Characterization of the ATLAS Liquid Argon Front-End ASIC ALFE2 for the HL-LHC upgrade
summary
The gist
ALFE2 is an ATLAS Liquid Argon Calorimeter (LAr) Front-End ASIC designed for the HL-LHC upgrade The test results indicate that ALFE2 fulfills or greatly exceeds all specifications on gain…
In short
ALFE2 is an ATLAS Liquid Argon calorimeter ASIC designed for the HL-LHC upgrade. Testing confirmed it meets or exceeds all specifications for gain, noise, linearity, and radiation tolerance. The chip achieves a 16-bit dynamic range using dual gain outputs and demonstrates high resilience to radiation and single-event effects.
Key concepts
- Dual Gain Outputs (LG/HG)
- ALFE2 uses two separate output channels with different gains: Low Gain (LG) and High Gain (HG). This allows the system to cover a wide dynamic range of 16 bits, providing high resolution for measuring signals across different energy levels.
- Fully Differential Architecture
- The ASIC is built using a fully differential circuit design. This technique doubles the voltage swing of the signal, which significantly reduces sensitivity to common-mode noise—noise that affects both signal paths equally—improving overall measurement precision.
- Radiation Hardness (TID/SEE)
- Testing showed ALFE2 is highly resistant to radiation. After being exposed to a high dose (1.4 kGy), performance changes in gain and the trigger sum baseline were less than 1%. Furthermore, it successfully withstood proton beam testing without detecting any Single-Event-Upsets (SEUs).
- Integral Non-Linearity (INL)
- INL measures how much the output signal deviates from a perfectly straight line across its full operating range. The results show excellent linearity: LG outputs have an INL below ±0.62% across the full dynamic range, and HG outputs are even more precise at less than ±0.13%.
Terminology used across episodes
This episode discusses
- Characterization of the ATLAS Liquid Argon Front-End ASIC ALFE2 for the HL-LHC upgrade · Paper Radio
The paper
Characterization of the ATLAS Liquid Argon Front-End ASIC ALFE2 for the HL-LHC upgrade · Read on arXiv
D. Matakias, G. Carini, H. Chen, M. Dabrowski, G. Deptuch, L. Duflot, J. Kierstead, T. Liu, H. Ma, Á N Morange S Rescia S Tang H Xu
Physics Department Brookhaven National Laboratory Instrumentation Division KU Leuven IJCLab Université Paris-Saclay CNRS IN2P3
DOI: 10.1088/1748-0221/19/04/C04019
Transcript
Introduction to the show: ident: Physics. The day in general physics: instruments, methods and results across the field.
Kai: We're looking at the paper: "Characterization of the ATLAS Liquid Argon Front-End ASIC ALFE2 for the HL-LHC upgrade".
Mira: Here's the gist of the paper. ALFE2 is an ATLAS Liquid Argon Calorimeter (LAr) Front-End ASIC designed for the HL-LHC upgrade The test results indicate that ALFE2 fulfills or greatly exceeds all specifications on gain, noise, linearity, uniformity, and radiation tolerance Design of ALFE2 ALFE2 integrates quad-channel transimpedance PAs and CR-(RC)2 shapers A key feature of ALFE2 is to cover a dynamic range of 16 bits To achieve such a high dynamic range, each channel of ALFE2 has two outputs with different gains, Low Gain (LG) and High Gain (HG) Both outputs can be simultaneously read out Each HG or LG output has a CR-RC shaping stage and an RC shaping stage To be compatible with the current trigger system
3: , ALFE2 includes a Trigger-Sum (TS) output based on the low gain outputs of the PAs ALFE2 is implemented as a fully differential circuit architecture The differential architecture doubles the voltage swing, thus making the ASIC much less sensitive to commonmode noise The PA, which performs the single-ended to differential conversion plays a crucial role in achieving the required noise and linearity Each PA consists of an input stage (A0) and an output stage (A1) The controlled terminating input resistance is realized by means of an asymmetric negative feedback
5: as shown in Figure 1(a) To achieve low noise while minimizing power, the input stage utilizes an inverter-based differential circuit
6: as shown in Figure 1(b) ALFE2 operates with two power domains, 1.2 V for the input pair of the PAs and the shapers and 2.5 V for the output stage of the PAs Test Setup of ALFE2 The block diagram of the test setup is shown in Figure 2(a) The ALFE2 chip under test is soldered or held in a socket mounted on a test board The FETB uses dual octal-channel 16-bit ADCs (Texas Instruments ADS52J65) with an Effective Number of Bits (ENOB) of 13 and a sampling rate of up to 125 MS/s The test setup is used in characterization measurements, irradiation tests, and Quality Control (QC) evaluations Test Results of ALFE2 The transient responses of the LG outputs at 25 Ω and various input currents are shown in Figure 3(a) The relative gain ratio between the HG and the LG outputs was measured at 22.9 on average The Integral Non-Linearity (INL) measurement of ALFE2 was performed The INL of the LG outputs is below ±0.62% across the full dynamic range and ±0.21% within 80% of the dynamic range The Equivalent Noise Current (ENI) values shown in Figure 4(b) are slightly lower than those shown in Figure 4(a) due to the inclusion of ADC anti-alias filters The power dissipation is 610 mW on average or 153 mW