The Linac4, PS Booster and PS all achieved remarkable reliability last week, with beams delivered to users 99% of the time. At the SPS, beam availability was also very good until 12 September, when an electrical fault in one of the static VAR compensators (devices that stabilise the electrical supply) caused a major interruption. The fault triggered a 26-hour downtime of the SPS and required the reconfiguration of a spare static VAR compensator. Thanks to the rapid and coordinated efforts of experts from several groups, beam operation was successfully restored in the early afternoon of 13 September.
The LHC switched to machine development (MD) mode on 1 September, running a demanding programme of studies until 5 September. Some of these studies involved testing HL-LHC-like beams, with bunch intensities similar to those planned for the high-luminosity upgrade but with a lower number of bunches. Although the overall beam intensity remained below operational limits, the tests produced some vacuum pressure rises in the non-compliant RF finger module. As a precaution, the MD team postponed further high-bunch-intensity beam tests. The non-compliant RF finger module will be replaced during the next year-end technical stop. Encouragingly, X-ray inspections carried out after the MD block confirmed no further degradation of the RF finger module, which is reassuring for the safe continuation of beam operation and luminosity production.
The return to luminosity production was delayed by another challenge: a vacuum leak in one of the LHC secondary collimators at Point 7. Collimators play a crucial role in cleaning the beam halo and protecting the machine but, because they absorb high-energy particles, they become radioactive over time. Radiation protection experts carefully determined the cooldown time required before repair work could take place. Fortunately, the preceding days of running with lower-intensity MD beams had already reduced radiation levels, allowing the repair to begin sooner.
On Friday, 5 September, specialists entered the tunnel, opened and blocked the collimator jaws, securing free passage for the beam. Subsequently, the leaking bellow was sealed with varnish that usually holds well, provided the bellow is not moved (hence the blocking of the jaws). After successful vacuum pumping, the machine was tested to ensure safe operation without the affected collimator. Dedicated "loss maps" confirmed that other collimators and protection devices could take over its role, enabling safe operation to continue.
With these issues resolved and the staged intensity ramp-up completed, the first full luminosity production fill, with 2460 bunches per beam, was successfully injected and brought into collision on 9 September, marking a strong return to physics, a few days later than scheduled.
