Ponente
Descripción
The radiation environment (e.g. gamma, neutrons) can damage or destroy electronic devices or sensors, corrupt signals in analogue or digital circuits, corrupt data or programs in digital circuits (memories, microprocessors, microcontrollers, FPGAs, etc. ). These effects can appear gradually due to cumulative phenomena, or instantaneously due to a single particle (e.g. a neutron) causing a phenomenon called “Single Event Effect” (SEE).
In a modern tokamak (e.g. ITER) or in a modern particle accelerator (e.g. the LHC), the most difficult radiation problem to solve is that of SEEs induced by thermal and intermediate neutrons (up to 14 MeV in a deuterium-tritium fusion installation, and up to 20 MeV in a particle accelerator). To perform the regulatory safety demonstration of safety electronics, or the reliability demonstration of critical electronics (essential for investment protection and operability of the machine), it is necessary to know the reliability of the constituting electronic semiconductor devices. This reliability established by their manufacturers is only valid in the environment in which it was established, which is the natural terrestrial environment at ground level. In a non-natural neutron environment, it must be determined by neutron irradiation tests. This approach is generally not feasible due to the large amount and diversity of electronics involved. The possibilities of relocation to radiation-free regions are limited due to constraints on cable lengths. The possibilities of redundancy (to increase reliability and ensure a safe state) are limited due to size and cost constraints, and generally do not make it possible to dispense with the need for neutron irradiation tests. In a tokamak or in an accelerator, for the vast majority of electronics, the only realistically feasible solution is to install the electronics in Radiation Protected Areas (RPAs), whose neutron environment must be demonstrated to be equivalent, in terms of impact on the reliability of general modern electronics, to the natural atmospheric terrestrial environment at ground level.
This demonstration of equivalence requires a detailed study of the sensitivity of electronics to thermal and intermediate neutrons, covering a panel of electronic components and SEE mechanisms representative of general modern electronics. This panel must include the families of components most sensitive to neutrons and must include for each family different technological nodes representative of contemporary and advanced electronics.
CERN, in collaboration with the IM2NP Institute of the University of Aix-Marseille and the IRFM Institute, operator of the WEST tokamak at the CEA in Cadarache, is currently preparing, within a broad collaborative framework, a detailed study of the sensitivity to SEEs of the general modern electronics, the results of which will provide the input data necessary for the optimization of Radiation-Protected Areas in facilities such as tokamaks and particle accelerators. This approach will make it possible to replace the case-by-case qualification of the electronics of a facility for its specific neutron environment, which is not feasible on a large scale, by the optimization and the qualification of the neutron environment for any modern electronic circuit.
This invited talk will present the motivations for this detailed study, the preliminary study which made it possible to validate the method and the models which will be used for the detailed study, and the guidelines for the detailed study.