per channel Radiation Hardness and Yield The ALFE2 underwent radiation hardness testing (TID) at Brookhaven National Laboratory, utilizing a 60Co gamma-ray source at a dose rate of 0.28 Gy/s Less than 1% performance change was measured on the gain and the TS baseline after the dose was beyond the specification of 1.4 kGy, whereas no change was observed on other metrics A SEE test, using a 226 MeV proton beam was performed at Massachusetts General Hospital in Boston Four chips were irradiated to a total fluence of 1.7´1014 cm-2, 17 times of the specification, with the auto-refresh feature enabled, and no Single-Event-Upsets (SEUs) were detected in any chip The error rate for the entire ATLAS LAr calorimeter is extrapolated to be under 5 errors per day (95% confidence level) The ALFE2 ASIC, designed for the ATLAS LAr calorimeter upgrade, has been extensively characterized The measured ENIs are 48 nA and 153 nA at 50 Ω and 25 Ω, respectively So far, 120 pre-production chips have been tested and all meet the specifications The test results indicate that ALFE2 fulfills or greatly exceeds all specifications on gain, noise, linearity, uniformity, and radiation tolerance Based on the characterization results, about 80,000 ALFE2 chips will be produced for the HL-LHC upgrade Acknowledgments We wish to extend our sincere appreciation to the individuals who contributed to the success of the ALFE2 chip design project First, we would like to acknowledge G.
Kai: de Geronimo who designed HLC1, a predecessor to the ALFE We would like to express our gratitude to the IN2P3/OMEGA team, comprising Sylvie Blin, Selma Conforti, Christophe de la Taille, Mowafak El Berni, and Nathalie Seguin-Moreau In addition, we acknowledge the contributions made by Vamshi Manthena and Sandeep Miryala from the Instrumentation Division at Brookhaven National Laboratory and Stefano Michelis from CERN Finally, we would like to thank Ethan Cascio from the Proton Therapy Center at Massachusetts General Hospital in Boston References
1: ATLAS Collaboration, Technical Design Report for the Phase-II Upgrade of the ATLAS LAr Calorimeter, CERN-LHCC-2017-018, ATLAS-TDR-027, CERN, Geneva, 15 June 2018
2: K.A.
Mira: Khoury et al., Characterization of preamplifier and shaper prototypes for the Phase-II upgrade of the LAr Calorimeter readout, ATLAS Note ATL-LARG-INT-2019-001, 27 September 2019
3: G.
Kai: Aad et al., The Phase-I trigger readout electronics upgrade of the ATLAS Liquid Argon calorimeters, 2022 JINST 17 P05024
4: N.
Title and authors: Kai: Let's get into the specifics of who wrote this paper, "Characterization of the ATLAS Liquid Argon Front-End ASIC ALFE2 for the HL-LHC upgrade." The authors are D Matakias, G Carini, H Chen, M Dabrowski, G Deptuch, L Duflot, J Kierstead and T Liu.
Mira: These folks are from a mix of places—Brookhaven National Laboratory and KU Leuven in Belgium—which tells you they’re bringing together different expertise in the front end electronics design.
Lev: What this team is doing is taking a complex piece of hardware, an ASIC, and rigorously testing it against the very demanding requirements set by the physics experiment itself.
Kai: The implication here is that they are not just designing a theoretical circuit; they are proving that this physical chip can actually operate reliably in the extreme environment needed for high-luminosity physics.
Mira: It speaks to how much effort goes into making sure the readout electronics can keep up with the massive data rates expected from the HL-LHC upgrade.
Lev: The fact that they've characterized it on a board using a Zynq UltraScale plus mp soc shows they’re testing it in a way that mimics how it will eventually be deployed in the actual detector system.
Kai: It’s about moving from simulation to reality, making sure the hardware matches the theoretical performance targets before committing to building eighty thousand of these chips <ref:2402.08901#pg2>.
Mira: So, this isn't just an academic exercise; it’s a critical step in verifying that the entire readout chain for liquid argon calorimeters is ready for operation at that high luminosity.
Lev: The authors are showing their work in terms of performance exceeding specifications by a large margin, which sets a high bar for what the rest of the production run needs to achieve.
Kai: It’s about establishing that baseline; if this chip performs better than expected in the lab, it gives engineers confidence that it will perform well under real detector conditions.
Mira: And they're also providing concrete numbers on noise and linearity, which is what experimentalists need to know to plan their data acquisition strategy.
Lev: It’s the validation of the design against the physical constraints of building something that needs to handle high energy deposition signals from those collisions.
The paper's summary: Kai: So, let's get into what the paper actually summarized about ALFE2. It boils down to it being a fully differential circuit with two gain outputs for sixteen-bit dynamic range coverage <ref:2402.08901#pg1,16-bit dynamic range coverage>.
Mira: They detailed the design choices, like using an asymmetric negative feedback for termination and an inverter-based differential circuit in the input stage to keep power low and noise down.
