Arkadiusz Gorzawski

Beam Loss Measurements for Recurring Fast Loss Events During 2017 LHC Operation Possibly Caused by Macroparticles

The availability of the LHC machine was adversely affected in 2017 by tens of beam aborts provoked by frequent loss events in one standard arc cell (16L2). In most of the cases, the dumps were triggered by concurrently developing fast beam instabilities leading to particle losses in the betatron cleaning insertion. Many of the events started with a distinct sub-millisecond loss peak comparable to regular dust particle events, which have been observed along all the LHC since the start-up. In contrast to regular dust events, persistent losses developed in cell 16L2 after the initial peaks which can possibly be explained by a phase transition of macroparticles to the gas phase. In this paper, we summarize the observed loss characteristics such as spatial loss pattern and time profiles measured by Beam Loss Monitors (ionization chambers). Based on the measurements, we estimate the energy deposition in macroparticles and reconstruct proton loss rates as well as the gas densities after the phase transition. Differences between regular dust events and events in 16L2 are highlighted and the ability to induce magnet quenches is discussed.

Beam Offset Stabilization Techniques for the LHC Collision Points

Maintaining head–on collisions over many hours is an important aspect of optimizing the performance of a collider. For the current LHC operation where the beam optics is fixed during periods of colliding beam, mainly ground motion induced perturbations have to be compensated. The situation will become significantly more complex when luminosity leveling will be applied following the LHC luminosity upgrades. During β* leveling the optics in the interaction region changes significantly, feed–downs from quadrupole misalignment may induce significant orbit changes that may lead to beam offsets at the collision points. Such beam offsets induce a loss of luminosity and reduce the stability margins for collective effects that is provided by head–on beam–beam. It is therefore essential that the beam offsets at the collision points are minimized during the leveling process. This paper will review sources and mitigation techniques for the orbit perturbation at the collision points during β* leveling, and present results of experiments performed at the LHC to mitigate and compensate such offsets. BEAM POSITION AT THE INTERACTION POINT The stability of the beam position at the LHC interaction points (IPs) must be at the level of one rms beam size or better in the most critical phases within the operation cycle. The most critical dynamic phases concern future modes of operation that imply a mixture of dynamic optics changes and colliding beams. Luminosity leveling by β*, where the betatron function at the IP is adjusted with colliding beams during experiments data taking, is one of those modes. Colliding the beams during the betatron squeeze (collide– and–squeeze) to profit from the Landau damping from headon beam-beam collision is another mode. In both cases the beams must remain colliding head-on within roughly one rms beam size. The main disturbances of the beam positions are ground motion, noise sources that generate orbit drifts as well as feed-down from the changing quadrupole gradients during the optics changes that take place during β* leveling or collide–and–squeeze. The disturbances are compensated by the LHC orbit feedback. Some residual perturbations of the IP position can however not be excluded, for example when some beam position monitors malfunction and due to the limited accuracy of the beam position monitors. Table 1 presents simulated IP beam separations that can build up during the full optics change associated with β* leveling or collide–and–squeeze. For quadrupole rms misalignments of δQ of 100 μm, the separation reaches tens of rms beam sizes. Table 1: Impact of the initial misalignment and BPM errors on the beam separation dIP when the LHC optics is changed from injection to physics β* (squeeze). Different assumptions are presented for the perturbation and the correction (with or without the common orbit correctors near the IP – MCBX). r.m.s. δQ BPM errors Used MCBX Max dIP 100 μm 20 μm no 11σ 100 μm 20 μm yes 6σ 100 μm 100 μm yes 20σ Although the values presented in Tab. 1 indicate huge beam separations at the IPs, the real shifts from ground motion are much smaller. An analysis of the orbit drifts and orbit corrector strengths in LHC Run 1 (2011–2012) [1] indicates that the beam separation due to ground motion does not exceed 0.1σ on the time scale of 12 hours, a typical inter-fill time. Ground motion is therefore not expected to be an issue, but the beam position monitor reproducibility of around 50 μm and possible larger reading outli ers can have a more important impact. During LHC operation in 2015 large and unexpected orbit drifts were observed. The origin was rather quickly localized in one of the LHC low-beta quadruples in IP8 (LHCb experiment) whose radial position was oscillating by around 30 μm] with a period of 8 hours [2]. The root cause of the movement was eventually tracked down to the regulation of the thermal shield of the quadrupole (mal-functioning valve). This effect caused some luminosity oscillations until an orbit feedback was introduced during physics data taking. FEEDBACK ON LUMINOSITY The principle of the method presented here to measure and correct possible orbit shifts is based on a small modulation of the luminosity that is superposed to the beam orbit. The modulation consists of a circular beam position scan, a rotation scan: a circular rotation of one beam around the other (with a small separation – scan radius) will cause a modulation of the luminosity. In case the beam position is stable the luminosity is constant. If an offset is present between the two beams the luminosity will be modulated, the offset and its evolution is encoded in the modulation amplitude and phase. The scan concept is illustrated in Fig. 1: the direction α, initial value and average speed v∆ of the drift ∆(t) may be extract from the time evolution of the luminosity. TUPMW012 Proceedings of IPAC2016, Busan, Korea ISBN 978-3-95450-147-2 1438 C op yr ig ht