⚛️ high-energy experiments

Operation and performance of the Probe for Luminosity Measurement at LHCb

This paper reports on the operational performance of the PLUME detector, a dedicated Cherenkov-light-based luminosity monitor at LHCb, during the 2024–2026 Run 3 data-taking period, highlighting its role in real-time luminosity control and recording an integrated luminosity of $(26.71 \pm 1.07)$ fb $^{-1}$ at $\sqrt{s} = 13.6$ TeV.

Original authors: F. Alessio, S. Barsuk, C. Beigbeder, A. Bellavista, O. Bezshyyko, I. Boiaryntseva, C. Bourgeois, A. Boyarintsev, M. Briere, L. Burmistrov, A. Carbone, G. Cavallero, V. Chaumat, S. Cholak, D. Douillet

Published 2026-09-10

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: F. Alessio, S. Barsuk, C. Beigbeder, A. Bellavista, O. Bezshyyko, I. Boiaryntseva, C. Bourgeois, A. Boyarintsev, M. Briere, L. Burmistrov, A. Carbone, G. Cavallero, V. Chaumat, S. Cholak, D. Douillet, O. Duarte, P. Durante, F. Ferrari, E. Franzoso, C. Gaspar, L. Golinka-Bezshyyko, L. A. Granado Cardoso, E. Graverini, D. Hohov, G. Iaquaniello, V. Kushnir, R. Matev, P. Mayencourt, V. Mytrochenko, V. Orlov, V. Puill, D. Reynet, P. Robbe, P. Rosier, L. Shchutska, Y. Song, E. Spedicato, L. Toscano, M. van Dijk, A. Villa, G. Vouters, V. Yeroshenko, V. Zhovkovska

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Inside the vast, circular tunnel of the Large Hadron Collider at CERN, two beams of protons race in opposite directions at nearly the speed of light. When these beams collide, they create a shower of new particles that scientists study to understand the fundamental laws of nature. To make sense of these collisions, researchers need to know exactly how many times the protons hit each other. This measure, called luminosity, is not just a number; it is the key that unlocks the value of every experiment. Without a precise count of collisions, scientists cannot tell if a rare event they observe is a genuine discovery or merely a statistical fluke. Because the LHCb experiment focuses on rare processes involving heavy particles, it must operate with extreme precision, keeping the collision rate steady and well-understood. If the rate fluctuates too much, the delicate detectors can become overwhelmed, losing the ability to record the subtle signals they are designed to find.

To solve this problem, a specialized team has built and operated a dedicated monitor called PLUME, which stands for Probe for LUminosity MEasurement. This device acts as a highly sensitive counter, sitting near the point where the proton beams crash together. Instead of trying to catch the particles themselves, PLUME watches for a faint flash of light known as Cherenkov radiation. This light is produced when charged particles, created by the collisions, travel through a block of fused silica glass faster than light can travel through that same glass. The monitor is equipped with forty-eight photomultiplier tubes, which are essentially ultra-sensitive eyes capable of detecting single photons of light. By counting these flashes of light, the system can calculate exactly how many collisions are happening at any given moment, providing a real-time feed of data that keeps the entire experiment running smoothly.

The paper details how this system performed during the 2024 to 2026 data-taking period, a time when the LHCb experiment was running at its highest capacity yet. The researchers report that the detector worked with remarkable stability, successfully tracking the collisions as the machine delivered a total of 26.71 inverse femtobarns of data. This unit of measurement represents the total amount of collision data collected, and the team calculated this figure with a high degree of confidence, accounting for various sources of error. The system proved so reliable that it became the primary tool for controlling the luminosity leveling process, a procedure that adjusts the beam intensity to ensure the detectors are never overwhelmed. The team also demonstrated that the system could measure the timing of the collisions with incredible precision, down to a few billionths of a second, which is essential for synchronizing the different parts of the massive detector.

