Report on the Fire in the Second Experimental Hall
of the Materials and Life Science Experimental Facility of J-PARC


19 February 2015

Incorporated Administrative Agency - Japan Atomic Energy Agency
Inter-University Research Institute Corporation - High Energy Accelerator Research Organization


Subject : Fire at the Second Experimental Hall of the Materials and Life Science Experimental Facility, J-PARC


  1. Time and Date of Incident : Approximately 15:01 on 16 January 2015

  2. Incident Location : Muon D-line area in the Second Experimental Hall of the Materials and Life Science Experimental Facility, J-PARC

  3. Category of Incident : Fire in the facility

  4. Description of incident : 
    The remodeling work of the power supply for septum magnets of a muon beam line had been carried out since 9:00 of January 16 2015.

    ⅰ.   Around 14:30 : Electrical testing of the power supply with a newly installed transformer started. Due to unusual smell from the supply, doors of the power supply cabinet were opened around 14:57 to find the cause of the smell. Around 15:01 the transformer was found to catch fire. The fire was immediately extinguished with a hand extinguisher.

    ⅱ.   15:01 : Called 119 (the fire department).

    ⅲ.   15:11 : Fire engines arrived at the MLF

    ⅳ.   15:26 : The fire department judged this incident was "fire". At the same time they confirmed that it was already extinguished. The burned area is ~30 cm x 20 cm (600 cm2) and the transformer was damaged.

    ⅴ.   It was confirmed that there was no leakage of radioactive material to both inside and outside of the facility.

    Immediately after the incident, the accelerators stopped the beam operation and no muon beam was used.

  5. Effect of the incident
    (1) No leakage of ratioactive material to the environment
    (2) No radiation exposures
    (3) No injuries
    (4) The newly installed transformer burned and was damaged

  6. Effect to the facility
    The fire was immediately extinguished with a fire extinguisher. There was no spread of the fire at all.

  7. Direct Causes of the Fire
    7-1 Following the direction of the fire department, the transformer was taken off from the power supply for inspection. The fire was presumed to be originated from the transformer itself. (Fig. 4 shows the picture of the burnt transformer.)

    7-2 Presumed reason for the firing : 
    In the configuration of the power system with the new transformer, the primary 420 V AC was supplied also to the secondary output of the transformer (rated voltage 138 V) while the primary input line was open. Because the transformer was not designed to withstand to 420 V, unexpectedly high electric-current flew to the transformer. As a result, the transformer reached to an extremely high temperature, lost the magnetic property, and then flew a larger current to the coil. Eventually, the wire caught fire even though the main breaker of the power supply worked properly and cut off the current.

  8. Causes of the Incident
    8-1 Cause in electric circuit
    The original power supply (PS) could deliver 4,000 A (40V) to the system. In order to enable to reduce the current down to 200A (4 V), a new transformer was installed. As shown in Fig.5, the primary line was designed to switch between the high current and low current modes using a by-pass line switch therein. However, the secondary line was directly connected to the primary line, and primary 420 V AC was imposed to the secondary line.

    8-2 Cause in the design of electric circuit
    The circuit in which the transformer was newly installed was designed and confirmed by the manufacturer. The designer considered that the transformer would withstand to 420 V input voltage. (Actually, the fire occurred by applying 420 V on the secondary line.) Besides, the manufacturer did not carry out a test in a configuration that was similar to the actual environment.

    8-3 Cause in the confirmation procedure of the electric circuit design
    The J-PARC staff member who is in charge of the PS, confirmed the procedure of the on-site test, but did not confirm the safety, because he assumed the circuit was designed by the manufacturer (who designed the main PS) and thereby to be safe.

    8-4 Summary of causes
    There was a failure in the circuit design and it was not recognized in J-PARC center.

  9. Countermeasures
    9-1. Measures for safety management
    At the J-PARC Center, safety checking has always been done in situations such as new installation or repair of instruments/equipment. However, in this case, because a scale of the change was considered to be small, the risk was not recognized and this case was not subjected to safety checking. Learning a lesson from this incident, J-PARC Center has revised its rules, and stipulated in writing the procedures for safety work, including procurement methods. Furthermore, in each Division, a process has been established, as well as a Safety Review Committee to check safety from multiple perspectives. Particular attention must be paid to safety checking in the following cases :

    When planning new installation or modification of instruments/equipment
    When ordering items other than consumer products

    •  Technology shall be reviewed in order to evaluate the capabilities of the company.
    •  In the specifications, it shall be mandatory for the company who receives the order to submit design documents and inspection records, etc. J-PARC shall carry out a thorough safety review based on these materials.

    At the stage of planning work, it shall be confirmed whether or not there is any high-risk work involving the 3H keywords (Hajimete "first time," Henko "change," or Hisashiburi "first time after a long interval"). In cases such as when testing can only be done on-site, multi-faceted, specialized safety checking shall be carried out for the work posing a particularly high risk.

    9-2. Implementation of safety checking (involvement of peer groups)
    In cases that were on order, we invested if there were any cases where a factory test would be difficult and would be carried out a test for the first time on-site. It was confirmed that safety has been ensured. In cases where ordering is done in the future, safety shall be checked according to 9-1 above.

    To assess the appropriateness of the measures for safety management (9-1) and implementation of safety checking (9-2), three outside experts were invited as special members to the safety review committee. An assessment was thereby obtained that these measures are appropriate.

    9-3. Implementation of education based on lessons from this event
    A J-PARC Center Safety Meeting was held for all members of the J-PARC Center community.

    9-4. Checking safety measure implementation
    The effectiveness of these safety measures shall be checked.

    9-5. Checking soundness of the electromagnet power supply
    The pertinent electromagnet power supply will have its fire-extinguishing agent washed off and parts replaced at the factory. The soundness of the electromagnet power supply will be checked by testing parts with the same process as that used during fabrication. In addition, it will be confirmed that newly added parts incorporate protective circuits to ensure safety, and the equipment will be carried into the MLF after conducting a performance test at the factory, witnessed by members of J-PARC Center.

[Top Page]