Improvement of reliability in long-term operation of the LHD cryogenic system

The Large Helical Device at National Institute for Fusion Science in Japan is a heliotron-type fusion plasma experimental device. In the device, superconducting magnets, the supporting structures and superconducting bus lines are cooled by a large-scale helium liquefier/refrigerator with total equivalent cooling capacity of 9.2 kW at 4.4 K. The cryogenic system has been operated stably since 1997 thanks to appropriate annual maintenance and 25 plasma experimental campaigns have been conducted successfully. The total operating time is 110,521 hours. An investigation into causes of failures in the system up to 2010 showed that most of the downtime was caused by the control system and helium compressors. For higher reliability, the previous control system using duplicated VME controllers and reflective memory was replaced to new control system using duplicated compact PCI controllers and duplicated remote I/O in 2013, because problems with shared reflective memory often caused the entire system to stop previously. Two redundant compressors were also added in 2011. It is expected to reduce recovery time by allowing alternative operation against the event that any existing compressor breaks down. In the present paper, the long-term reliable operation of the cryogenic system due to appropriate improvement is reported. After the improvements, the control system has never stopped because the probability of failure decreased. Although the failures of the compressors occurred twice, average downtime was reduced to 0.65 h. Consequently, the mean time between failures drastically increased to 19,751 h, while the availability became almost 100 %.