Beyond just counting collisions, the PLUME detector faced the challenge of aging in a harsh environment. The intense radiation near the collision point can damage electronic components over time, causing them to become less sensitive. To counter this, the team developed a sophisticated calibration system. They used a combination of light pulses and direct measurements of the electrical current flowing through the tubes to constantly adjust the voltage. This allowed them to keep the sensitivity of the detectors perfectly tuned, even as the equipment accumulated significant radiation damage. The study shows that the system could maintain its accuracy to within a few percent, ensuring that the data collected over several years remained consistent and trustworthy.

The researchers also explored ways to keep the experiment running even if the primary counting system faced technical difficulties. They developed a backup method that measured the electrical current drawn by the photomultiplier tubes directly, rather than counting individual light flashes. While this backup method was slightly less precise because it could not easily distinguish between collisions in the main beam and those in a separate gas cell, it provided a crucial safety net. The team found that this alternative approach could still provide a reliable estimate of the collision rate, ensuring that the experiment would not lose valuable data during unexpected interruptions.

In addition to counting, the PLUME system played a vital role in keeping time. The LHC operates on a strict clock cycle, and the detectors must be perfectly synchronized with the arrival of the proton bunches. The team used the PLUME detector to measure the exact moment the collisions occurred relative to the machine's master clock. They found that the detector could track these timing shifts with a precision of about eight picoseconds, which is a tiny fraction of a nanosecond. This level of precision allowed the operators to correct for any drift in the system, ensuring that the detectors were always looking at the right moment. The measurements from PLUME agreed well with other independent timing systems, confirming that the detector was providing a reliable and independent check on the experiment's timing.

The paper concludes that the PLUME detector has successfully fulfilled its role as a robust and versatile instrument for the LHCb experiment. It has provided a continuous, real-time stream of luminosity data that has been essential for the stable operation of the machine. The team's work demonstrates that with careful calibration and monitoring, even detectors operating in extreme radiation environments can maintain high precision over long periods. The integrated luminosity recorded by the system serves as a foundation for all the physics results that will be published from this data-taking period. While the final numbers used for scientific papers will come from more detailed offline analyses, the online measurements from PLUME have already proven their worth in keeping the experiment running efficiently and safely. The success of this system highlights the importance of dedicated monitoring tools in modern particle physics, where the ability to measure the intensity and timing of collisions is just as critical as the ability to detect the particles themselves.

Technical Summary: Operation and Performance of the Probe for Luminosity Measurement (PLUME) at LHCb

Problem Statement
The LHCb experiment, dedicated to flavour physics, operates in a high-occupancy forward acceptance environment. During Run 3, the detector operates at luminosities approximately five times higher than in previous runs. To maintain optimal detector performance and manage high interaction rates, the instantaneous luminosity delivered to LHCb is stabilized via a dedicated luminosity-levelling procedure. This process requires a precise, real-time, bunch-by-bunch determination of the instantaneous luminosity. Furthermore, variations in the LHC clock phase due to temperature fluctuations and fibre length changes can lead to timing misalignments between sub-detectors, causing efficiency losses. A dedicated system was required to provide low-latency luminosity monitoring and independent beam-phase timing measurements to ensure stable data-taking conditions.

Methodology
The Probe for LUminosity MEasurement (PLUME) is a hodoscope detector installed between $z = -1485$ mm and $z = -2085$ mm along the beam axis, covering a pseudorapidity range of $-2.4 < \eta < -3.1$ . The detector comprises 48 Hamamatsu R760 photomultiplier tubes (PMTs) arranged in a cross-shaped, two-layer configuration. Each PMT is coupled to a 5 mm fused silica tablet to enhance the yield of Cherenkov light produced by charged particles emerging from the collision region.

