The DMAPS upgrade of the Belle II vertex detector Babeluk, M.; Barbero, M.; Baudot, J. ...
Nuclear instruments & methods in physics research. Section A, Accelerators, spectrometers, detectors and associated equipment,
07/2024, Volume:
1064
Journal Article
Peer reviewed
The Belle II experiment at KEK in Japan considers an upgrade for the vertex detector system in line with the accelerator upgrade for higher luminosity at long shutdown 2 planned for 2028.
One ...proposal for the upgrade of the vertex detector called VTX aims to improve background robustness and reduce occupancy using small and fast pixels. VTX accommodates the OBELIX depleted monolithic active CMOS pixel sensor (DMAPS) on all five proposed layers. OBELIX is specifically developed for the VTX application and based on the TJ-Monopix2 chip initially developed to meet the requirements of the outer layers of the ATLAS inner tracker (ITk).
This paper will review recent tests of the TJ-Monopix2 chip as well as various design aspects of the OBELIX-1 chip currently under development.
The OBELIX chip for the Belle II VTX upgrade Babeluk, M.; Auguste, D.; Barbero, M. ...
Nuclear instruments & methods in physics research. Section A, Accelerators, spectrometers, detectors and associated equipment,
October 2024, Volume:
1067
Journal Article
Peer reviewed
The OBELIX depleted monolithic active CMOS pixel sensor (DMAPS) is currently developed for the upgrade of the vertex detector of the Belle II experiment located at Tsukuba/Japan. The pixel matrix of ...OBELIX is inherited from the TJ-Monopix2 chip, but the periphery includes additional features to improve performance and allow the integration into a larger detector system. The new features include a trigger unit to process trigger signals, a precision timing module and a possibility to transmit low granularity hit information with low latency to contribute to the Belle II trigger. Additionally, low dropout voltage regulators and an ADC to monitor power consumption and substrate temperature is developed. This paper will focus on the trigger contribution capabilities of the OBELIX chip.
The Belle II experiment searches for beyond-the-standard-model physics using the Belle II detector and SuperKEKB collider. The silicon vertex detector (SVD) is crucial for particle tracking. After ...the 1.5-year shutdown from June 2022, Run 2 began in January 2024; Run 2 operation shows stable noise levels, high signal-to-noise ratios, and hit efficiency over 99%. To manage higher beam background from increased luminosity, new techniques such as hit-time selection and cluster grouping are being developed. These methods increase the acceptable level of occupancy by distinguishing hits from triggered collisions and other sources.
The silicon vertex detector of the Belle II experiment Gabrielli, A.; Adamczyk, K.; Aihara, H. ...
Nuclear instruments & methods in physics research. Section A, Accelerators, spectrometers, detectors and associated equipment,
07/2024, Volume:
1064
Journal Article
Peer reviewed
The silicon vertex detector (SVD) is a four-layer double-sided strip detector installed at the heart of the Belle II experiment, taking data at the high-luminosity B-Factory SuperKEKB since 2019. SVD ...has been operating smoothly and reliably, showing a stable and above-99% hit efficiency, and a large signal-to-noise ratio in all sensors. In June 2022 the data-taking of the Belle II experiment was stopped for the Long Shutdown 1, primarily required to complete the vertex detector (VXD) with the inner two-layer DEPFET detector and to upgrade several components of the accelerator. This article reports on the excellent performance of SVD in terms of the signal-to-noise ratio, the hit position resolution, as well as the hit-time resolution. We briefly describe the challenges and delicate phases of the VXD re-installation and the SVD status for operation starting in early 2024. In SVD layer 3, which is closest to the interaction point, the average occupancy has been less 0.5%, well below the estimated limit for acceptable tracking performance. However, higher machine backgrounds are expected at increased luminosity, and so also increased hit occupancy. To enhance the robustness of offline software in a high-background environment, new algorithms of background suppression using the excellent SVD hit-time information have been developed, which allows a significant reduction of the fake rate, while preserving the tracking efficiency.
With the increasing luminosity also the radiation levels are expected to increase, with possible deterioration of the sensor performance. The SVD integrated dose is estimated by the correlation of the SVD occupancy with the dose rate measured by the diamonds of the radiation monitor and beam-abort system.
The effects of radiation damage are starting and in good agreement with our expectations. So far, no harmful impact due to the radiation damage on the detector performance has been observed.
The Silicon Vertex Detector of the Belle II experiment Wang, Z.; Adamczyk, K.; Aggarwal, L. ...
