The summer of luminosity production is well under way. As of 19 August, the LHC had delivered 55 fb⁻¹ to both ATLAS and CMS, compared with the initial target for this date of 60 fb⁻¹. This means we are still slightly behind the predicted luminosity curve. The gap is slowly closing, but we are not there yet. When the LHC is running smoothly, the production curve rises more steeply than forecast, and catching up can take just a week or two. Unfortunately, several periods of beam stop have slowed progress, caused by a variety of issues within the LHC and the injector chain (e.g. electrical disturbances, loss of cryogenic conditions, and the failure of an SPS magnet that required replacement). One particularly intriguing event may have simply been a coincidence, but it nevertheless raised some questions. At 1.25 a.m. on 30 July, a powerful 8.8-magnitude earthquake struck off the eastern coast of Kamchatka in the Russian Far East, just as the LHC was preparing to inject beam. Just after 2.00 a.m., the orbit of the freshly injected beams began to oscillate, most noticeably in the horizontal plane, as illustrated in the image below. Unfortunately, this beam was dumped around 4.45 a.m. due to a disturbance in the electrical network, which caused some issues at Point 8 and required an operation in the tunnel. During the orbit excursions, beam losses were also recorded. These were traced to a sudden and brief rise in vacuum pressure of about a hundredfold near Point 2 but on the side towards Point 1. After the spike, the pressure gradually decreased again. Similar vacuum activity was observed in subsequent fills, prompting the vacuum experts to request access in order to take X-ray images of the affected part of the LHC machine. The X-ray scans revealed a problem in a radiofrequency (RF) finger module, a component that ensures electrical continuity at the join between two vacuum chambers. The images showed that the spring, which should press the fingers tightly against the chamber wall, had lost its tension and no longer provided reliable contact. Without this contact, electrical sparks can occur as the beam passes, which in turn trigger the sudden spikes in vacuum pressure. This may sound familiar, as we faced a similar issue on 25 May 2023 when two successive beam dumps occurred during acceleration due to slow local beam losses. The source was ultimately traced to an RF finger module in a warm section near Point 1 that was heating up, or arcing. At that time, the beam performance was limited by the issue, and the RF finger module was replaced in the days following the event. Only five days after the event occurred, beams were back in the LHC. As a preventive measure, many of the most critical RF finger modules were replaced during the subsequent year-end technical stop with improved versions designed for the HL-LHC. However, not all the modules could be changed, including the one responsible for this month’s issue. This time, following the detailed procedure established after the 2023 event, the LHC was initially filled with fewer bunches, while vacuum activity was closely monitored by experts. A careful, stepwise ramp-up in the number of bunches then took place. Some vacuum activity was observed, but never to a level requiring the immediate replacement of the RF finger module. The LHC is now running with this non-compliant RF finger module at nominal performance as planned, with 2460 bunches per beam, each containing 1.6×10¹¹ protons at the start of collision. The vacuum activity remains low and acceptable and, in some cycles, no noticeable activity at all is observed. What will happen to the module is still to be determined. While repeated X-ray inspections have not shown any change in its condition, its behaviour could still evolve over time. Plans for the final quarter of 2025 also include an increase in the number of protons per bunch. The LHC Machine Committee (LMC) together with the ATS management will now take these considerations into account and decide on the best course of action. One final question remains: was the timing of the RF finger module issue pure coincidence, or triggered by the Kamchatka earthquake and a subtle shaking of the LHC tunnel and machine? Personally, I lean towards coincidence, but we will never be entirely sure…