly unrealistic assumptions to derive the aperture requirements for the D1 magnets, 10-15 mm are missing in the gap height of the present D1. It was generally agreed that this magnet needs to be  modified. Assuming that the normal conducting option is kept, the gap height of D1 needs to be increased from 63 mm to 90 mm or 100 mm, for a horizontal or vertical beam-crossing at the IP, respectively.

• Q1-Q3: In order to warrant a normalized aperture of n1=9 both in the LSS magnets and in the inner triplet, the  optimum inner coil diameter of the Q2 and Q3 quadrupoles  is equal to 113 mm (compatible with  an operating gradient of about 140 T/m for NbTi magnets). An additional aperture margin, say D_margin, still to be specified, might however be needed for a thicker beam screen, or cold absorbers (see the discussion following the presentation by Vincent). Then, if the triplet aperture is greater than 113 mm + D_margin, and there are no changes in D2-Q4-Q5 (but the reorientation of the beam-screens mentioned above), the Phase I machine performance will be limited by the mechanical acceptance of the matching section and will degrade (lower gradient and longer triplet resulting in an increase of the beta functions at least up to Q4). On the other hand, assuming that the aperture of the two-in-one Q4 quadrupoles can reach 90 mm and that D2's, Q4's and Q5's are replaced by wider aperture magnets ("Phase Ib", see details in the Stephane's slides) while keeping the "Phase Ia" NbTi inner triplet quadrupoles, the "final" performance of the machine will degrade if the triplet aperture is greater than 145 mm+D_margin.

• Other equipments: The aperture of all other equipments related to the triplet must be reviewed, e.g. TAS, DFBX, BPM, triplet corrector packages, but also the tertiary collimators in between D1 and D2.

Following the  presentation, there was a clear consensus that D1 dipoles need to be replaced, possibly by cold magnets satisfying the aperture requirements of phase II upgrade (140-150 mm coil aperture). The hardware modifications of Q5, Q4 and D2 were also thought to be possible in the background of other activities. However,  OB reminded that the latter actions must be scheduled within the three month shut-down period which is currently foreseen for installation of other equipment in IR1 and IR5 (inner triplets, D1, absorbers). The range derived by Stephane for the triplet aperture was found relevant and triggered two important questions:

• What is the maximum aperture of the LHC two-in-one quadrupoles (i.e. the 90 mm guessed by Stephane)?

=> ACTION: Ezio, Jean-Pierre.

• Are we sure that installing cold absorbers in all triplet quadrupoles Q1/Q2/Q3 (requiring D_margin) is the most effective way of protecting the triplet?

This later issue could not be answered and requires input from specialists. This important point will be discussed in one of the next meetings.

    4. A.O.B.

The next meeting is scheduled for 18 October 2007. Tentative agenda: Conceptual design of the triplet cryostats and interconnects (V. Parma);  Possible optics solutions for Phase I (R. De Maria).

S. Fartoukh and R. Ostojic