Online program Overview Session: N-LT-03

Abstract

Ensuring reliable RF operation in radiation-prone is critical for wireless monitoring systems in nuclear and space applications. This work presents a radiation-hardened-by-design (RHBD) voltage-controlled oscillator (VCO) implemented in a 28 nm Fully-Depleted Silicon-On-Insulator (FDSOI) process and experimentally evaluated under total ionizing dose (TID) conditions. The proposed RHBD VCO is compared with a conventional design up to 370 kGy(Si) using a Cobalt-60 gamma irradiation source at KAERI. Measurement results demonstrate that the RHBD VCO significantly improves frequency stability, achieving up to 57% reduction in frequency drift compared to the conventional counterpart. In addition, enhanced output power stability and reduced supply current variation are observed under irradiation. The results highlight the combined benefits of FDSOI technology and circuit-level hardening techniques, demonstrating a robust solution for radiation-tolerant RF circuit design.

Abstract

The next generation of imaging atmospheric
Cherenkov telescope (IACT) cameras is transitioning to silicon
photomultipliers (SiPMs) due to their stability, robustness, and
superior sensitivity compared to PMTs. Specifically, we are
developing an ASIC for the advanced SiPM camera (AdvCam).
This ASIC, designated as PRESSEC (Preamplifier Readout Electronics
for Summing SiPMs Enhanced Circuit), is implemented
in 65 nm CMOS technology. The PRESSEC utilizes a 32-
input scheme for large-area sensor readout. Its primary function
is to generate analog outputs by summing signals from four
sensors into a single channel, effectively reducing the total output
count to eight channels. The main functionality of this ASIC is
housed in the energy path, which integrates a single-ended-to-differential
converter and a pole-zero cancellation circuit. This
structure is instrumental in reducing the pulse’s Full Width at
Half Maximum (FWHM) to 3 to 5 ns, thereby mitigating pileup
effects in high activity environments. Simultaneously, these
signals feed a trigger system for rapid event identification via a
fast comparator. Additionally, a dedicated Night-Sky Background
(NSB) monitoring circuit is included to measure ambient noise
from sources like moonlight, employing a slow integrator with
adjustable high-value pseudo-resistors to achieve integration time
constants in the order of the milliseconds. Rigorous performance
benchmarks, validated using Hamamatsu and FBK SiPMs, show
a signal-to-noise ratio of at least 5, a dynamic range of 300
photons, and a linearity error below 3% for both energy and NSB
paths, while presenting a power consumption of approximately
10 mW per input channel.

Abstract

The experimental characterization of MAGIC (Multichannel Analog Gamma- ray detection Integrated Circuit) is here presented. This ASIC is designed in CMOS 110 nm technology and it is intended to readout the current signals coming from SiPMs coupled with monolithic scintillators that are used in medical imaging applications. The circuit is divided into three main sections: an energy path that provides a differential voltage 
proportional to the energy of the gamma photon, a timing path that gives information about the time of arrival of the photon, and a validation path for event discrimination. For the experimental characterization two PCBs, a mainboard and a carrier, have been designed. In this testing phase all the functionalities have been validated and the key performance of the ASIC have been characterized, in particular the power dissipation (5 mW/channel), the linearity of the energy chain to demonstrate the direct proportionality between the measured charge and the one generated by photon interaction, and the time jitter which affects the propagation delay through the timing path. The results are within the specifications and confirmed by simulations: the linearity error is below 1% and the time jitter is below 20 ps. Such a low value enables its use also in TOF-PET applications.

