Data Analysis of Transient Energy Releases in the LHC Superconducting Dipole Magnets

Premature training quenches are caused by transient energy released within the LHC dipole magnet coils while it is energized. Voltage signals recorded across the magnet coils and on the so-called quench antenna carry information about these disturbances. The transitory events correlated to transient energy released are extracted making use of continuous wavelet transform. Several analyses are performed to understand their relevance to the so called training phenomenon. The statistical distribution of the signals amplitude, the number of events occurring at a given current level, the average frequency content of the events are the main parameters on which the analysis have been focalized. Comparisons among different regions of the magnet, among different quenches in the same magnet and among magnets made by different builders are reported. Conclusions about the efficiency of the raw data treatment and the relevance of the parameters developed with respect to the magnet global behavior are finally given.