Lev: And they specifically mentioned that this dual power domain design, operating at one point two volt for the input stages and two point five volt for the output stage, was chosen to enhance both transimpedance gain and minimize power consumption.
Kai: It also covered how they make things flexible by making the TS output gain relative to the low gain outputs adjustable between one, two, or three times their level.
Mira: Plus, they showed that the Integral Non-Linearity for the low gain outputs stays below zero point six two percent across their whole dynamic range and under zero point two one percent within eighty percent of it.
Lev: The noise current values they report are forty eight nano amperes at fifty ohm and fifteen three nano amperes at twenty five ohm, which gives a concrete picture of the noise floor in different impedance scenarios.
Kai: They also covered the timing constants for the shapers being configurable, allowing peaking times to be set from twenty nine nanoseconds up to forty six nanoseconds at fifty ohms, and then different ranges at twenty five ohms.
Mira: So, they’ve mapped out all these tunable parameters so that the system can be configured with an I2 C interface that has eight bit internal registers for configuration.
Lev: That configuration interface includes triple modular redundancy to help with resilience against single event effects and a built in auto refresh feature for error bits.
Kai: The summary is really about showing this isn't just a fixed design, but a flexible system where you can tune the gain, timing constants, and baseline voltages via that interface.
Mira: And they’re making sure that all these tuning knobs are well within their specified limits for linearity and noise performance.
Lev: It confirms that the fundamental design choices they made to balance noise, linearity, power use, and dynamic range coverage are sound for the intended application.
The paper's improvements: Kai: Now let’s talk about what they suggest as improvements or key takeaways from their testing. They highlight how the system can achieve a sixteen bit dynamic range by using those separate low and high gain outputs for optimal resolution <ref:2402.08901#pg1>.
Mira: That’s important because it shows that having two paths instead of one path gives you more control over the signal quality, allowing for better resolution.
Lev: From an engineering standpoint, that dual path design is what lets them get the sixteen bit range, and it's a direct consequence of how they structured the circuit to handle those different signal levels effectively <ref:2402.08901#pg1>.
Kai: They also emphasize that their fully differential architecture for noise reduction is key because doubling the voltage swing makes it much less sensitive to commonmode noise.
Mira: That’s a crucial point because in any real setup, you have noise sources that we don't always perfectly know about, and this design helps mitigate those unknown issues.
Lev: And they also stress how the configurable shaping and timing constants mean you can precisely tailor the waveform reconstruction to what the detector is seeing at different input impedances.
Kai: It’s not just a fixed setup; it’s adaptable to different operating conditions, which is important when you are trying to get accurate measurements.
Mira: And they also point out that their radiation hardness results show less than one percent performance change on the gain and the TS baseline after a dose of one point four kilogray.
Lev: That resilience under those conditions is what makes them viable for the long-term operation in a high radiation environment like the LHC.
Kai: Plus, they confirm that they haven't seen any single event upsets in any of their chips during the proton beam testing, which is a huge win for reliability.
Conclusion: Mira: So to wrap up, this paper on "Characterization of the ATLAS Liquid Argon Front-End ASIC ALFE2 for the HL-LHC upgrade" shows that they’ve done a thorough job characterizing this chip under lab conditions and radiation testing.
Kai: They’ve shown it meets or exceeds all their stated specifications across gain, noise, linearity, uniformity, and radiation tolerance.
Lev: The main implication is that based on these characterization results, about eighty thousand of these ALFE2 chips will be produced for the final production run for the HL-LHC upgrade.
Mira: It means this design is validated enough to move forward into mass manufacturing with high confidence in its performance metrics.
Kai: So, what we’re hearing is that they have a well-validated component ready to be integrated into the larger detector system.
Lev: From an engineering viewpoint, this chip proves that the design choices for balancing noise and power consumption are solid for this kind of high energy deposition readout.
Mira: And it confirms that these specific tuning knobs they built into the ALFE2 ASIC are working exactly as intended across all their operating conditions.
Kai: It’s a strong signal that this front end electronics is ready to be put into the next phase of building out the ATLAS detector.
More episodes
- 2610.34250-Yb/Sr optical clock comparison at 2.08E-18 uncertainty over a 1350-km fiber link
- 2610.36514-Integer electron-count cuts in variational second-order reduced-density-matrix theory. I. Molecular dissociation with fixed real-space and orbital partitions
- 2611.14082-Two atoms with strong light-induced dipolar interactions
- 2611.14040-Characterization of BNLs 20 and 30 micrometer thick LGADs for the Electron Ion Collider
- 2611.13399-Lift-element-based dynamic network of vortex gust-airfoil interactions
- 2611.13476-Hopfion Superlattices in 3D Twisted Moir'e Light Fields
- 2611.13642-Investigation of the timing performance of double-sided p-type 3D silicon sensors produced by CNM
- 2611.13830-SiPM optical readout of a PICOSEC Micromegas precise timing detector
- 2611.14152-Where are the shortest-path nodes in spatial networks?
- 2611.14219-Sub-Ice: semi-automated mapping of basal channels from ice shelf surface elevation data