Abstract

Achieving 100ps FWHM coincidence time resolution (CTR) in time-of-flight positron emission tomography (TOF-PET) is benefited by electronics that can resolve individual scintillation photons with high temporal precision. We report a proof-of-concept readout in which silicon photomultiplier (SiPM) single-photon pulses are discriminated by a high-speed comparator and digitized directly by GTY transceivers of a Xilinx Virtex  UltraScale XCVU095 FPGA (VCU1287 board) at 16Gbps, encoding photon arrival times as bit positions with 62.5ps resolution. Signal integrity analysis showed that an initial LVDS comparator produced common-mode voltage and differential swing below GTY receiver specifications; switching to a CML comparator substantially reduced bit errors. Room-temperature SiPM dark count bitstreams were captured with a random FPGA clock trigger and post-processed to produce a finger plot showing 1-, 2-, and 3-photoelectron peaks, providing a preliminary validation of the bitstream-based photon-counting concept. These results support the feasibility of FPGA gigabit transceivers as scalable digitizers for single-photon pulse streams in next-generation TOF-PET detector modules.

Abstract

Abstract—Traditional radioactive source testing involves health risks, regulatory limitations, and operational inconvenience. To address these issues, this paper presents a nuclear-like signal generator capable of reproducing the waveform, amplitude, and timing characteristics of real nuclear radiation signals. The proposed system is implemented on a Xilinx K7 FPGA and supports parameter configuration through a touch screen, USB, Ethernet, or host computer software. In the waveform-generation module, photon emission timing is modeled according to scintillator light yield and decay time, while photoelectric conversion jitter and Gaussian pulse responses are introduced to synthesize realistic macroscopic waveforms. In the amplitude-generation module, a cumulative distribution function threshold table derived from measured energy spectra is used to reproduce statistical energy distributions. In the timing module, event occurrence is modeled as a Poisson process with exponentially distributed intervals. The system supports multiple radioactive sources and scintillator materials, including Na, Cs, LYSO, NaI, CsI, and BGO, with independent dual-channel output capability.

Abstract

Preamplifiers play a primary role in the resolution, linearity, and sensitivity of ionizing radiation spectrometers, converting the small charges collected by photo detectors into usable voltage signals that are further amplified, digitized, and processed to obtain the intended spectra. Additionally, the accuracy and thermal stability of such charge-to-voltage conversion must be assured to keep a constant performance over time and across different detectors, wiring configurations, and radiation scenarios. This article presents electronic simulations and the first laboratory results showing recent improvements in a new preamplifier designed for High-Purity Germanium spectroscopy systems, in terms of dynamic response, noise, linearity, thermal stability, symmetry, tolerance to reactive loads, piling-up, saturation recovery, and power consumption. The impact of these improvements is also discussed.

Abstract

The latest version of the low-power 64-channel Front-End readout chip for SiPM Array (FESA4.0) with the preliminary test results is presented in this paper. Using a row-column summing readout method, FESA compresses the 64 input signals into 16 output digital pulses, which then is digitized by the followed field programmable gate array (FPGA). In the pathway of each summing output, there are two threshold current comparators used for signal pickup, one with a low threshold for signal timing and the other with a high threshold for signal validation. Since the signal amplitude is converted into the pulse time-width via the linear discharge method, both the signals timing and amplitude can be measured by the time-to-digital converter (TDC) in the FPGA. The new version of FESA incorporates a time-over-threshold (TOT) measurement capability, which allows to identify the signals originating from different scintillators. The chip offers prominent advantages of simple architecture, low power consumption, and high performance. Through electronics tests, FESA4.0 has a measurement linearity of 99. 1% with a dynamic range of 140. The intrinsic time measurement precision is measured better than 30 ps.  Combined with a TOF-PET detector with an 8×8 LYSO crystal coupled to an 8x8 SiPM, the energy resolution and time coincidence resolution are measured as 279 ps and 12.4% respectively. These test results show that FESA4.0 is very promising for the construction of highly integrated TOF-PET scanners.

