Review of BLM thresholds at tertiary LHC collimators

The Large Hadron Collider is designed to accelerate protons at the unprecedented energy of 7 TeV. With a total stored energy of 360 MJ, even tiny losses can cause machine downtime or induce damage to sensitive accelerator components. The Beam Loss Monitors (BLMs) are an important component of the complex LHC protection system. They consist of a series of ionisation chambers located all around the ring to detect secondary particle showers induced by beam losses. The monitors are assigned thresholds such that if the radiation generated by the loss is too high, the BLM triggers a beam dump, preventing the loss to grow excessively. BLM signals are recorded for different integration windows, in order to detect losses on very different time scales, ranging from the extremely short ones (taking place over half a turn) to those very close to steady state (i.e. lasting for more than a minute). The LHC is equipped with a complex collimation system, to provide the machine with passive protection in case of transient losses. Among the different families populating the system, the tertiary collimators (TCTs) are located close to the experiments to protect the magnets needed to squeeze the colliding beams. These collimators are made of tungsten to maximise absorption capabilities at the expenses of robustness. Thresholds at collimator BLMs, aimed at preventing damage to the jaws, have been first set based on simulations and empirical scaling laws, and then optimized based on operational experience as a trade-off between the required protection of the metallic collimators and the rate of spurious beam abort triggers. This work reviews and proposes further optimisation of the current thresholds of the BLMs at the TCTs. The review is accomplished by means of numerical simulations, where a single TCT collimator is set as aperture bottleneck and the losses concentrate there. Two steps are carried out; in the first one, the population of protons hitting the collimator is evaluated by means of cleaning simulations, where single-particle beam dynamics and particle-matter interactions are taken into account. The second step consists of the actual energy deposition calculations carried out by means of a Monte Carlo transport code, for the evaluation of the peak energy deposition in the collimator jaw and the corresponding BLM signal. Thanks to these two quantities, and knowing the maximum energy deposition that a TCT can stand before experiencing damage in different time domains, it is then possible to compute the BLM thresholds on the different integration windows. The work is complemented by a benchmark of the simulation results against measurements gathered in 2016 and 2017. This allows to verify experimentally the BLM response per hitting proton, for a couple of scenarios of controlled losses on different collimators.