Nuclear instruments & methods in physics research. Section A, Accelerators, spectrometers, detectors and associated equipment,
04/2024, Volume:
1061
Journal Article
Peer reviewed
The Belle II experiment located at KEK, Japan takes data from asymmetric e+e− collision provided by the SuperKEKB accelerator. The Silicon Vertex Detector (SVD), which is part of the Belle II Vertex ...Detector (VXD), has been operating smoothly and reliably since the start of data taking in March 2019. In this article, we report on the performance of the SVD in terms of the large signal-to-noise ratio, the good hit position resolution as well as the good hit-time resolution. New algorithms based on hit-time information are under development to improve robustness of tracking performance within the anticipated high background environment. The Background situation of the SVD has been constantly monitored and no degradation in performance is observed so far. To investigate the SVD performance at high luminosity runs in the future, simulation as well as an irradiation campaign are launched and their results are summarized. During the first long shutdown of the Belle II experiment, which starts from June 2022, the VXD has been refurbished with a new two-layer DEPFET pixel detector located inside the SVD. All the delicate phases of the disassembly, re-assembly and installation of the new VXD have been successfully completed. The new VXD commissioning phase began in Sept 2023 to get ready for beam operation starting in early 2024.
Punzi-loss Abudinén, F.; Bertemes, M.; Bilokin, S. ...
The European physical journal. C, Particles and fields,
2022/2, Volume:
82, Issue:
2
Journal Article
Peer reviewed
Open access
We present the novel implementation of a non-differentiable metric approximation and a corresponding loss-scheduling aimed at the search for new particles of unknown mass in high energy physics ...experiments. We call the loss-scheduling, based on the minimisation of a figure-of-merit related function typical of particle physics, a Punzi-loss function, and the neural network that utilises this loss function a Punzi-net. We show that the Punzi-net outperforms standard multivariate analysis techniques and generalises well to mass hypotheses for which it was not trained. This is achieved by training a single classifier that provides a coherent and optimal classification of all signal hypotheses over the whole search space. Our result constitutes a complementary approach to fully differentiable analyses in particle physics. We implemented this work using PyTorch and provide users full access to a public repository containing all the codes and a training example.
The silicon vertex detector of the Belle II experiment Irmler, C.; Adamczyk, K.; Aggarwal, L. ...
Nuclear instruments & methods in physics research. Section A, Accelerators, spectrometers, detectors and associated equipment,
01/2023, Volume:
1045
Journal Article
Peer reviewed
Open access
The Belle II experiment is taking data at the asymmetric SuperKEKB collider (KEK, Japan), which operates at the Υ(4S) resonance. The vertex detector is composed of an inner two-layer pixel detector ...(PXD) and the silicon vertex detector (SVD), made of four layers of double-sided silicon strip detectors. A deep knowledge of the system has been gained since the start of operations in 2019 by assessing the high-quality and stable reconstruction performance of the detector. The very high hit efficiency and large signal-to-noise ratio are monitored via online data-quality plots. The good cluster-position resolution is estimated using the unbiased residual with respect to the track, and it is in reasonable agreement with the expectations. The SVD dose is estimated by the correlation of the SVD occupancy with the dose measured by the diamond sensors of the radiation-monitoring and beam-abort system. First radiation damage effects are measured on the sensor current and strip noise are shown not to affect the performance. Six samples of the shaped particle signal are recorded utilizing the multi-peak mode of the APV25 front-end chip and used to determine the hit timing with a precision of 2 to 3 ns. Recently a method to compute the time of collision from SVD hit time information has been implemented and verified with simulations and on data.
Punzi-loss Abudinén, F; Bertemes, M; Bilokin, S ...
European physical journal. C, Particles and fields,
02/2022, Volume:
82, Issue:
2
Journal Article
Peer reviewed
Open access
We present the novel implementation of a non-differentiable metric approximation and a corresponding loss-scheduling aimed at the search for new particles of unknown mass in high energy physics ...experiments. We call the loss-scheduling, based on the minimisation of a figure-of-merit related function typical of particle physics, a Punzi-loss function, and the neural network that utilises this loss function a Punzi-net. We show that the Punzi-net outperforms standard multivariate analysis techniques and generalises well to mass hypotheses for which it was not trained. This is achieved by training a single classifier that provides a coherent and optimal classification of all signal hypotheses over the whole search space. Our result constitutes a complementary approach to fully differentiable analyses in particle physics. We implemented this work using PyTorch and provide users full access to a public repository containing all the codes and a training example.