High-order corrector magnets will be required for the magnetic system of the HL-LHC inner triplets. These magnets are based on a superferric design thus the saturation of the iron poles affects the field generated in the aperture, i.e., the magnetic transfer function shows a nonlinearity. One of the challenges for the operations of these magnets is to find a suitable fit of the magnetic transfer function able to predict the field generated, given the current, within the acceptable level of 1%. In the LHC, the magnet operations rely on a magnetic field model (FiDeL) for deriving the current level from the required field strength. This paper presents a first iteration of the field modelling for the new high-order corrector magnets.

The MQXFB magnets are superconducting quadrupoles with nominal peak field on the conductor of 11.3 T. With their magnetic length of 7.2 m, they stand as the longest Nb3Sn accelerator magnets designed and manufactured up to now. Together with the companion MQXFA 4.2 m long units, built by the US Accelerator Research Program, they are at the heart of HL-LHC, as they shall replace the inner triplet quadrupoles at either side of the ATLAS and CMS interaction regions of the LHC. This technology has benefited from many years of development, and this specific design was validated with successful short models (MQXFS, 1.2 m long). More recently, several MQXFA magnets were shown to satisfy HL-LHC requirements. In this paper, we report on the cold test results of four MQXFB magnets, focusing on performance, training, behavior after thermal and powering cycles, and field quality. We then provide an update of the overall status, including ongoing verifications of design changes at the level of the coil fabrication.