Rink

November/December 2018

Issue link: http://read.uberflip.com/i/1051049

Contents of this Issue

Navigation

Page 31 of 55

32 / NOVEMBER.DECEMBER.2018 USICERINKS.COM by pipe couplings that were not supported to resist pipe movement from pressure or mechanically applied forces. The rising pressure within the brine system eventually exceeded the strength of the joint and one of the couplings separated. This separation suddenly depressurized the brine system and caused the ammonia in the system to rapidly expand, propelling the pipe contents into the room. As ammonia was released, it quickly expanded within the room, reaching estimated concentrations over 20,000 PPM. II. OPERATIONAL DECISIONS THAT CONTRIBUTED TO THE INCIDENT In October 2010, seven years prior to the incident, the City of Fernie received a recommendation from its maintenance contractor to replace the curling system brine chiller due to its age. Analysis of evidence gathered during the investigation identified a series of key decisions during this seven year period that contributed to the incident. Potential inf luences of these decisions were identified, including: A. facility management and organizational priorities; B. failure to include safety risk criteria from aging infrastructure risk assessment; C. operational management structure; D. employee capacity and turnover; E. incomplete maintenance planning; and F. an industry practice of run-to-failure or run-past-failure for brine chillers. The City of Fernie initially scheduled funding to replace the curling brine chiller for 2013. This funding was deferred to 2014 and then deleted from further financial planning. At the end of the 2016/2017 operating season, an ammonia leak was detected in the curling system, indicating a potential failure of the chiller. A decision was made to monitor the leaking chiller in the summer of 2017, followed by a decision to put the leaking chiller back into operation on October 16, 2017. Available evidence did not indicate that there was an awareness of any safety risk associated with the continued operation of the chiller by any parties involved. The decision to operate the leaking chiller is pivotal in the development of the incident. Once the leaking chiller was put back into operation, additional actions and decisions were a response to cascading failures and were beyond the scope of training and situational awareness of those involved. III. IMPACT OF INADEQUATE VENTILATION AND DISCHARGE SYSTEMS In addition to analysis of the cause, the investigation also evaluated the role of the ammonia detection, ventilation and discharge systems following the ammonia release. This analysis determined the ventilation system could not have prevented the large concentration of ammonia within the room, and that it may have contributed to the spread of ammonia to other areas of the arena. It was also determined that the deployment and configuration of the emergency discharge system introduced risk while not reducing the amount of ammonia released into the mechanical room. The investigation found that the configuration and capacity of the ventilation system could not have prevented extremely high levels of ammonia from accumulating within the mechanical room due to the rate of ammonia release. Additionally, the fan belt for the leak/rupture fan was worn and the location of this fan was such that ventilation performance was inefficient; extending the period of time required to reduce the ammonia concentration within the room following the leak. , Figure 1: Outside of failed tube showing hole and corrosion due to brine spray. , Figure 2: Build-up of salt crystals from brine around lower compressor cylinder heads of compressors #1 and #2. , Figure 3: Curling chiller opened showing contamination. The leaking upper tube is identified by the white arrow—second row, third tube from the right side of the photo. 1 2 3

Articles in this issue

Links on this page

Archives of this issue

view archives of Rink - November/December 2018