Speakers
Description
We studied the Total Ionizing Dose (TID) response of LSF0102 2-channel voltage translators intended for the upgrade of the ATLAS Muon Barrel read-out system for HL-LHC. TID tests were carried out at the CERN CC60 facility using a 60Co gamma source. The devices showed no degradation in key performance metrics, including supply current, eye diagram quality, rise/fall time, jitter, and bit error rate, up to the ATLAS-recommended dose. Results of a second irradiation at 20 kGy are under evaluation, with additional analysis using impedance spectroscopy modelling and techniques to assess potential degradation of specific device sections.
Summary (500 words)
HL-LHC will pose challenges for the reliability of detectors and electronics, requiring upgrades of the major physics experiments. In this regard, a new read-out system of the ATLAS Muon Barrel will be installed. It is based on Data Collector Transmitter (DCT) boards [1] that process RPC hit data and send it to the Barrel Sector Logic via optical fibers. The reliability of the DCT components must be verified through dedicated irradiations at the expected radiation levels. In this work, we study the TID response of LSF0102 2 Channel voltage translators. These are suited for translating signals with fast edge rates, working up to 100 MHz. They operate by allowing their FET switches to conduct during the low pulse of an input signal and shutting them off during the high pulse. These characteristics make them useful to match the front-end logic levels (1.2 V) to those of the output FPGA (3.3 V). For the TID test, we designed a setup made of three boards [2]. An Input Board (I-B) supplies voltage from an Agilent N6705B DC Power Analyzer and signals from a Tektronix AFG31152 function generator. The signals are distributed across four channels using a 1-to-4 fan-out buffer, then transmitted to the Device Under Test (DUT) board (DUT-B). According to ATLAS recommendations, the DUT-B (Fig. 1-left) houses ten DUTs. Fig. 1-middle shows the DUT-B placed in front of the 60Co gamma source at a distance ensuring a dose rate of 32.9 Gy/h, reaching the required TID=83 Gy in 2.5 hours. Output signals are driven to the Readout Board (R-B), shown in Fig. 1-right together with the I-B, both placed in a buffer zone.
DUT selection is done by four 16-to-1 multiplexers, controlled by the GPIO pins of a Raspberry Pi. Connections between the boards are made via 4-meter flat cables. Two tests have been performed: one at the ATLAS target dose and another at 20 kGy and 360 Gy/h. For the first test, we did not observe significant variations in the current drawn the DUTs, as well as in the eye diagram, rise/fall time, jitter and bit error rate. For the high-dose test, data analysis is ongoing. We foresee applying impedance spectroscopy, a non-destructive technique recently used to study the capacitive behavior of Si pn diodes [3] and the TID response of LVDS receivers [4], to assess potential damage on specific decive sections.
[1] M. Bauce et al., ATLAS Level-0 Muon Barrel Trigger system status and integration tests for Phase-II, Nucl. Instrum. Meth. A 1069 (2024) 169843.
[2] P. Casolaro et al., TID damage assessment on LVDS links for the ATLAS muon barrel spectrometer readout system, 2025 JINST 20, C01023
[3] P. Casolaro et al., Modeling the diffusion and depletion capacitances of a silicon pn diode in forward bias with impedance spectroscopy, J. Appl. Phys. 136 (2024) 115702.
[4] P. Casolaro et al., TID analysis of LVDS receivers with impedance spectroscopy for the ATLAS muon trigger, in proc. 2024 IEEE NSS MIC RTSD, 2024