Luminosity Determination: PLUME utilizes the "logZero" method, calculating the average number of interactions per bunch crossing ( $\mu$ ) based on the probability of detecting zero events ( $P(0)$ ). To ensure accuracy, the system performs background subtraction to remove contributions from the SMOG2 fixed-target gas cell and beam-gas interactions. The instantaneous luminosity ( $L_{inst}$ ) is derived from the background-subtracted interaction rate and a visible cross-section ( $\sigma_{vis}$ ) determined via van der Meer (vdM) scans and emittance scans.
Timing Monitoring: Four of the 48 PMTs are dedicated to measuring the beam-phase timing. The analogue signal from these PMTs is split into eight replicas, each delayed by 3 ns, allowing the leading edge of the pulse to be sampled at multiple time offsets. The arrival time is reconstructed offline by fitting the cumulative response with an error function to determine the inflection point.
Electronics and Readout: The detector employs Front-End (FE) boards identical to those used for the LHCb calorimeter, utilizing the ICECAL ASIC for signal amplification and integration. Data are processed by TELL40 back-end boards hosting Intel Arria 10 FPGAs. Custom firmware handles pedestal subtraction, thresholding, and the accumulation of counts per bunch crossing. A dedicated monitoring system uses LED pulses injected via quartz fibres to calibrate PMT gain, supplemented by a data-driven technique using the TELL40 RAM memories for higher statistics.
Redundancy: An alternative luminosity monitoring method based on PMT anode currents was developed to ensure operational continuity during non-nominal conditions (e.g., TFC signal loss), though it is limited by the inability to subtract SMOG2 contributions.

Key Contributions

Detector Implementation: Successful commissioning and operation of a 48-channel PMT-based hodoscope for real-time luminosity and timing monitoring in the harsh radiation environment of the LHCb interaction point.
Firmware Development: Implementation of custom firmware on TELL40 boards to perform online luminosity calculation with a latency of 2.4 seconds (well within the 3.0-second LHC requirement) and high-precision timing reconstruction.
Calibration Strategies: Development of a dual-approach gain calibration system combining LED-based monitoring with a data-driven method utilizing full-rate TELL40 readout to overcome statistical limitations during short fills or low-rate collisions.
System Integration: Integration of PLUME into the LHCb Experiment Control System (ECS) and Data Acquisition (DAQ) framework, enabling both low-latency feedback for luminosity levelling and offline analysis for integrated luminosity determination.

Results
During the 2024–2026 data-taking period of Run 3, PLUME operated stably, providing luminosity measurements for over 99% of the stable-beam time.

Integrated Luminosity: The total integrated luminosity recorded by LHCb in $pp$ collisions at $\sqrt{s} = 13.6$ TeV, as measured by PLUME online counters for performance monitoring, is $L = (26.71 \pm 1.07) \text{ fb}^{-1}$ . This breaks down to $9.56 \pm 0.39 \text{ fb}^{-1}$ (2024), $11.81 \pm 0.48 \text{ fb}^{-1}$ (2025), and $5.34 \pm 0.22 \text{ fb}^{-1}$ (2026).
Uncertainty: The total relative systematic uncertainty on the online luminosity determination is 4.0%. The dominant contribution (3.1%) arises from the determination of the visible cross-sections during vdM scans. Other contributions include gain calibration (0.4%), compatibility of measurements (1.7%), and consistency among online counters (1.9%).
Gain Stability: The PMT gain was maintained with a precision of approximately 1% relative to the target value. High-voltage adjustments were sufficient to maintain constant gain, with extrapolations indicating that maximum voltage limits would not be reached until well beyond the expected exposure of Run 4.
Timing Performance: The PLUME timing measurement achieved a resolution of approximately 8 ps. The measured collision-clock phase difference showed good overall agreement with the Beam Pick-Up Timing (BPTX) system, though differences up to 200 ps were observed, attributed to the different physical quantities measured by the two systems.

Significance
The paper establishes PLUME as a reliable and versatile instrument for the LHCb experiment during Run 3. It successfully fulfills its primary role as the detector for controlling the luminosity levelling process, providing the necessary real-time feedback to stabilize the instantaneous luminosity. Additionally, PLUME serves as a critical tool for continuous timing alignment, offering an independent measurement of the beam phase that complements existing beam instrumentation. The paper notes that while PLUME provides the online luminosity values used for detector performance monitoring, the luminosity values used in physics analyses are determined separately through dedicated offline calibrations and will be reported in a separate publication. The operational stability and performance demonstrated during Run 3 validate the detector design and the associated readout and calibration systems for future high-luminosity operations.

More like this