Nb-Ti Symmetric Triplets for the LHC luminosity Upgrade

Abstract

We study a Nb-Ti lay-out for the triplet in the low-beta interaction regions of the Large Hadron Collider, based on a stretched version of the present baseline. The triplet length is increased from the present value of 32 m up to about 60 m. The quadrupoles are based on a two layer coil made with the LHC main dipole cable. A parametric analysis of the dependence of the optics and magnet performances on the triplet length and aperture is carried out. INTRODUCTION The possibility of increasing the focusing in the interaction point of the Large Hadron Collider using a wider and longer Nb-Ti triplet has been considered in several studies [1,2,3]. In this paper we update the results of a parametric analysis developed according to the approach proposed in [4], and presented in [5] and [6]. The triplet lay-out is a stretched version of the today baseline, with quadrupoles of equal gradient and aperture, and different lengths (“symmetric option”). We extend the analysis up to triplet lengths that are ∼25 m longer than the baseline, and we consider quadrupoles made up of two layers of the LHC main dipole cable. Moreover, we improved our analysis in a few points, namely i) we correct an overestimate of the LHC cable performance as given in [6], ii) we use a stronger focusing to have smaller beam size in Q4, iii) we increase the distance between Q2a and Q2b to take into account of the interconnection space needed for magnets in separate cryostats, and iv) we include a scaling of the cold bore thickness with the magnet aperture. The paper presents plots giving the main magnetic and optic properties as a function of the length of the triplet, which is taken as the free parameter. OPTICS CONSTRAINTS Triplet structure We consider a triplet whose structure is similar to the LHC baseline [7], i.e., is made up of two focusing quadrupoles Q1, Q3 of equal length l1, and with two defocusing quadrupoles Q2a and Q2b, each of length l2, in between. We use the nominal distance l=23 m of Q1 from the interaction point (see Fig. 1). With respect to the calculations presented in [5] and [6] we increase the distance between Q2a and Q2b from 1 to 1.6 m to take into account the fact that the two magnets will have a separate cryostat, and not a common one as it is today in the baseline. 0 25 50 Distance from IP (m) Q1