Firmware Development and Integration for ALICE TPC and PHOS Front-end Electronics: A Trigger Based Readout and Control System operating in a Radiation Environment

The readout electronics in PHOS and TPC - two of the major detectors of the ALICE experiment at the LHC - consist of a set of Front End Cards (FECs) that digitize, process and buffer the data from the detector sensors. The FECs are connected to a Readout Control Unit (RCU) via two sets of custom made PCB backplanes. For PHOS, 28 FECs are connected to one RCU, while for TPC the number is varying from 18 to 25 FECs depending on location. The RCU is in charge of the data readout, including reception and distribution of triggers and in moving the data from the FECs to the Data Acquisition System. In addition it does low level control tasks. The RCU consists of an RCU Motherboard that hosts a Detector Control System (DCS) board and a Source Interface Unit. The DCS board is an embedded computer running Linux that controls the readout electronics. All the mentioned devices are implemented in commercial grade SRAM based Field Programmable Gate Arrays (FPGAs). Even if these devices are not very radiation tolerant, they are chosen because of their cost and flexibility, and most importantly the possibility to easily do future upgrades of the electronics. Since physical shielding of the electronics is not possible in ALICE due to the architecture of the detector, the radiation related errors need to be handled with other techniques such as firmware mitigation techniques. The main objective of this thesis has been to make firmware modules for the FPGAs reciding in different parts of the readout electronics. Because of the flexibility of the designs, some of them have, with minor adaptations, been applied in different devices surrounding the readout electronics. Additionally, effort has been put into testing and integration of the system. In detail, the work presented in this thesis can be summarized as follows: - Firmware design for radiation environments. All firmware modules that are designed are to be used in a radiation environment, and then special precautions need to be taken. Additionally, a state-of-the-art solution has been designed for protecting the main FPGA on the RCU Motherboard against radiation induced functional failures. - Implementation of Trigger Handling for the TPC/PHOS Readout Electronics. The triggers are received from the global trigger system via an optical link and are handled by an Application Spesific Integrated Circuit (ASIC) on the DCS board. The problem is that the DCS board might have occasional down time 6 due to radiation related errors, so a special interface module is designed for the main FPGA on the RCU Motherboard. This module decodes and verifies the information received from the trigger system. As it is a generic design it has also been implemented as part of the BusyBox. The BusyBox is an important device in the trigger path of the TPC and PHOS sub-detectors. - Testing and Verification of all firmware modules. All firmware modules have been extensively verified with computer simulation before being tested in real hardware. - Maintenance of the DCS board for TPC/PHOS and of the different Fee firmware modules in general. - System Integration and System Level Tests. A big contribution has been done integrating and testing all the modules and sub-systems. This concern both locally on the RCU and the BusyBox, as well as making all the devices play together on a larger scale. - Testing and Verification of all firmware modules. All firmware modules have been extensively verified with computer simulation before being tested in real hardware. - Maintenance of the DCS board for TPC/PHOS and of the different Fee firmware modules in general. - System Integration and System Level Tests. A big contribution has been done integrating and testing all the modules and sub-systems. This concern both locally on the RCU and the BusyBox, as well as making all the devices play together on a larger scale. As the presented electronics are located in a radiation environment and are physically unavailable after commissioning, effort has been put into making designs that are reliable, scalable and possible to upgrade. This has been ensured by following a systematic design approach where testability, version management and documentation are key elements. Some parts of the work described in this thesis have been published and presented in international peer reviewed publications and conferences.