The standard operating procedures for Hall B are described in a separate document 4 .
There are emergency response procedures which are documented in the Emergency Procedures binder located in the Hall B counting room. Among these are procedures for responding to a fire alarm or pre-alarm in the hall during data taking conditions. There ar e two types of procedures: those for `pre-alarms', and those for full building fire alarms. A pre-alarm is defined to be the output of the locally-designed Hall B Central Alarm Processor (`HuBCAP'), which gathers information from a number of systems and m akes decisions as to the severity of the conditions. A full building fire alarm has a different set of procedures, the first action of which is a call to 911. The procedures are described here; the official procedures are those written in the Emergency Pr ocedures book, which may be revised as appropriate.
Due to the low thresholds used to trigger a pre-alarm, there is a high probability that the alarm was caused by electronic noise or other brief signals not related to a true fire condition. Therefore, the essence of the action on pre-alarm is to investiga te the problem; during data taking, the shift leader is in charge of the situation. The problem may be investigated by a rapid access to the hall, if the alarm severity warrants it. A visual inspection of the area by a CCTV camera precedes entering the ha ll; the counting house, MCC, and the staff investigating the problem in the hall remain in contact by cordless emergency phones at all times, so that each party can immediately be aware of any new information such as additional alarms reporting. Electrica l power may be removed from any and all areas from the Hall B counting room or locally in the hall, as warranted by the situation.
On a full building fire alarm, the crew chief is in charge of the situation. The first action taken is to call 911. If data is being taken, the beam is then removed from the hall, and the CCTV cameras are consulted. If smoke or fire are seen, no one enter s the hall, but information about the situation is gathered and communicated to the crew chief. If no smoke or fire is seen, and if one suitably trained person is available, that person and one other may enter the hall on a voluntary basis to investigate the problem in parallel with the fire department driving to the site. These two investigators will be joined by two investigators from the Machine Control Center (MCC). All participants remain in contact via the emergency response phones. If, during the i nvestigation, an actual fire is apprehended, the standard Jefferson Lab policies apply (as found in the EH&S manual). These permit employees to attempt to extinguish the fire, as long as they 1) elect to do so voluntarily, 2) they have been trained in the use of the appropriate fire extinguisher, 3) the fire is judged to be limited in size and spread, 4) an escape route is guaranteed, 5) an extinguisher is close at hand, 6) their safety is guaranteed.
If the building fire alarm occurs during maintenance periods, the standard Jefferson Lab policies apply. This includes evacuation by all occupants, calling 911, notification of the crew chief and the guard at Post 2. In this situation the area safety ward en has the same role as the trained person during data taking; on a voluntary basis, coordinated with the crew chief, he or she may investigate the alarm with one assistant, and they will be joined by MCC staff and will communicate via emergency phones.
There is a training program which acquaints the trainee with all safety hazards in Hall B, with an emphasis on fire safety. The course is called the `Hall B Safety Walkthrough' because it consists of actually walking through the hall to see the hazards an d safety devices. This course is part of the Jefferson Lab staff training program, course number SAF 132. This course is required for anyone entering the hall without a constant escort. An additional requirement for fire alarm investigators is to complete a written test including a detailed walk-through tour of Hall B, and to undergo additional classroom training. Final approval to investigate fire alarms is at the option of the Hall B program manager.
Because of the number and variety of hazards, and because of the complicated layout of the area, a buddy system has been implemented for Hall B. Under this system, no one may enter the hall without having a clearly identified `buddy'. The two people invol ved are required to communicate at least every five minutes.
There is a maintenance plan for the systems developed by Hall B staff related to fire safety 5 . The staff to carry out this maintenance plan are designated by the Hall B program manager.
The Hall B facility has three modes of operation: data acquisition periods, maintenance periods, and public tours. The data acquisition periods are scheduled in conjuction with Halls A and C, and the accelerator; approximately 35 weeks of the year (67%) w ill be operation in this mode. Maintenance periods occupy the rest of the year, with public tours being occasional events which are organized in advance.
For most of the data acquisition periods, no personnel are present in Hall B because of the radiation in that area. This is enforced by Radiation Control Group procedures for sweeping the areas before introducing beam, and by the Personnel Safety System a fter the sweep is completed.
At times there are accesses to the hall under the conditions of `controlled access,' meaning that the number and identity of the people in the hall is controlled by key access. The exact information as to who is in the hall at a given time is recorded in the Machine Control Center (MCC), and is also generally known by staff in the counting house.
During maintenance periods the hall is in `restricted' access (there is no formal record kept of who is present). A number of restrictions apply to those who enter the hall; the current requirements are posted on the entrance doors. The type of restrictio ns include requirements for personal safety equipment to be worn, for mandatory training, and restrictions on types of activities. These requirements may change depending on conditions; typically they exclude smoking, eating or drinking; require hard hats , safety glasses, and steel-toed shoes; require Hall B safety orientation; and require Oxygen Deficiency Hazards (ODH) training. During maintenance periods, all major power supplies (for mini-torus and main torus, tagging magnet, and pair spectrometer) ar e switched off to minimize risks and power consumption.
Public tours are coordinated through the director's office and through Hall B management. There exists an Operational Safety Procedure (OSP) for tours which permits groups of up to 15 people to tour the hall when accompanied by one escort who is trained i n Oxygen Deficiency Hazards. The members of the tour group are required by the OSP to wear hard hats and safety glasses.