Speaker
Description
The outer radii of the inner tracker (ITk) for the Phase-II Upgrade of the ATLAS experiment will consist of groups of silicon strip sensors mounted on common support structures. Lack of space creates a need to remotely disable a failing sensor from the common HV bus. We have developed circuitry consisting of a GaNFET transistor and a HV Multiplier circuit to disable a failed sensor. We will present two variants of the HV Mux circuitry and show irradiation results on individual components with an emphasis on the GaNFET results. We will also discuss the reliability of the HV Mux circuitry and show plans to ensure reliability during production.
Summary
To increase the physics reach of the Large Hadron Collider (LHC) a major upgrade of the collider, known as the High Luminosity LHC (HL-LHC) is planned to be operational around 2026. The ATLAS detector will need to undergo major upgrades to cope with the increased luminosity. One of the major upgrades is the Inner Tracker (ITk), which will be an all-silicon detector consisting of pixel detectors at the inner radii and silicon strip sensors in the outer radii.
The silicon strip detector will be fabricated from staves in the barrel region and petals in the endcap regions. They consist of multiple silicon sensors mounted on a common thermal-mechanical support and cooling structure. Due to a drive to reduce material and a lack of space, it will not be possible to bring individual high voltage bias to each sensor. Groups of sensors will need to share a common high voltage bias. Should a sensor fail due to its developing a low breakdown voltage or a short, other working sensors will also be lost.
We began a program to develop a method to disconnect a malfunctioning sensor from the HV bias bus. The main requirements on the high voltage multiplexing circuitry (HV Mux) is that it be capable of switching -500 V, operate in a 2 Tesla magnetic field, and survive fluences of 2 x 1015 hadrons/cm2 and ionizing doses ~ 50 Mrad.
We initially focused on two emerging technologies, Gallium Nitride and Silicon Carbide, for an HV switch that could potentially be radiation hard. Additionally, we considered a unique design using 3D Trench technology for a custom JFET. We have concluded that only Gallium Nitride technology is capable of meeting our radiation requirements.
We will report on an extensive irradiation campaign on a variety of Gallium Nitride Field-Effect Transistors (GaNFETs) that have survived doses as high as 1 x 1016 neutrons/cm2 and over 200 Mrad. Results from irradiations with neutrons, protons, pions, and gamma rays will be presented.
We have developed an HV Multiplier circuit that controls the state of the GaNFET. The multiplier circuit accepts a ~50 kHz square wave from a radiation hard ASIC operating at 2.5V and generates a gate-source voltage of several volts but offset by the bias voltage to the sensor (as much as -500 V). The components of the HV Multiplier circuit must meet the radiation hardness requirements of the GaNFET. We will report on their radiation hardness and demonstrate that we believe we have developed an HV Multiplier circuit that is sufficiently radiation hard. We have developed a dual-stage variation of the HV Mux circuit than can operate at -700 V or even lower.
The introduction of such circuitry should not introduce more failures or loss of sensor operation due to HV Mux circuitry failure than if the HV Mux circuitry was not implemented and we instead rely on low rates of sensor failures. We will present a risk analysis demonstrating the tradeoffs of implementing HV Mux vs not implementing. We will present our plans to ensure the reliability of the HV Mux circuitry in our continuing R&D phase as well as during production.