Abstract

We report a channel-level characterization of a calibrated 64-channel FPGA Time-to-Digital Converter (TDC) for fast-timing TOF-PET, with explicit decomposition of the single-shot precision (SSP) into quantization and jitter contributions. The TDC is implemented on an Altera Cyclone 10 GX device and operates with a 256-tap, 64-bin tapped-delay-line (TDL) at ~400 MHz, calibrated offline via code-density histograms. Two implementation strategies are compared: the default arithmetic carry-chain inference, and a placement-aware variant in which the TDLs are constrained to regular structures. Across all 64 channels, controlled placement reduces the ultra-bin width from 160 ± 76 ps to 101 ± 12 ps, the equivalent LSB (LSBeq) from 83 ± 32 ps to 63 ± 4 ps, and the quantization noise σ_Q from 24.5 ± 7.3 ps rms to 18.1 ± 0.8 ps rms. The placement-constrained implementation also substantially improves channel-to-channel uniformity. The constrained routing increases σ_jitter which makes jitter the dominant SSP contribution once σ_Q is reduced. A system-level projection onto previously characterized UTOFPET monolithic modules indicates that the TDC accounts for 59% of the measured CRT² in a 12 mm LYSO/256-channel module (CRT ≈ 150 ps FWHM), but only ~8% in 16–20 mm BGO/64-channel modules (CRT ≈ 400 ps FWHM). This identifies σ_jitter as the dominant architectural ceiling and motivates wave-union/multi-TDC averaging or finer-node FPGAs for full LYSO exploitation.

Abstract

Atom Switch-based FPGAs (AS-FPGAs) are being developed as next-generation FPGAs with expected high radiation tolerance. Their application to various radiation environments is anticipated, and we are also investigating their potential use in detector electronics for future high-energy physics experiments. In this study, neutron irradiation tests were performed to evaluate the Single Event Upset (SEU) tolerance of digital resources implemented in a 65 nm AS-FPGA. A prototype slow-control board equipped with the AS-FPGA was developed and used for the irradiation test. Two independent block RAM (BRAM) blocks and D-type flip-flop (DFF) blocks were implemented in the FPGA for SEU evaluation, and data mismatches detected during irradiation were counted as SEU events. Neutron irradiation tests were carried out using the tandem accelerator at Kobe University. The neutron fluence reached 1.2 × 10¹¹ n/cm². No SEUs were observed in the DFF blocks, while approximately 10 SEUs were detected in the BRAM blocks. The preliminary SEU cross section for the BRAMs was estimated to be on the order of 10⁻⁹ cm²/Mbit. In parallel with the irradiation studies, stable acquisition of temperature and pressure data from a commercial environmental sensor connected to the prototype slow-control board was successfully demonstrated during neutron irradiation, indicating the applicability of the AS-FPGA to detector slow-control systems in radiation environments.

Abstract

Correlating molecular neuroimaging with behavior in preclinical models requires acquisitions in awake and freely moving animals, whereas most current positron emission tomography procedures are still performed under anesthesia.
The MAPSSIC project addresses this limitation with an implantable intracerebral probe based on monolithic active pixel sensors (MAPS) for direct positron detection in targeted brain structures of awake, freely moving rats, thereby avoiding the pharmacological bias of anesthesia. Following sensor characterization and simulation studies previously reported, this contribution presents the transition to a fully wireless system. The acquisition chain combines four IMIC2 sensor chips, a STM32L4 microcontroller, 1 Gbit on-board flash memory, and a bidirectional 2.4 GHz radio link to a Raspberry Pi ground station. The complete software stack is implemented in Rust with the Embassy asynchronous framework, providing memory safety and concurrent peripheral management without dynamic allocation. Planned validation with ¹⁸F, ¹¹C phantoms, and anesthetized implanted rats will mark the shift from sensor feasibility to system-level wireless validation.

Abstract

The COMET (COherent Muon to Electron Transition) Phase-I experiment at J-PARC searches for coherent muon-to-electron conversion in aluminum with a single-event sensitivity of 3 x 10^-15. A key detector challenge is the high instantaneous beam rate, which produces background hits in the Cylindrical Drift Chamber (CDC). In particular, low-energy protons can generate large ionization signals that drive the readout electronics into nonlinear regions, producing waveform saturation and pileups. These effects may reduce hit efficiency by masking later signal or background hits and therefore must be modeled realistically in performance studies. We developed a detector response model for the CDC that combines Geant4-based ionization with the Photon Absorption Ionization (PAI) model, Garfield++-based drift time calculation, and an electronics response emulation based on measured response functions. The model explicitly includes nonlinear saturation effects in the circuit and was validated using cosmic-ray test data. The measured-response model reproduces the main features of the data more accurately than a simple linear-response model. The remaining discrepancy corresponds to a limited uncertainty in the pileup-effect evaluation, providing a realistic basis for COMET Phase-I reconstruction and performance studies.

