Speaker
Description
Summary 500 words
The use of programmable logic devices such as FPGAs has become a very attractive solution for High Energy Physics applications. With the ability to be reprogrammed in the field, FPGAs offer the flexibility to continuously develop and update already deployed instrumentation. However, a major caveat of SRAM based FPGAs are their susceptibility to radiation. As the configuration of the FPGA is stored in SRAM memory, Single Event Upsets (SEUs) in this configuration memory may lead to a malfunction of the FPGA. In the main tracking detector of ALICE, the Time Projection Chamber, an SRAM based FPGA from Xilinx is implemented in the Readout Control Unit (RCU) of the front-end electronics and controls the read out of
data from the detector. This FPGA will experience SEUs induced by the radiation produced when the LHC beams are collided in ALICE experiment. A dedicated reconfiguration solution has therefore been implemented to continuously detect and correct any SEUs in the configuration memory of the the RCU FPGA. The solution is based on partial reconfiguration, an option offered by Xilinx for a number of FPGAs to reconfigure the configuration
memory without interrupting the operation of the FPGA. Due to the sophisticated implementation of this solution, every SEU is registered and logged by the detector control system. This allows us to monitor and correlate the number of SEUs to the number of collisions or luminosity seen by the ALICE experiment. In fact, as the SEU cross section of the FPGA configuration memory is known from previous irradiation tests, the network of 216 FPGAs installed on the TPC detector can be used as an online radiation monitor.
This paper will present the first measurements of radiation induced SEUs in the RCU FPGAs. The results will be compared to simulations and discussed in light of the expected integrated luminosity. These results are of high importance as they provide the first measurement-based predictions of the SEU rate that can be expected for the RCU FPGA in the coming years of operation.