Abstract

This work reports on the experimental characterization, modeling and correction of high intensity effects on the pixel response of the DEPFET Sensor with Signal Compression (DSSC) camera, an active pixel detector targeted at soft X-ray imaging experiments at the European XFEL. Experimental tests with up to 5 MeV/pixel deposited energy have been performed at the European XFEL, showing evidence of a shift of the pedestal value growing with X-ray intensity. Moreover, the signal compression provided by the DEPFET (DEpleted P-type Field Effect Transistor), integrated in each pixel, enhances the impact of this systematic effect on the measurement accuracy. The observed shift is assumed to be proportional to the signal dependent drop of the power supply lines along each ASIC column and can be modeled by two parameters. We devised a procedure to efficiently measure the two model parameters by means of a custom mask that shields a fraction of each ASIC column, irradiated with a flat field. The main results of the characterization of the signal dependent pedestal shift parameters and the application of the resulting correction to X-ray images of different patterned masks (dots, stripes, 2D shapes etc.) acquired at increasing intensities will be presented and discussed.

Abstract

The increasing brilliance of fourth-generation synchrotron sources drives a growing demand for energy-dispersive X-ray fluorescence (XRF) spectrometers capable of handling extreme photon fluxes.
We present the design and characterization of the ARDESIA-64 instrument, a complete X-ray fluorescence (XRF) spectrometer for high-throughput synchrotron applications. 
The instrument integrates a highly scaled 64-channel monolithic Silicon Drift Detector (SDD) array, featuring a pixel pitch of 2.5 mm, within a compact, vacuum-compatible mechanical housing tailored for beamline integration.
At the conference, we will present the instrument design and comprehensive spectroscopic characterization results. 
The collimated 450 μm-thick sensor variant demonstrates an excellent average energy resolution of 125.9 eV FWHM at 5.9 keV. Preliminary high-rate tests confirm exceptional throughput capabilities, sustaining Input Count Rates (ICR) exceeding 800 kcps per channel while preserving an energy resolution of ~150 eV. The readout is performed by a 64-channel Digital Pulse Processor (DPP) based on the CAEN DT5560SE digitizer with firmware specifically adapted for high-throughput X-ray spectroscopy in synchrotrons.

Abstract

We have been organizing schools on radiation measurements for students from ASEAN countries. Hands-on experience with radiation measurements is very important, and we prepare radiation measurement systems for practical exercises.

To host as many students as possible, we considered organizing the schools in ASEAN countries. However, the shipping cost of the equipment was extremely high. As a result, we could not conduct hands-on exercises on radiation measurement.

Driven by recent large-scale experiments, electronics for radiation measurements have become compact, highly integrated, and low-cost. This development now makes it possible to carry equipment to schools.

Today, compact detector systems are available at affordable prices, making it much easier to conduct practical radiation measurement exercises in schools.

However, modern compact electronics have introduced a new problem: they are too small to probe easily with an oscilloscope. In modern electronics, waveform processing is increasingly performed digitally. Nevertheless, we believe that analogue processing remains important and highly educational for students. This motivated us to develop electronics specifically for educational purposes.

We have developed an “education kit” for radiation measurement education. By plugging in a SiPM and a small scintillator crystal module, the board functions as a complete radiation measurement system. An on-board FPGA interfaces with a PC through a USB connection. The board is the size of a Raspberry Pi board. Optionally, it can be plugged onto a Raspberry Pi.

Although the board is compact, it provides sufficient space for probing signals. The design of the board will be released as open-source hardware. Students can modify the design and develop their own versions of the board. The “education kit” will serve as an entry point to radiation measurement electronics.